Table of contents

In the process of technological evolution of products, some functions may evolve slowly, which hinders the evolution of products as a whole. In order to compensate for the imbalance of technological evolution and find the corresponding technical shortcomings, a product design process combining technology-effect matrix with TRIZ is proposed. Through searching and screening patents related, the target patent pool is obtained and each patent is labeled. We, at the same time, summarize the technical means and effects, construct the technology-effect matrix and deduce the design objectives by extracting the technical weaknesses. Finally, the design model is obtained through using TRIZ tool to analyze the design objectives. Taking the innovative design of rotary cultivator as an example, the technology-effect matrix of rotary cultivator was constructed and a technical weakness was extracted. Based on this, the innovative design of rotary cultivator was carried out with the objective of reducing the wear of rotary cultivator blade, and thus the feasibility of the design process was verified.

Earthwork calculation occupies an important position in all aspects of engineering construction, and its calculation accuracy and efficiency will have a significant impact on engineering construction. This article analyzes and summarizes the calculation results of various methods and software used in calculating earthwork; points out the shortcomings of the existing earthwork calculation methods; discusses the future development direction of earthwork calculation, which requires not only high accuracy, but also Faster and smarter information feedback provides timely and effective reference and guidance for project construction.

In the era of the construction engineering industry informatization, the collection, storage, and reuse of big construction data is the only way to apply and inform big data in the construction industry. Relying on the reverse logistics management system, modern construction technology can simulate the timeliness of materials, funds, and time limits of the actual construction site according to the logistics reverse engineering principle. Then, reverse the active management status of the internal logistics model of the actual project. It is one of the critical problems for construction engineering informatization to study applying big data and cloud computing technology to solve problems from massive and complicated data or mass data. Since the research on big data and cloud computing technology started late in construction engineering, it is worth studying which aspects of these technologies can be used in the whole life cycle of this field. This paper proposes a structural construction optimization identification method based on IPEM rapid reanalysis. USES the co-solubility of reverse logistics structure construction technology and applied the data mining method to solve many problems such as refined construction management leftover material surplus in the later construction stage. Finally, a construction optimization model of reverse logistics building system based on improved data mining is proposed according to the actual engineering requirements.

The Awakening of sleeping members is the focus of smart marketing. The cost of losing a sleeping member in existing pharmacies is about 1/4 of the cost of adding a new member. This paper abstractly classifies the problem of whether the sleeping members are easy to be woken up in drugstore into a dichotic problem, and solves the problem that the existing sleeping members' wake-up model is applied to the users of the sleeping members in drugstore and the prediction accuracy of the users' consumption in the store is not high. According to sleep members have rich behavior characteristics, behavior, medical conditions, gender, age, payment, the membership card integral, active level characteristics such as properties, through the analysis of these sleep member attributes in order to distinguish between normal users and sleep, characteristics and the actual scene at the same time, using the method of feature selection to remove redundant features to form a new expression model. On the basis of multiple perspectives, multiple algorithm models were integrated to wake up the sleeping members. The actual store-to-store consumption data of the sleeping member users were used to verify and compare the effectiveness of the algorithm model, and the characteristic expression model with the best discrimination and accuracy was selected. Through the experimental study of the abstract dichotomy algorithm, it is found that the sleeping members with what characteristics are easy to be woken up. The experiment proves that the feature expression model has a high degree of differentiation for the sleeping users, and has a high accuracy in judging the users who are likely to be woken up among the sleeping members of the pharmacy.

When the traditional collaborative filtering algori- thm is applied to drug recommendation, the recommendation effect is not good due to the sparsity of data. In view of the above problems, this paper proposes a collaborative filtering recommendation algorithm based on user behavior and drug semantics (UBDS-CF). Firstly, we construct the purchasing behavior matrix of users and drugs, and use the weighted cosine similarity to calculate the basic similarity between drugs; then construct the category label matrix of drugs, calculate the category similarity, and extract the feature vector of drug function text using the word vector model; category and main function together constitute the drug semantic information; finally, the above two similarities are integrated in a linear weighted manner to obtain the final similarity of drugs. Experimental results show that the average absolute error and root mean square error of the algorithm are significantly reduced, which proves the effectiveness of the proposed algorithm.

Current approaches of using neural networks in lane departure warning systems are expensive. And it is difficult for neural networks to process 2K and 4K images. In this paper, we use a series of image preprocessing techniques such as perspective transformation, threshold processing and mask operation to process high-resolution images and the optimized sliding window method to fit the lane lines. Compared with neural network method, we can not only reduce hardware cost, but also quickly process high-resolution images. In addition, compared with the traditional lane line detection algorithm, we extract the region of interest through perspective transformation, which not only greatly reduces computation, but also converts images into an aerial view for subsequent processing. Especially, we carry out threshold operation and mask operation after perspective transformation, which greatly improves the performance of our algorithm in a strongly interfering environment. As can be seen from the experimental results, our method has good detection effect and can be applied to various road sections in different environments.

Objective: As a carrier, bridge cranes are widely used in port cargo handling. Due to the structure of the crane itself, the load inevitably produces displacement deviation and lifting weight swing. Therefore, the optimization investigation of the bridge crane control system is carried out to guarantee the fast and secure operation of the crane. Method: According to the characteristics of bridge cranes, combined with the experience of on-site operating staff, the fuzzy rules that meet the working conditions of the site are designed by using the advantages of fuzzy PID (Proportion-Integration-Differentiation) control algorithm without relying on an accurate mathematical model, to set the parameters of the fuzzy PID controller. The speed control technique of frequency converter is adopted to design the system to improve the safety of the overall system. Results: The results show that the adjustment time of the swing angle of crane weight of the traditional PID control requires 6.79s, and the maximum swing angle of crane weight is 0.039rad, and the adjustment time under the fuzzy control is 4.89s, and the swing angle of crane weight is 0.013rad. The fuzzy PID controller surpasses the traditional PID controller in the positioning accuracy. It is not sensitive to changes in the internal parameters of the system. Also, it has strong adaptability to changes in operating conditions. The swing angle of crane weight is smaller, which further improves the robustness. The vector frequency conversion speed control has a more precise control effect on the speed and running track of the operating mechanism of the bridge crane. This method can be applied to the automatic operation mechanism of bridge cranes and achieve good control performance. Conclusion: The conducted optimization investigation of the bridge crane control system makes the control system flexible, simple and robust, providing important theoretical support for related explorations.

With the advancement of modern society, information exchange is convenient and fast, and the increase in data volume makes data more convincing. Therefore, in the context of big data, the evaluation of college physical education quality is more scientific and objective. In the context of the development of the era of big data, college physical education teaching models mainly involve three micro-classes, MOOCs, and flipped courses. Some colleges and universities use the Internet to innovate physical education teaching methods or introduce innovative teaching software to effectively improve college physical education. The individualized level of class teaching makes it more in line with the physical learning needs of students in the era of big data. Based on this, starting from the big data era and the development of college physical education, combined with specific teaching cases, several college physical education teaching models are introduced, in order to provide help for college physical education teachers to carry out class teaching.

In this paper, a kitchen waste illegal component recognizer based on embedded device is proposed, which can intelligently identify and record illegal components in kitchen waste. The recognizer is based on SSD-MibileNetV2 algorithm. The Quantization-aware training model is established in TensorFlow framework. The whole model is quantized by uint8 and transformed into TensorFlow Lite structure, and then deployed to the embedded device based on raspberry Pi 4b, which can finally achieve 4.5FPS and 76.35mAP. The recognizer can meet the requirements of real-time and accuracy in detecting illegal components of kitchen waste, and can effectively reduce labor costs and meet the requirements of intelligence.

For the current medical block chain, the access rights of electronic medical records and agent encryption are mainly controlled by permission contract. The way of protecting electronic medical records realizes the sharing of medical data, but neglects the management and protection of user identity. In this paper, this article studies the privacy protection technology of the blockchain, combined with the stealth address and zero-knowledge proof technology, and proposes a plan to weigh anonymity and security to protect the privacy of users in the medical blockchain.

In order to overcome the limitations of traditional optical display technology, a novel optical display technology based on optical image virtual reality technology is proposed. The optical display technology combines with virtual reality technology, relying on the Internet platform, fully and comprehensively simulates the visual display effect of human eyes. The results show that the optical display technology can break through the limitations of traditional optical display technology, improve the display effect, and build a high immersion virtual reality world for users.

To verify the brake effect and the mechanism of exerting the brake of the electronic mechanical brake (EMB) system in the brake process of electronic vehicles, the working environment and function of each component of EMB are analysed based on the current relevant laws and regulations. Each component of EMB is calculated and selected. The mechanical relationship between the components of the EMB system are analysed, the preliminary EMB structure is designed, and the key parameters such as EMB brake pressure and EMB brake clearance elimination time are tested. Results show that the difference between the experimental value and theoretical value of the brake pressure test of the preliminary designed EMB structure aren't large, and the brake clearance force increases with the increase of the locked-rotor voltage. When the voltage increases to 12V, the theoretical brake clearance force is 16510N and the experimental value is 15998N. During the EMB brake clearance elimination time test, the maximum speed of the brake block running at a constant speed during the elimination of brake clearance phase is 2.39mm/s, which is similar to the theoretical value Vmax=2mm/s. The experimental and theoretical motor brake speed is reduced to 88 rpm/min at 0.15s and at 0.13s. With all indicators considered, the EMB system can be used to control electromechanical brakes and provide certain guidance for its development.

A new Harris-AE feature matching algorithm based on image edge information combined with SLIC superpixel segmentation method is proposed to achieve high-precision registration of visible and infrared images. First, perform edge enhancement and histogram equalization processing on visible light and infrared images, superpixel segmentation technology is used to eliminate sub-images with low information entropy, and then by constructing a multi-scale Gaussian pyramid to detect the Harris corners of the image edge information, using autoencoder neural network to generate feature descriptors corresponding to feature points, and match the feature points of two images through the fast nearest-neighbor algorithm with bidirectional matching strategy. Finally, the RANSAC algorithm is used to purify the matching points, and the spatial geometric change parameters between the two images are estimated by the least-squares solution to complete the registration of visible light and infrared images. Experiments have proved that this algorithm improves the accuracy of matching while shortening the registration time.

This article combines the new characteristics and changes brought by the widespread use of computer technology to the environmental design industry, and analyzes the basic characteristics of computer-aided design in view of some of the concerns of environmental design in the network age, and points out that the new era willfurther actively develop computer an effective way to assist design.

Chaotic secure communication has always been a hot topic in nonlinear circuit research. In this paper, an improved jerk chaotic circuit is designed by using cell circuits and nonlinear function circuits. Simulation results show that the circuit can generate multi-scroll chaotic attractors. The position and number of chaotic attractors can be controlled by adding or deleting the number of nonlinear function circuit or by adjusting its parameters. Theoretical analysis and numerical calculation show that the circuit can describe jerk system, and the calculation results are consistent with the circuit simulation results. The circuit is not only simple in structure, easy to operate and high in experimental accuracy, but also easy to control chaos, which is widely used in secure communication.

Generic cabling is an important basic work for enterprises to realize networking, automation, digitization, and intelligence. This article introduces the structured integrated wiring system of the network, especially the 6 modules included in the structured wiring system, and the network wiring engineering circuit in a more detailed discussion, and combined with engineering application examples, analyses the structured wiring system Composition and engineering design process.

The continuous development of big data technology has brought many new ideas and challenges to power system analysis and control. With the development of distributed power sources, energy storage, monitoring, and protection devices, the traditional distribution network has gradually evolved into much Active distribution networks with control ability. At present, the power quality of the power system is a hot issue in power system research. There are many types of power quality disturbances, mainly including single type and compound type disturbances. The thesis first uses Matlab to model and simulate the power quality disturbance signal and then obtains a complex number matrix through S transform, then calculates its modulus, and extracts the corresponding characteristic parameters to form the characteristic vector. Secondly, the thesis uses a simple and efficient decision tree to correctly classify and identify power quality disturbance signals. Finally, the thesis's simulation analysis results show that this research method combining S transform and decision tree has high recognition accuracy and strong anti-noise ability, which is a very suitable method.

With the development of computer software and hardware technology in colleges and universities, the application of multimedia computers has become increasingly widespread. Graphics and image processing occupy a major position in the application of multimedia computers in colleges and universities. Image display is the basis of image processing. The key of special effect display is to divide the image into different image area blocks. The usual image special effect display method is to divide the image into different image blocks according to certain rules. Based on the analysis of the principle of image special effects display, this paper proposes a template-based image special effects display method. Experimental results show that compared with traditional interpolation algorithms, such as bilinear interpolation and bicubic polynomial interpolation, this algorithm has better interpolation effects and overcomes the speckle interference in single wavelet interpolation.

Based on the high-speed development of computer technology, VLSI and communication technology, this paper studies an anti stealing wireless meter reading system which can improve the power supply enterprise's anti stealing ability and collect, count and analyze the power consumption of large users timely, effectively and accurately, and analyze the characteristics and components of the system design to meet the market demand. The system studied in this paper is mainly used for power supply enterprises to control power stealing for large users.

In order to guide students to master Arduino development board, sensors, unity3d and other software and hardware application methods, familiar with the development process of virtual reality and intelligent hardware control platform. Learning from the WeChat "jump and jump" finger game, based on virtual reality technology and Arduino intelligent hardware control platform, the virtual intelligent fitness device with good immersion, strong interactivity, and combining entertainment, exercise and virtual experience is developed and designed. The "virtual reality + Arduino" immersion interactive experimental platform provides immersive and interactive immersive helmet and desktop screen display modes, uses a variety of intelligent control methods such as in-situ jumping action or grip to replace the finger pressing action in WeChat to control the jump of the game model, and can choose the sky, ocean, land and other fitness Scene, to achieve physical exercise and intellectual purposes. In the process of case design and production, virtual reality immersion equipment, grip and running pad are combined with fitness to cultivate students' comprehensive practical innovation ability.

There is a problem that the amount of users' preference data on the mobile is small, and users are unwilling to disclose the preference data for the recommendation system about mobile users, so the server can't centrally train a large amount of users' preference data for a personalized recommendation. This paper proposes a personalized federated matrix factorization algorithm by introducing a federated matrix factorization model. The algorithm introduces users' and items' biases to modify the predictive rating model on each mobile; At the same time, conformity is introduced to give different weights to the preference data. In the case that the preference data does not leave the mobile, but the user preference data of the multiparty mobiles is shared, the multi-party mobiles and the server jointly train personalized matrix factorization model. The experimental results show that the algorithm in this paper still has high recommendation accuracy under the premise of correct updating in the federated matrix factorization model that uses bias and conformity.

Practical Byzantine Fault-tolerant Algorithm (PBFT) is the most widely used consensus algorithm in alliance chain, which has the advantages of fault tolerance and large throughput. However, PBFT also has some problems that can't be ignored in specific blockchain applications, such as bad behavior of master node, high network communication overhead and low system flexibility. In this paper, the improvement measures of PBFT consensus algorithm are summarized from many aspects, and then compared with traditional PBFT algorithm. Finally, the development direction and trend of PBFT algorithm are prospected, hoping to provide reference for the research and innovation of PBFT consensus algorithm in the future.

This paper studies the completion of power communication technology and transmission network planning. It summarizes the basic research and current situation analysis of transmission technology and business network planning. First, it focuses on the functions of power communication business, information flow, transmission channels and bearing requirements. The characteristics of various power grid technologies and grid communication business models in China are analyzed. Secondly, it carried out analysis and research on the basic design principles of power communication technology and transmission network business planning, evaluation of main indicators, and focused on elaborating the digital communication network management system of synchronous transmission of power communication, the characteristics of ason power communication technology, network protection and management and basic principles of power communication technology and transmission network such as restoration.

This article is based on a comprehensive analysis of application bearing work methods and specific business models according to various types of power grid transmission services. For different application methods of power grid transmission services on each service transmission network and data transmission to the service network channel bearing work the way of analysis is different. Based on the security level of the power grid transmission service, the method of carrying the service transmission channel network, the method of carrying the service transmission network, the direction and flow of the service, this paper analyzes the way of carrying the channel on the business network.

Based on the current new situation, not only to more scientific assessment of the supplier's equipment quality and technical level, but also on this basis to take into account the equipment in the supply chain life cycle of the production, procurement, operation and decommissioning of the cost, through the evaluation of equipment quality and cost, to achieve the asset-wide life cycle management of the supplier intelligent evaluation, the establishment of supplier intelligent evaluation management information system in order to achieve the establishment and perfection of supplier classification performance evaluation management system.

Among the target detection algorithms, YOLOv3 algorithm has fast detection speed and high accuracy, but it is difficult to directly deploy to small embedded devices because of its high requirements for the computing power of the device. In response to this problem, this paper combines the characteristics of EfficientNet-lite network and YOLOv3, and proposes an improved model of YOLOv3 combined with EfficientNet-lite network. This model takes advantage of the small size and high efficiency of the EfficientNet-lite network to reduce the size of the model, so that it can be applied to wearable devices to help blind people traveling to detect environmental information. Experimental results show that the model greatly reduces its size and its dependence on equipment performance under the premise of a small decrease in detection accuracy.

To realize image information hiding, we train a multi-layers autoencoder includes an encoder and a decoder. Using encoder architecture, we input the secret color image and generate the cover image where the size stays the same directly. In addition, we reconstruct the secret color image by inputting the stego images passing on the decoder which accomplish the secret image blind extraction. Our method realizes a high resolution and high payload color image steganography. Experimental results show that we can realize the high capacity blind specified secret images information hiding automatically by our method.

In view of the various instruments, equipment, survey methods and current problems involved in the deep sea survey, based on the analysis of the characteristics of the deep sea survey data, this paper puts forward a design method for the application framework of multi-source heterogeneous data in the deep sea survey. This method uses XML language to analyze, extract and fuse the multi-source heterogeneous data generated by marine survey equipment, so as to meet the practical application requirements of data. The practice shows that this method is feasible and easy to use as an application method of multi-source heterogeneous data in deep and deep sea survey.

Wisdom supply chain is a new supply chain model developed in the new generation of information technology revolution, which makes the organization and management mode of supply chain undergo an inevitable change. As an important intersection connect "One Belt and One Road" and the Yangtze river economic belt, integrating wisdom supply chain industry into the regional resource construction system and cultivating wisdom supply chain management talents are the urgent requirements for the Yangtze River Delta urban agglomeration to adapt to the adjustment and upgrading of industrial structure and pursue higher quality development. This survey analyzes the new characteristics of wisdom supply chain and the new demand for human resources by using the methods of literature analysis and enterprise research. Then, explores the demand characteristics of market-oriented wisdom supply chain management talents by mining the recruitment information of supply chain and logistics posts in the Yangtze River Delta and its surrounding areas based on clustering, statistics, text mining and other data analysis methods.

Recording the electricity consumption of users is a monthly routine of the electric power department, which often requires a lot of material and manpower. Therefore, using new technical methods, with the help of automatic remote and local meter reading technology, to achieve the purpose of automatic meter reading to monitor the electric energy meter, work efficiency is improved, and the real automatic power management is realized, and the remote electric meter reading system becomes Alternative meter reading problem solutions for grid companies. This system can meet the remote data recording, collection and real-time monitoring functions of the user's electric energy meter in the remote power meter reading system within a certain range.

With the continuous development of the times, chip technology has also been innovated. In order to meet the needs of edge computing networks and data communication networks, it is necessary to make chips with shorter delay time and lower power consumption, which in turn brings challenges to chip architecture design. In the field of chip design, the problem of chip performance has been overcome. In many edge design networks, the core of the design is to allow the chip to meet low power consumption requirements, thereby extending the use time of edge devices. This article conducts research on a low-power master chip architecture to enable it to meet the needs of edge computing networks and enhance product competitiveness.

In this paper, aiming at the design of the sorting device for rod-shaped vegetables and fruits, the research status of grading device at home and abroad is analysed, and a variety of grading device mechanisms are compared, and the design scheme of a variety of carrot grading device is determined. Through analogy analysis, a carrot grading device is designed. The grading device is composed of cleaning and feeding device and automatic grading device. The cleaning uniformity device is mainly to put carrots in the pool for cleaning, and then the motor operates, and the conveyor belt with inclined partition starts to work, and the carrots are transported to the grading device through the partition board. The principle of automatic grading device is to grade carrots by adjusting the gap between two adjacent conveyor belts. The gap size is arranged from small to large in order to realize automatic grading according to the diameter of carrots. In this paper, the following aspects are analyzed: by comparing the mechanism of other grading device, the carrot grading device is analyzed and designed. The mechanical part is designed to be three-dimensional using Pro/E modeling, as well as part modeling and assembly. At present, there are some defects in the design of rod-shaped vegetable and fruit grading device in our country. The research has filled the defect and has high practicability.

Convolutional Neural Network (CNN) has shown its strength in image processing task, and Hidden Markov model (HMM) is a powerful tool for modeling sequential data. This paper presents a new architecture for audio-based chord recognition using a CNN-HMM mixture model. This architecture replaces the Gaussian mixture model (GMM) and Deep Neural Network (DNN) layers of GMM-HMM and DNN-HMM models with CNN. The model performance is evaluated through a dataset using different combinations of chroma vectors (STFT, CQT, CENS) as features, based on that, a scale recognition sub-model is tested.

This paper combines the characteristics of the real Washington DC house price estate market and the theoretical model of housing prices to conduct an empirical analysis and comparison of the current mainstream housing price index prediction models. It is found that the current mainstream model only studies the trend of the housing price index itself, and is not sensitive to the characteristics of the house itself. Therefore, this paper uses a multiple regression model to integrate the advantages of external factors and use the improved housing price composition establishing a multiple regression prediction model with the particle swarm optimization. It not only makes up for the disadvantages of poorly determined housing price regression indicators and lack of statistical data in multiple regression prediction, but also enables the model to reflect the inflection point of housing prices in advance. PSO is used for selection of affect variables and regression analysis is used to determine the optimal coefficient in prediction.

Cross-modal retravel has been used in many fields, such as business and search engines. Most search engines for business are text-based, but text-based search engines are limited by equipment and the strict requirement for knowledge. Text-based search needs keyboards to finish the search process, which requires users to have the knowledge of using keyboards. Compared to the text-based search, audio-based search has advantages. First, it avoids the traditional ways of inputting information. And it gets rid of the gap in time between inputting information for searching and getting useful information. In this paper, we propose a way to use audio to search images for business applications. We use deep learning to implement cross-modal retrieval systems between images and audio. We first extract features from images and audio respectively. And then we implement a neural network with two identical networks to learn the correspondence between images and audio. The first network extracts the features from images and audio further for calculation, and the second network learns whether two features from different modalities are related. This research provides a new way for business applications to search for information more instantly.

Machine learning is known as the most popular methodology to do prediction on large data set while NBA's data sets consists of plentiful statistics. Since predictions of various events are important, our research would investigate whether machine learning algorithms are efficient in doing prediction on certain NBA data sets and tasks. We are focus on mainly three supervised tasks, namely: All-Star Prediction, Playoff Prediction and Hot Streak Fallacy. For Playoff Prediction, we predict the team performance by doing machine learning on two data sets consisting of distinct well-selected features and compare the result to show which data set are more suitable for the machine learning to work. The results show that advanced statistics outperform the elementary ones. For Hot Streak Fallacy, we build the model based on multiple-linear regression to address the question: is hot streak a fallacy? It turns out that there is a lack of evidence to support 'Hot Streak Phenomenon'. For the NBA Trend, we try to view how the games involve for the past decade, and analyze the correlation of playoff tickets and other data.

A sequential recommendation system is war-ranted, and several models have been created to meet this need. However, sparsity is a major problem in this field. In order to deal with the space-time sparsity problem in a sequential recommend-dation system and avoid the unreliability caused by manually generated data, we have established the UBS structure. We treat user behavior sequences in batches. Each batch gets the inte-grated data set UBS after K iterations through a two-step cycle of clustering and merging. The results are connected as inputs of a sequential neural model. We use three baseline deep learning algorithms –Caser/SASRec/HGN to evaluate the model. We compared the UBS model with the original model in terms of normalized discounted cumulative gain ms(NDCG) and Recall. It is found that when the UBS model is applied to a short sequence of user behaviors and the iteration number K=2, it can improve the recommendation efficiency compared with the original algorithm (represented by K=0).

Machine learning has been extensively used in diverse healthcare settings since the 21st century. Statistical models are proven to be powerful in detecting early disease symptoms and could potentially aid decision-making in the healthcare system. To help improve medical resource allocation during COVID-19 pandemic, we aim to develop machine learning models that predict each patient's length of stay (LOS) in hospital. Three machine learning models, namely, K-nearest Neighbors Algorithm, Logistic Regression and Random Forest are implemented and optimized on the same healthcare dataset. The final accuracy of each model is 0.3442, 0.3524 and 0.3541 respectively, which are not very high. Our subsequent correlation analysis on the healthcare dataset shows the patients' features used do not provide sufficient information for accurate LOS prediction. Yet, machine learning approaches could potentially yield much better results if the data quality can be improved by including additional relevant patient features and breaking LOS into more appropriate intervals. More detailed healthcare data should be obtained to make the LOS prediction useful for healthcare management.

Genetic algorithm and neural network are both ideas and algorithm skills. The flexibility of this method provides a stage for many researchers and applications. In view of the complexity of general construction engineering valuation problems and the advantages of genetic algorithm and neural network algorithm, this article proposes a new construction engineering valuation model. Taking the actual residential project as an example, the feasibility and reliability of the genetic algorithm to optimize the neural network model are verified, and it is better than the simple neural network model, which provides an effective method for actual project investment estimation.

Aiming at the problems of low prediction accuracy and premature convergence of traditional BP network prediction models, firefly algorithm FA is introduced into BP neural network model to optimize the optimization process of neural network weights and thresholds. This paper presents a BP neural network prediction model based on firefly algorithm. Taking Fujian port throughput data as an example, the application verification and comparative analysis of the improved prediction algorithm are carried out. The experimental results show that the improved FA-BP prediction algorithm has better performance in prediction accuracy and stability.

At present, the primary method of sports biomechanics research in track and field events is to use a high-speed image analysis system to monitor the training process and obtain kinematics information through special or general analytical software. To improve the training efficiency and correct the defects of technical movements in time, coaches urgently need a kind of equipment that can feedback the movement information in time during the training. Because of this situation, a track and field training information acquisition and feedback system based on a digital track are proposed based on sufficient condensing project requirements. The system USES flexible array sensors obtain contact interaction information during the run-up, and USES special analytical software to process kinematic parameters in an integrated manner, providing real-time biomechanical parameters such as step length and speed step frequency, and takeoff force in the process of moving. The system realizes the comprehensive analysis of sports representation and internal mechanical factors, which helps coaches and athletes to deeply grasp the internal law of the project. Its effectiveness and scientificity have been preliminarily verified through the athletes' testing in the national track and field team's long jump event.

Recently, Wireless sensor network is regarded as a new information acquisition platform, its relevant technologies have been widely used as the basis of the Internet of Things. In order to realize the campus information construction, a campus stereo monitoring system based on wireless sensor network is designed in this paper. Such technologies as the intelligent information processing technology of wireless sensor network, the remote WIFI wireless data communication technology, the cloud services technology and the management query technology for monitoring points based on the internet display platform are adopted in this system synthetically, realize on line real time monitoring and location, which can not only reduce the human cost, but also has important significance in maintaining campus safety and campus management.

With the substantial increase in the amount of data information and the development of 5G technology, boundary computing technology based on 5G, as a real-time and efficient data processing method, has broad application prospects. The aggregation gateway further aggregates the power injection data of all electric energy storage units and sends it to the power company. The power company uses its own private key to access the aggregated data. Row decryption to obtain the total power data recovered in each time slot. The paper proposes a new solution using Lagrangian interpolation to ensure data privacy and security. The security shows that the algorithm can not only protect the privacy of the electric energy storage unit, but also realize the confidentiality and integrity of the injected data. Efficiency analysis shows that this algorithm can significantly reduce computational complexity and improve communication efficiency.

The smart grid covers the traditional power grid and the Internet of Things network, including smart meters, which helps manage and monitor various customer consumption patterns in near real-time. To achieve data reading and effective management of terminal, smart meters connected to the power metering system, the article proposes using low-power wireless network (LPWAN) technology to connect terminal, smart meters to data gateways and private power clouds to achieve smart power metering Technical solutions. The solution uses the LoRa wireless module to realize the two-way interaction of power metering data and has the advantages of reliable operation, low cost, low power consumption, and remote access. This article also focuses on using emerging LPWAN technology to help build smart meter wireless access solutions and demonstrates the system implementation for smart meter end-to-end LoRa connection.

Segmenting 3D mesh is a difficult topic in computer graphics. This paper proposes a new mesh segmentation algorithm. First, we over-segment the input mesh to generate a series of superfacets. Second, we introduce a novel region fusion algorithm to obtain the segmentation result. Various experiments show that our algorithm is able to handle 3D meshes with complex shape and rich details. Even though the input mesh has noises or holes our algorithm can still achieve satisfactory results.

To improve network programming efficiency, the paper proposes a network automatic programming method based on big data analysis technology. Build a big data analysis model for automatic network programming. The dissertation uses the distributed information fusion method to extract the significant data feature of automatic network programming, extract the associated feature quantity of automatic network programming big data, and combine the fuzzy membership analysis method to carry out the synthesis operation and fuzzy comprehensive evaluation decision of automatic network programming and build network automation Programming multi-level index evaluation system to realize automatic network programming optimization.

In this paper, the framework of one-stop application service system of Hunan Regional Seismological Network is introduced, taking the fast earthquake reporting of Hunan Regional Seismological Network as an example, in order to realize the automatic, one-stop application service of graphic and text data display, which integrates "obtaining earthquake parameters-issuing earthquake message-completing website earthquake situation update-automatically sending fax".

With the development of tourism, vehicle traffic scheduling in scenic spots has gradually become a research hotspot. Aiming at the problem of vehicle scheduling and route planning at multiple service points in the scenic spot, we propose a method to minimize the total cost of vehicle scheduling and route planning tasks at the service point under the constraints of the weight of the required goods and the vehicle volume. First, we classify the demand points and assign the corresponding service points; then we aim to minimize the cost of vehicle scheduling and establish a nonlinear programming model under the constraints of the weight of the demand product and the volume of the vehicle; then, we will The model is reduced to a single service point TSP problem, and based on the improved ant colony algorithm, a solution model based on TSP is designed. The experimental results show that our model and algorithm can make the distribution cost of multiple service points the lowest and the iteration speed is faster.

In this paper, a method of real-time vehicle speed and vehicle type recognition is proposed. The vehicle type recognition is based on residual network to increase the convergence speed and improve the feature expression ability, and attach the centre loss to improve recognition accuracy of similar vehicles. The speed recognition is based on moving object detection. Experimental results show that the average error of vehicle speed is no more than 5%, and the average precision of vehicle type recognition is 85%, towards minibus and cars, the precision reaches 98.7%, which is superior to the traditional recognition method.

Accurate diagnosis of Alzheimer's Disease (AD) is of great importance to patient care, especially in the early stage, because it enables patients to take preventive measures before irreversible brain damage occurs. Obviously, if computer-based diagnostic models can make better predictions than doctors, the future of health care will change dramatically from the current system. Based on magnetic resonance imaging (MRI), this paper realizes a classification algorithm to distinguish AD patients from normal people. OASIS Brain Imaging Data Set is an open source online data set, which is a coronal bitmap of the human brain. Canny edge detection algorithm is used to extract image information, and K-Means and convolution neural network (CNN) are used to train the data set. The results show that the recognition accuracy of the training set is 99%, and 26% higher than that of the K-Means algorithm, which can be effectively used in the early diagnosis of Alzheimer's disease.

Clouds have a huge impact on the energy balance, climate and weather of the earth. Cloud types have different cloud radiation effects, which is an important indicator of cloud radiation effects. Therefore, determining the type of cloud is of great significance in meteorology. In this paper, the Convolutional neural network with Squeeze & Excitation Networks (SENet) are mainly used to solve this probelm. CNN can automatically learn the filters that need to be manually set before, and can learn complex edge, spatial and texture information in the image which are difficult for traditional methods to learn and extract. Moerover, a website and a deep learning framework are established to showcase the results of this article and to further develop our models and methods through open source methods.

As a characteristic of big data, the individual data in it is no longer isolated, and the data and its underlying mechanisms have complex associations, which make all data into an indivisible whole. The dynamic generation and disappearance of data will change its original relationship and affect the overall characteristics of the data. This feature of big data makes the subject-oriented analysis methods such as data mining present limitations: the presupposition of the subject and the analysis by subject split the interaction relationship between the subjects, leading to the loss of the implicit mechanism in these relationships. Aiming at the problems of traditional network big data multi-resolution acquisition methods such as high acquisition cost, long completion time and low acquisition accuracy, a JA-va3D-based big data network multi-resolution acquisition method is proposed, and average interactive data is introduced. The extraction method estimates the power spectral density of the network data multi-resolution acquisition, and uses the ADASYN algorithm to remove the invalid multi-resolution data, and realizes the large data multi-resolution accurate acquisition. Experimental results show that the proposed method has lower acquisition cost, shorter completion time, and higher acquisition accuracy; it has certain practical value and can be widely used in various fields.

Aiming at the problem that oral hygiene is often overlooked and oral treatment is often convenient, this article proposes a new type of oral medical vehicle based on digital technology. The article starts with the application range of the dental medical vehicle, clarifies the necessity of the design research, and introduces the structure and key technologies of the currently designed medical vehicle, and then introduces the working mode of the medical vehicle and the optimal path algorithm carried. Finally, based on market surveys, it is concluded that dental medical vehicles will definitely attract more people's attention in the future, and their market demand will increase.

In recent years, more and more people are keen to express their feelings and opinions in the form of pictures and texts on social media at the same time, which makes the multimodal data with pictures and texts as the main content growing. Compared with monomodal data, multimodal data contains more information and can reveal the real feelings of users. The analysis of the emotion of these massive multimodal data is helpful to better understand people's attitudes and viewpoints and has a wide range of application scenarios. In order to solve the problem of information redundancy in multimodal emotion classification tasks, based on the tensor fusion scheme, a multimodal emotion analysis method based on attention neural network is proposed. This method constructs a text feature extraction model and an image feature extraction model based on attention neural network, which highlights the key areas of image emotional information and the words containing emotional information, which makes the expression of single-modal features more concise and accurate. The tensor product of each mode is regarded as the joint feature expression of multimodal data, and the redundant information of joint features is eliminated by principal component analysis, and then the emotion category of multimodal data is obtained by support vector machine. The proposed model is evaluated on two real Twitter image data sets. The experimental results show that, compared with other emotion classification models, this method has a great improvement in classification accuracy, recall rate, F1 index and accuracy.

As an important part of intelligent traffic management, license plate recognition has very important research value. In order to solve the problem that the location of traditional license plate detection methods is not accurate and the detection results are easily affected by the environment, this paper proposes a license plate detection algorithm based on Faster R-CNN and Inception ResNet_v2, which realizes accurate license plate detection on VOC license plate data set by means of transfer learning. The effective extraction of a single character is realized by using the pixel statistics method in the detection results, and then the mLeNet5 convolution neural network model is designed to recognize the single character. The experimental results show that the algorithm can achieve efficient and accurate recognition of license plate characters with occlusion and angle tilt.

This paper raises some method to make neural network training can work on distributed computing clusters well.

Assessment is the core and fundament of rehabilitation, which can guide the whole treatment process. In the rehabilitation, doctors or therapists need to assess the function of patients in upper/lower-limb, based on subjective assessment and objective assessment. While it may cause large error and high cost by traditional ways. Therefore, artificial intelligence technology is applied to the field of medical rehabilitation. This review will summarize the application of objective assessment methods which are base on artificial intelligence including not limited, trajectory error feature, joint angels and joint angular velocity, sEMG signal feature. Finally, the review concludes that existing objective methods are generally affected by the scale of data and the number of feature. This review will give instruction for a lager application in rehabilitation field.

The task template system of this system combines the business requirements of the work and provides the classification management function of work packages and work package templates. At the same time, data integration with other systems such as the main data system and resource library is realized. Through this system, users can instantiate a single work package to form a single work item, and can also establish the front and back relationship of the work package through the work package template, thereby realizing the instantiation of the work package process.

The paper firstly analyze the current problems in the information services integration, then the architecture of information services integration based on SOA is designed. The paper also gives a integration method based on ESB and Web Service. Compare the differences between SOA and traditional integration methods. We know that the integration method based on SOA makes full advantage of the information services system. Results show that the method is scientific and easy to use.

Boundary pillars are important symbols of the junction, and at present, the most common way of managing boundary pillars in China is to employ a full-time boundary pillar maintainer in order to maintain the pillars themselves better. Damage to or loss of reference materials such as the "borders" can cause considerable problems in resolving boundary disputes and controversies.

In recent years, the Geographical Information System (GIS) has been in a stage of rapid development, both theoretically and in terms of application. With the advancement of network information technology and the increasing maturity of GIS technology, WebGIS combining the two has gradually been widely used in various industries, while the Internet of Things (IoT) is also developing rapidly, on which a more complete set of intelligent electronic boundary pillars based on IoT technology has the theoretical basis to emerge.

Traffic lights play an important role in the distribution of conflicting traffic flow, and the correct detection and identification of traffic lights play an important role in reducing the occurrence of traffic accidents. Since the 1980s, China has begun the research and development of intelligent self-driving vehicles, which have higher requirements for all kinds of traffic sign recognition. Based on the above situation, this paper has done some research on traffic light image detection and recognition, which has far-reaching significance in the field of auxiliary driving and self-driving. The technical scheme adopted in this paper first uses the initial segmentation of RGB space and HSV color segmentation to get the selection of traffic lights, and then uses the lightweight network model to identify traffic lights and finally get the information of traffic lights. The experimental results show that the recognition accuracy and speed of this scheme can meet the requirements of the actual driving process.

As an important tool of scope management, boundary pile is also an important basis for judging boundary violations and an important means for the supervision of industry authorities. However, the external environment where many boundaries are located is so bad that it is difficult to measure the manpower and material resources wasted in manual inspection and maintenance management every year. In view of this situation, a new type of boundary pile is designed in this paper, which can accurately locate the specific location of the problem by collecting GNSS (Global Navigation Satellite System) positioning information, and can also collect the surrounding state by carrying sensors, so as to save a lot of manpower and material resources for inspection.

This paper proposes a driving road slope measurement algorithm to provide a reference for adjusting the vertical angle of headlights. It uses a combination of Canny Edge Detection and Hough Transform to process images from camera to detect and fitting lanes. The angle of ramp can be calculated by Mathematical Model.

With the rapid development of society, the continuous expansion of cities an d the speeding up of urbanisation, the rapid growth and destruction of all kinds of boun dary pillars is a common occurrence. Up to now, there has been little product develop ment and application of intelligent electronic boundary pillars in China, and the cost of managing open boundary areas is high because many special boundary pillars are loca ted in harsh external environments that are difficult to inspect manually, thus leading t o untimely discovery of problems and even greater losses. Therefore, a digital boundar y pillar equipped with suitable software products and high-quality hardware that can a dapt to a variety of environments has become a major need in the current boundary pill ar market.

According to the principle of UPS, an AC sine wave online uninterruptible power supply based on STM32 is designed. The system adopts mains power, outputs corresponding DC power through isolation transformer, autotransformer and rectifier module, and then connects to single-phase full-bridge inverter circuit after boosting through booster module to supply stable AC voltage to the load, which significantly improves the load regulation rate, The voltage sampling module collects the output voltage of the boost circuit in real time and feeds it back to the single-chip microcomputer, and uses PID algorithm to realize the voltage closed-loop control, which improves the reliability of the uninterruptible power supply system. The system has functions such as overcurrent protection, instant power switching, and battery undervoltage alarm.

With the continuous improvement of the development level of the socialist m arket economy, education has been gradually pushed into a new historical period. The development of computer science and technology and the progress of language teachin g and learning methods, especially the rise of artificial intelligence, have opened up ne w ways for English language learning. The core of computer-assisted English language learning is speech recognition and evaluation technology, and speech recognition tech nology is the key. This paper takes deep learning as the main research perspective, co mbing and elaborating the concept, training process, specific models and applications of deep learning, in order to apply deep learning technology to English speech recognit ion system more maturely, thereby helping English learning Enhance the learning abili ty of English pronunciation.

Aiming at the problems of slow convergence speed and easy to fall into the optimal solution of ant colony algorithm, genetic algorithm and nonlinear optimization are used to optimize ant colony algorithm. After the initial iteration of the ant colony, the solution formed by all paths is the initial population, and then the genetic algorithm is used for selection, crossover and mutation to improve the ability of global search. Finally, the nonlinear optimization algorithm is used to increase the ability of local search of the algorithm. Through this improvement, the convergence speed of the ant colony algorithm is improved and the problem of easy to fall into the optimal solution is solved, which is applied to the traveling salesman problem.

In the past, the external costs, such as transportation costs and transit costs, are usually considered in the model of container sea-rail multimodal transport, without considering the time value cost of goods and the social costs such as traffic accident cost and environmental pollution cost caused by the transportation process. Goods transportation is a kind of social and economic activity. The whole transportation process of goods from the place of departure to the destination needs time. When time is connected with specific social and economic activities, it has value, that is time value. [1] The time value cost of goods will become a large part of the cost in the process of transportation. Improving the efficiency of goods transportation is a major goal of goods transportation. However, while improving the efficiency of goods transportation, we should also take into account the social costs brought about by the transportation of goods. In this paper, the time value of goods and social cost are considered to optimize the operation scheme of container sea-rail multimodal transport trains.

The continuous development of network technology in the information age has provided hardware guarantees for networked teaching, and the deepening of computer-assisted teaching research has promoted the development of networked computer-assisted teaching. With the popularization and application of informatization in colleges and universities, the use of modern information technology to change the traditional teaching mode has become the focus of attention. The C++ network-assisted teaching system uses computers and networks as the main means to develop network-based teaching and learning auxiliary teaching methods. This system is a mixed mode of network teaching and traditional teaching, which can enhance the communication between teachers and students, and can also play the role of the student-based teaching mode in network teaching.

Recently serverless cloud computing which was proposed by Amazon in 2015 is getting more favored by developers because of its charging model of pay-by-usage, applicability for fine-grained services and transparency of the servers to developers. Although a lot of study has proved that the performance and cost of serverless platform is more ideal than the traditional cloud computing service, the new metering model of serverless which depends on execution time, usage count and memory footprint is still unfamiliar to most developers and researchers. In this paper, we firstly make an introduction to the new metering model and the advantages of serverless over traditional cloud services. Then according to the cost affecting factors, we make a comprehensive survey of some constraints that mainly impact the cost of serverless and related methods to reduce cost in three levels—function level, container level and cloud platform level. Based on the analysis, we also propose some directions worthy of in-depth study in the future.

With the support of vehicular networking technology, the rating of UBI auto insurance rate has a certain guiding significance for accurate pricing of the rate and personalized demand. Based on CNN-Softmax algorithm, a rating model of UBI auto insurance rate is proposed in this paper. The model first performs a series of operations such as convolution, pooling, and non-linear activation function mapping on CNN algorithm to enable it to extract features based on UBI customers' driving behavior data, and then uses Softmax algorithm to classify customers according to their driving behaviors, thereby obtain UBI customer's auto insurance rate rating. The empirical results show that compared with, CNN algorith, BP neural network algorithm and SVM algorithm, CNN-Softmax algorithm has a higher discrimination accuracy in the risk assessment of UBI customers' driving behavior. And it is easy to implement in establishing rating model of UBI auto insurance rate. What's more, it also has a good robustness, which can achieve better grade evaluations.

According to the characteristics of strong coupling and highly nonlinear of manipulator, a trajectory tracking control method based on neural network is proposed. This paper makes full use of the neural network's self-learning characteristics, parallel processing ability, nonlinear mapping ability, fault tolerance and so on, and combines it with other control methods to design a controller which can improve the tracking performance of manipulator. The simulation results show that the control method can improve the effectiveness and accuracy of robot arm trajectory tracking.

When the SSD network uses an independent feature layer to detect objects, there is no connection between the layers, which leads to the problem of insufficient expression of contextual features. This paper proposes a pedestrian detection algorithm based on improved SSD network. The algorithm uses cross-layer feature adaptive fusion, and combines residual channel attention modules with different convolution rates of holes. While increasing the receptive field, this algorithm enhances important features and weakens unimportant features. Make the extracted features more directional, thereby improving the accuracy of pedestrian detection. Experiment on the improved network and algorithm on the INRIA pedestrian detection dataset, and the mixed pedestrian dataset extracted from the COCO dataset and the Crowd human dataset. The experimental results show that the average precision of pedestrian detection in the two datasets is improved by 1.7% and 4.0% respectively compared with the original network.

Phase-comparison of sum and difference angle measurement technique can be simply applied to uniform linear phased array radar. In contrast, it is not so simple for uniform circular phased array radar because the relation between ratio of difference to sum and target azimuth angle is complex. In this letter, we present this problem; based on derivation and approximate treatment, we propose an algorithm to solve this problem. The proposed algorithm is easy to implement and has good angle measurement performance. Computer simulations are shown to verify the efficacy of the proposed method.

Preceding vehicle detection is still a challenge for unmanned driving technology. Deep learning has achieved great success in target detection. Among them, the Faster R-CNN algorithm is a classic representative. However, the algorithm still has some room for improvement in detection accuracy. By analyzing the problems of Faster R-CNN in the detection of occluded vehicles, taking the target detection post-processing algorithm Soft-NMS as the research object, two new penalty coefficients, inverse proportional penalty coefficient and exponential penalty coefficient, were proposed. It further improves the algorithm's detection accuracy of the blocked vehicle in front.

Neutral section passing technology in railway power supply had always been a technical problem in current domestic traction power supply industry. In recent years, there are many technical schemes for passing neutral section. There are two main schemes of automatic neutral section passing at present: the vehicle control neutral section passing and automatic ground control neutral section passing, the automatic ground control neutral section passing is a relatively popular technical route in both schemes. Now incorporating fail-safe principle of signal specialty in automatic ground control neutral section passing and proposing the logic control system of automatic ground control neutral section passing based on the idea of "fail safe" in the interlocking system, and the risk identification and safety analysis are carried out based on the dynamic fault tree and HAZOP method. The fault tree is constructed based on the results of HAZOP analysis. Finally, the system reliability is quantitatively calculated through the fault tree.

In recent years, the application of neural networks in face recognition has improved its accuracy greatly. However, it still suffers from the disappearance of the network gradient and the training cost. In this paper, we propose a new hybrid method which combines the contrast divergence algorithm (CD) accelerated Restricted Boltzmann Machine (RBM) and the Integrated Learning Method (ILM) with the Boosting algorithm (BM). First, we use RBM to establish a minimum energy model of the sample distribution. Then we use CD to speed up the feature extraction of samples in RBM. And then we build a multi-classifier with heterogeneous samples for ILM by BM. Finally, we experience our proposed method on the ORL database and the AR database. The results show that, with a simpler neural network determined by CD-RBM and BM-ILM, the recognition accuracy is better than that without the proposed method.

China's large power grid enterprises often need to achieve optimal allocation of resources between different levels. This paper establishes a multi-level resource allocation model of large-scale power grid enterprises based on the difference of hierarchical demand, and puts forward the allocation strategy of multi-level resource optimal allocation, and thinks that it is necessary to establish and improve the cross level power market to promote the optimal allocation of resources at different levels of power grid enterprises.

The realization of the collection, transmission and processing of information is the three basic elements of a smart campus, among which data collection and transmission are inseparable from the participation of wireless sensor networks. In particular, the geographical area of the campus of colleges and universities is relatively large, the population is relatively dense, and the mobility and contact rate of personnel are high. Higher requirements are put forward for intelligent teaching management and campus safety management and control. In particular, the current large-scale outbreak of the new tube virus has been basically under control in our country, but because of the characteristics of college campuses, it is more likely to become a disaster area. Therefore, a large number of sensor nodes (monitoring, temperature, smoke, alarm, etc.) are deployed in the campus, and the emergency information of the school can be fed back and processed in time through the sensor network, thereby ensuring the safety of the campus. To build a large-scale sensor network, many technical issues must be considered. Among them, the time synchronization algorithm that can reduce network power consumption is one of the key technical indicators. This paper proposes time synchronization algorithm ideas from two aspects of cluster head synchronization and intra-cluster synchronization. It is applied to the time synchronization process of WSN nodes to reduce the energy consumption of WSN sensor nodes in smart campuses, thereby increasing the robustness of WSN networks.

Aiming at the problems of low control precision, poor anti-interference ability and poor stability of traditional PID controller, the improvement effect of fuzzy control and BP neural network in closed-loop control of PMSM is compared. In the case of variable speed and variable load, the stability, real-time, anti disturbance and robustness of PMSM system with traditional PID controller, fuzzy self-tuning PID controller and BP neural network PID controller as speed regulator are simulated and analyzed. The simulation results show that the fuzzy self-tuning PID controller has more advantages in the response speed of the system, and can make the system quickly recover to the initial state in case of sudden change. The BP neural network PID controller can ensure the relative stability of the system is better, and can restrain the sudden fluctuation.

With the development of sensor technology and computer vision, the speed of video generation is increasing exponentially. The workload of video browsing and the amount of stored data are increasing. In order to achieve fast browsing and efficient storage, video structure technology is increasingly important. This paper focuses on the video structure technology in the construction of smart cities, outlines the system process of video structure, and studies the extraction algorithm of human and vehicle target attributes based on deep learning. And combined with the actual video monitoring scene, the experiment is carried out. Compared with traditional image processing methods, the video structure target attribute extraction model based on deep learning has high detection accuracy and strong algorithm generalization ability. The intelligent analysis and processing of data by video structure technology is conducive to quickly increasing the speed of target search, reducing storage capacity, solving the problem of long-term storage, and realizing full automation of surveillance video.

Automatic Machine Learning (AutoML) uses automated data-driven methods to realize the selection of hyper-parameters, neural network architectures, regularization methods, etc., making machine learning techniques easier to apply and reducing dependence on experienced human experts. And hyper-parameter search based on automatic machine learning is one of the current research hotspots in the industry and academia. We mainly introduce the hyper-parameter search framework based on automatic machine learning and the common hyper-parameter search strategies. Combined with specific data sets, the classification accuracy of the model under different hyper-parameter search strategies is compared to find the model parameter configuration that can maximize the classification accuracy. Compared with the experience-based parameter adjustment method, the hyper-parameter search based on automatic machine learning can reduce labor costs, improve training efficiency, and automatically construct a dedicated convolutional neural network to maximize the model effect.

The single particle displacement damage effects and mechanisms in 8T Global Shutter CMOS image sensors (CISs) are studied. We provide radiation effects due to 129Xe ion irradiations of 8T Global Shutter CIS by the analyses of dark current spikes and dark current non-uniformity (DCUN). The experimental results show that low fluence irradiation-induced dark current distributions in 8T global shutter CIS exhibit a clear exponential hot pixel tail that appears difficult to match with cumulative radiation effect physical models. The degradation mechanism is a high electric field distribution exists at the overlap region between gate and pinned-photodiode (PPD). The emission rate of a defect can be dramatically enhanced via a high electric field. Irradiation-induced defects are the sources of the dark electron generation and the electric field acts as an amplifier.

In view of the current situation where the criteria for judging the infringement of art works are still vague and the court discretionary process is highly subjective, this article introduces the working principle of convolutional neural network (CNN) which is based on image recognition technology, and uses CNN algorithm to extract and store the characteristics of related works, And use the method of cosine similarity criterion to compare and calculate the degree of similarity, to assist in judging the standard of substantial similarity, and to improve the objectivity of the criterion for identifying copyright infringement of artworks

Linear phase-shift transformer is a new type of transformer, which can be used in multi-rectifier and multi-inverter system. Taking the design of a linear phase-shifting transformer for multiple superimposed inverters as an example, a layered magnetic field analysis model of the linear phase-shifting transformer is established. Firstly, the primary side slot is treated as a smooth structure, then the model is layered and the current density of primary side is calculated by the single Fourier method Taking into account the boundary conditions of the transformer, the distribution expressions of the vector magnetic potential and the magnetic flux density in the solution region are derived, the relative permeability function of the air gap is established and the influence of the slot effect on the air gap electromagnetic field is calculated. In this paper, the distribution of electromagnetic field without load is calculated. Finally, the feasibility of the method is verified by simulation, which provides basis for the following engineering calculation and optimization design of linear transformer.

In order to solve the difficult problem of partial discharge pattern recognition caused by large amount of partial discharge detection data and complex multi-source, a partial discharge pattern recognition algorithm based on VGG-16 convolution neural network is proposed. The parameters of VGG-16 network model are optimized in convolution layer, pool layer and connection layer by means of migration learning. The VGG-16 model is superior to LeNet-5 model and has higher recognition accuracy.

As a key economic index to measure the planning and management level of power grid company, line loss is not only the main content of completing the construction and transformation task of power grid company, but also an important assessment standard for the operation level of power grid company. In order to improve the generalization ability and prediction accuracy of extreme learning machine (ELM), particle swarm optimization (PSO) algorithm is applied to extreme learning machine (ELM), and a line loss prediction method based on PSO combined with ELM is proposed. The particle swarm optimization algorithm is used to select the optimal hidden layer deviation and input weight matrix to calculate the output weight matrix, so as to improve the accuracy and stability of the extreme learning machine. Experimental results show that the proposed method can effectively improve the accuracy of line loss prediction.

As one of the main application forms of energy coupling utilization, the electric gas integrated energy system has been widely concerned. State estimation plays an important role in the operation monitoring and optimal scheduling of the electric gas integrated energy system. The state estimation is the basis of energy management of the electro-pneumatic coupling network, which can provide a high-precision global consistent solution for the subsequent optimal scheduling and safety evaluation of the electro-pneumatic coupling network. The steady state estimation model of gas network is established for grid connected operation or island mode, and the state estimation of complex gas network is realized. On this basis, the coupled state estimation model of electricity and gas is further established, and its performance in state estimation accuracy and bad data identification is tested.

With the remolding of the automobile industry and the reconstruction of the value chain, the traditional automobile giants have embarked on the digital transformation and ecological layout of the new energy business, so the business model's innovation and reform are imminent. This paper makes a comparative analysis of the business models of Tesla and NIO by combing the current business models of automobile enterprises in the sales and use links with the four-elements model proposed by Professors Johnson and Christensen of Harvard University. Finally, based on some auto companies' current business models, this paper provides some new ideas for the business model innovation in the new energy business layout of Chinese auto companies.

Ship recognition is a key topic for intelligent shipping in canal. A new multi-decision algorithm based on Bayesian decision tree is proposed. The algorithm combine three Convolutional Neural Networks module - VGG, VGG19 and ResNet to get a better classify result. A quantization method is adopt to quantize the output of these network. A new method based on decision tree construct algorithm is proposed to improve the accurate of decision tree. In experiment, implement a various compared algorithms to prove the algorithm has better performance.

In the traditional digital identity authentication, there are potential security threats caused by the user's personal privacy data being too centralized and vulnerable to attack, and data centralization will lead to data redundancy, poor permeability, user's lack of master control, high overall cost, and potential threats of abuse of central rights. The digital identity authentication technology based on block chain can perfectly solve the problem of appeal. The main innovative work includes the design of a new digital identity ID and a new asynchronous Byzantine fault-tolerant algorithm dsbft using cryptography, computer network, asynchronous Byzantine fault-tolerant algorithm and other technologies, Block chain is designed based on ID and dsbft, and a decentralized identity authentication system is designed based on block chain. Hash algorithm is used for signature and acceptance to improve authentication efficiency and scalability. In order to solve the dust attack and quantum computation, the elliptic curve digital signature algorithm (ECDSA) encryption algorithm can be combined with other cryptography. The results show that the appeal identity authentication scheme can resist a variety of attacks and has high efficiency and security.

In the 21st century, with the development of information technology, the popularization of multimedia technology and application of network technology, teaching methods and education technology have undergone great changes. How to fully understand the characteristics of multimedia technology, how to correctly use it to assist language teaching, thus effectively improve the efficiency of foreign language learning, is worth exploring. On the basis of defining the meaning of multimedia teaching, this paper analyzes in detail the advantages of applying multimedia technology in language teaching, thus creating good conditions for improving the efficiency of language classroom learning.

The employment and Entrepreneurship of college students is an important aspect of people's livelihood. Under the background of "mass entrepreneurship and innovation", from the perspective of public management, this paper discusses and analyzes the current situation and construction strategies of College Students' Entrepreneurship service security system from the perspective of public management. This article mainly from the entrepreneurship skills training, funds and policy support measures, optimizing business environment, encouraging qualified college students to actively participate in the entrepreneurial tide, in order to provide certain reference for college students' Entrepreneurship service.

This paper introduces a set of professional data management system based on polymorphic graphics, which is established according to the requirements of software engineering management. It introduces the key technologies of the system, such as the main function module design, three-tier architecture quick design tool, online visio file and pdf file browsing. The system has strong practicability and good user interface. It can effectively improve the current professional data management mode.

With the continuous development of the two emerging fields of medical geography and spatial epidemiology, Geostatistical downscaling plays an important role in high-precision disease mapping and incidence prediction in a range of public health application This article combines the application cases of different diseases and lists a series of geostatistical downscaling methods, including Area-to-point and Area-to-Area Kriging, Area-to-point and Area-to-area Poisson Kriging, Area-to-point and Area-to- Area regression Kriging, Area-to-point and Area-to-Area co-kriging, Area-to-point and Area-to-Area binomial kriging, etc. By introducing the application progress of different methods in the field of public health, the advantages and disadvantages are compared. Finally, trends and some basic questions of methods proposed recently are discussed in the conclusion.

Nowadays, intelligent assistants, such as Apple's Siri, are becoming integral to our daily lives, and they significantly change the way users interact with systems. Meanwhile, with the development of technologies, people continue to bring the experience of using intelligent products to the car, and the in-car intelligent assistant is also gradually being used. While there have been many studies focused on technologies, there are fewer studies dealing with design issues at the initial stage of the design, the stage prior to system implementation, such as how the personality of intelligent assistant should be, how to design the appearance, and the way of interacting, the designers had to rely on imagination. Therefore, we present a design method to define intelligent assistants, using the MBTI model combined with brand tonality to define the personality of intelligent assistants, then we determined the design direction of the appearance design based on user research and benchmark. Furthermore, the initial design scheme was screened from brand fitting and flexibility, and experts and users were invited to determine the most suitable scheme from the two dimensions of personality and symbolization. This design method allows designers and engineers to explore more precise design solutions and to improve the emotional design and user satisfaction of intelligent assistants before all the complex systems are ready to run.

At present, we are in an era of rapid information development. The popularization of the Internet and mobile terminals makes information exchange more convenient. WeChat public platform, baidu, weibo, xiaohongshu and other platforms are still thriving. How to make the comic strip keep up with the trend of The Times, come into sight again, and become the popular spiritual and cultural products, is what the author wants to express in this article. Traditional paper comics can no longer meet people's needs for convenience, timeliness and interaction, and today's media era provides a broader platform for the development of comics. In the information age, everyone is creating images, reading images, and recording bits and pieces in the form of images, which has become a part of People's Daily life, and comic books can re-enter everyone's life in this way. It is the inheritance and development of traditional comics that endows them with the characteristics of the digital age, which can make them have the advantages that traditional comics do not have.

The integrated control system of a certain type of ship was imported from abroad, and its independent maintenance support system has been established in China. However, due to the complex internal structure of the system, there are problems in actual use such as high failure rate and insufficient number of replacement spare parts. The application of a large number of advanced technologies and specific encryption measures makes the maintenance very difficult, and the existing logistics support capability is far It is not enough to ensure the execution of its normal tasks and seriously affects the performance of the entire ship warfare technology. To this end, comprehensively analyze the factors affecting the logistics support capability of the system, and use the analytic hierarchy process to construct an index system for evaluating factors of the logistics support capability. On this basis, the fuzzy comprehensive evaluation method is used to construct an evaluation model, combining the actual use and maintenance of the equipment The validity of the model was verified. The results show that the lack of spare parts and the shortage of repair and debugging platforms are the main factors restricting the logistics support capability of the system. This comprehensive evaluation provides a reference for guiding the rational allocation and optimization of logistics support resources, improving the system's logistics support capabilities, and guiding the support construction of the same type of system, and has certain guiding significance for the later realization of the localization of the whole system.

With the rapid development of the financial market in the economic field, more and more financial problems need to be solved with mathematical knowledge. This article combines the development history and current status of each electrical appliance, from the three aspects of solvency, operating capacity and profitability. In terms of analyzing the company's financial ratio, the company's current ratio, asset-liability ratio, return on equity, and total return on assets of the company in the past three years were calculated.The analysis of these indicators revealed that the assets of Shanghai Liangxin Electric Co., Ltd. The debt ratio is extremely low, and the risk is small. However, in comparison with competitors in the same industry, we find that the company's profitability is at the middle and upper levels, and the overall development trend is good in the past three years, but it is generally higher than the industry's advanced level. Companies still need improvement.

Since the beginning of the 20th century, advanced business enterprises have focused on strengthening logistics management as a "third source of profit". In this context, that is, since the 1950s, the beginning of container sea transport, the superiority of container transport has been recognised worldwide, and become the main form of transport in domestic and foreign trade, the containerisation of such goods transport has become an irreversible direction of development. In the domestic occupy the transport of the mainstream of the car logistics is a greater demand for containers, in particular, can self-adjust the temperature of the container to achieve fresh demand, and the emergence of cloud servers and the Internet of things for this container design provides the possibility.

With the wide popularization of computer network at any time, students' computer technology operation level has been greatly improved than before, and students have more channels to obtain information. In the Internet environment, the development of high-quality student party members is an extremely important work for the party's construction, and has practical significance for the realization of the two centenary goals and the great rejuvenation of the Chinese nation.

To study the suitable heart rate of obese children with different physique in weight-loss exercise, so as to provide scientific and effective exercise for children. In this paper, a stochastic near parameter estimation and identification algorithm based on running of obese children is proposed. Based on the obese children's weight-loss exercise,a system of random near parameter estimation and identification is established. Finally, through the MATLAB simulation analysis, the results show that the method is suitable for children's running safety heart rate and physical quality. The simulation results verify the feasibility and effectiveness of the method.The obese children's weight-loss children's running exercise has a strong reference in physical exercise.

With the acceleration of urban construction, the urban electricity load has increased rapidly, and electricity shortages have also appeared in various places. However, the rapid development of cities has resulted in a shortage of land. The location of substations and the selection of transmission lines are particularly important in planning. In this paper, distributed power transmission and transformation project to optimize network access with locating and sizing problems, establish an annual loss of power as the objective, and consider the output load timing characteristics of multi-period optimal power flow model. Use representative scenes to simulate the situation throughout the year, and give a method for determining scenes and their weights.

This paper introduces the importance of cultivating pupils'oral arithmetic ability. This paper designs and realizes students' score management system based on Java technology. The system can automatically give oral arithmetic questions. When users answer questions, it can automatically judge whether the answers are correct or not, and put the questions they have done into the combo box for users to review.

Artificial intelligence technology is a new technology based on computer science and technology, combined with Internet technology. It has been widely used in all walks of life. With the development of information technology, the coverage of the Internet is wider, more and more users are involved, and the transmission and sharing of information is more and more convenient. This makes people's request for information security rise to a higher level. The arrival of the big data era has put forward higher requirements for the security protection of the information technology security personnel for the complex data. Artificial intelligence technology can solve this problem well in the application of big data security defense. Artificial intelligence technology can not only enhance the security of network information, but also will change the way people live in the future with the wider application in the network, which will profoundly affect people's life in the information age.

In order to solve the existing problems for the backup of massive unstructured data with its performance bottleneck of single server, in this paper we propose a distributed method for the backup of massive unstructured data. In the backup method, through using a combination of load balancing algorithm and distributed algorithm, the backup system assign task of backup to each production server. Multiple servers backup massive unstructured data stored in shared storage to the backend server. The backup system can make full use of the performance resources of servers in environment production, reduce the single server performance bottlenecks, to avoid the single server hunger, increased the speed of backup.

In recent years, with the rapid development of Internet technology, more and more online learning methods have emerged, and the main body introduced in this article is the integration of the original offline data and the status quo of online learning and database development. Under the premise of, under the review requirements of undergraduates at the end of the direction, the design of learning materials management system.

Socket based network communication is a flexible, easy to implement, low cost method. It can run on a variety of networks that use THE TCP/IP protocol as the communication protocol. So the design in VC++ environment, through the Socket to establish the original Socket, monitoring network traffic, and suspicious data through Access database records, to provide reference for administrators, so as to achieve network security.

In the era of information revolution, China's Internet construction has achieved unprecedented development. However, in the current public WiFi, the proportion of phishing WiFi with high risks is very high. This paper designs and implements a tool to detect wireless network security. It starts with the following: attack, defense, and detection. Firstly, based on the key technology of wireless attack, five functional modules of wireless attack system are designed. Then, by analyzing the shortcomings of traditional detection methods, a lightweight phishing attack detection scheme for the public is explored. The scheme is based on TMM algorithm and HDT algorithm. Finally, a phishing attack detection tool for operator network is designed and implemented based on the scheme, which can provide consumers with basic user information to connect WLAN, and detect the network effectively.The results show that all functions of the detection tool designed in this paper can run normally. It can meet the detection requirements of wireless phishing AP in the current environment and achieve the expected purpose, which has extensive market value and good scientific significance.

Seeker is the core of precision guided weapon. The performance of its servo mechanism has a decisive impact on the accuracy of guidance weapon. The controller design of seeker servo system is of great significance to improve the function of seeker servo system. The mathematical model of the seeker servo system in the form of transfer function is built. The traditional PID control and BP neural network PID control methods are used to design the controller respectively. Aiming at the limitation of BP neural network, the BP neural network is improved. Finally, the simulation experiment is carried out in Matlab/Simulink. Simulation results show that the neural network PID controller has better performance than the traditional PID controller, so that the system has better adaptability and robustness.

In the field of medical image processing, the convolution neural network (CNN) algorithm in depth learning solves many problems in machine vision.Aiming at the problems of low resolution of medical images and easy misjudgment of artificial recognition, in order to help medical workers to better identify the lesion area, this paper discusses the structure optimization, training and prediction ability of CNN, and trains the convolution neural network to judge the lesion area in the image.and get more accurate discriminant results on lung images, further verify the high efficiency, accuracy and reliability of convolutional neural network algorithm in medical image field.

In recent years, the learning method based on depth convolution neural network has achieved unprecedented results in image denoising, and it has become a research hotspot to obtain better image denoising effect by adjusting network structure and parameters. Noise reduction convolution neural network adopts residual learning method in depth neural network, which not only improves the effect of noise reduction, but also solves the problem of blind noise reduction to a certain extent. Its disadvantage is that the algorithm has a long convergence time. In this paper, the structure of noise reduction convolution neural network is further improved, and an image denoising algorithm based on deconvolution neural network is proposed. The main features of this paper are as follows: 1) in the original network structure, deconvolution neural network is introduced to optimize the residual learning method; 2) A new loss function calculation method is proposed. BSD68 and SET12 test data sets are used to verify the proposed method. The experimental results show that the noise reduction performance of this algorithm is compared with the noise reduction convolution neural network algorithm. Under the same noise reduction effect, the convergence time of this algorithm is shortened by 120%. 138%. At the same time, compared with the traditional deep learning image denoising algorithm, the denoising effect and running efficiency of this method are also improved.

Sentiment classification is a crucial problem in natural language processing and is essential to understand user opinions. There are two main approaches to solve this problem, one is the classified-based method, the other is the lexicon-based method; however, both methods perform not well on the long-sequence methods, and each method has its advantages and disadvantages. This paper introduced a new method called Lexiconed BERT, which cream off the best and filter out the impurities from the above two methods. The evaluation shows that our model achieves excellent results in the long sequence sentence and reduce resource consumption significantly.

Aim to solve the balance problem of robot production line, an improved double population genetic algorithm is proposed. The priority relation of production process is treated as the constraint condition, the selection function of crossover probability and mutation probability is designed according to the individual fitness of the population. Then the beat optimization algorithm is used to optimize the production rhythm of a switch shell welding line. The simulation result show that the balance rate of this robot welding line is improved and the production rhythm is optimized.

Aiming at the relationship between competition and cooperation among sensors in multi-target tracking task allocation, this paper proposes a game theory-based method to allocate sensor resources for multi-target tracking tasks. The dissertation first establishes a target tracking model and designs a game theory payment function based on the model. Then a multi-sensor multi-target tracking task allocation game model is designed. Finally, a multi-target tracking task allocation algorithm based on game theory is given. Compared with other allocation algorithms, the average tracking error of the proposed allocation algorithm is reduced from 68.5630m to 19.7601m through experiments.

In recent years, the Content-Centric Networking (CCN) has attracted much attention from the global Internet experts, especially it has demonstrated the outstanding effect on the application in the field of Internet of Things (IoT). At present, the routing technique of ICN is subjected to the dynamic change of network environment with the development of mobile Internet. Therefore, this paper proposes an Intelligent CCN routing strategy based on Bacterial Quorum pattern (ICBQ). The ICBQ tries to simulate the behaviors of bacteria, including quorum sensing and adaptive chemotaxis. Meanwhile, the quorum sensing can obtain the parameter information on bandwidth, delay and error rate to facilitate the subsequent forwarding of packets; The adaptive chemotaxis can select the optimal interface to forward the packets through the information measurement. The simulation is driven based on the real Netflix dataset over the GTS network topology, and the experimental results show that the proposed ICBQ has better performance in terms of routing success rate, routing delay and load balance.

Image processing technology has increasing important applications in sports in recent years, such as result prediction, technical analysis, and foul determination. For most billiards beginners, how to hit the billiard ball into the ball belt is not a simple matter. In order to assist the beginners in billiards training, this article proposes an assisted hitting system. The system is divided into three parts: bag detection, billiard detection, and hitting suggestions. In the ball belt detection, double-peak histogram threshold segmentation, edge detection and Hough transform are used to detect the edge line, and then the ball belt position is detected. In billiard ball detection, the normalized RGB method is used to extract the foreground, the Hough transform circle detection method is used to detect the center and radius of the sphere, and the neural network is used to determine the white ball. In the shot suggestion, the force and angle of the shot are given by the dynamic formula, and the path simulation of the ball in the background is given. Experiments show that the recognition method is effective and the path simulation is consistent with reality, which can assist beginners in the training of billiards to a large extent.

The potential and electric field responses in cylinder cavity with ceramics slice caused by distributed charge in X-ray environment were simulated with finite difference method. The numerical results indicate that the distributed charge in ceramics' middle section is negative or positive, and positive of both sides. For the photon energies from 30 keV to 1 MeV with same fluence, distributed charge and electric field were decreased at first and then increased. Electrode has great effect to the potential and electric field responses. For different plating thickness, there exists minimum to the electric field amplitude, and for the plating, there exist optimal diameter and thickness. Distributed charge is calculated using Monte-Carlo particle transport code.

Blockchain is a new application model that integrates distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, and other computer technologies. These technologies make blockchain traceable, reliable, and resistant to modification, characteristics that suit the need for data storage and exchange. As a result, in this paper, we design an energy internet big data system based on energy and blockchain technology systematically. Firstly, we discuss the basics that support for energy scheduling and policymaking of the management department, including energy data coordination, enterprise monitoring, and consumption monitoring, and trading services. Secondly, we develop the Unified Energy Reporting Service. It can generate the energy demand and supply table, production and sales table, demand forecast table, energy consumption monitoring table, environmental protection monitoring data, and energy data of critical customers. Finally, we ensure the storage and collection of evidence in our system, which would guarantee the electric security of the Winter Olympic Games.

As the "Outline for Building a Leading Transportation Nation" puts forward, the overall requirement of high quality development, the measurement work of transportation industry plays a much more important role. This paper introduces the measurement technology state of the FWD of the highway pavement bearing capacity measuring equipment, the technical state and error analysis of the equipment are introduced, and the calibration and comparison methods are sorted out, which provide a reference for the next generation of the equipment.

In recent years, with the enrichment of people's material life, the demand for various competitive sports has also increased, and the racing industry is one of them. In a racing competition, manual polishing of the steel pipe grooves of the racing frame is a task with huge workload, harsh working environment and long working cycle, which is laborious and laborious. The research direction of this paper is to develop a semi-automatic steel pipe bevel cutting machine to reduce the workload of steel pipe bevel grinding of racing car frame.

An improved ISAGWO algorithm is proposed to solve the problem that the traditional grey Wolf algorithm is easy to fall into local optimum, which leads to low search accuracy. By introducing the weight factor before the vector coefficient, the position vector updating formula and the adaptive search strategy of the dynamic modulation algorithm, the search accuracy of the algorithm and the ability of jumping out of the local optimum are improved. The efficiency of the improved grey Wolf algorithm is evaluated by statistical analysis, convergence rate analysis and classical reference function. The results show that the improved grey Wolf algorithm has better global search ability and solution robustness, and the optimization precision and convergence speed are also improved compared with the standard grey Wolf algorithm.

The three-stage Clos switching network is a kind of multi-stage switching network which is commonly used to build a high-speed and large-capacity switching network, and the scheduling algorithm of the switching network affects the performance of the switching network. The article introduces and analyzes the research ideas of the scheduling algorithm of the three-stage Clos network, and provides a reference for further research on high-performance scheduling algorithms.

With the development of society, the progress of science and technology, the level of computer has been improved. Digital imaging technology has also achieved unprecedented development. The most commonly used digital cameras are Charge-coupled device (CCD) and Complementary Metal Oxide Semiconductor (CMOS) image sensors. Bayer filters are often used to help sensors capture color information. However, only one component can be achieved per pixel. Full-color images must be obtained by color interpolation. Therefore, the color interpolation algorithm has attracted the attention of the academic community. In this paper, three classical color interpolation algorithms- Bilinear interpolation algorithm, COK Algorithm, Gradient Interpolation Algorithm- are compared and studied. The Optimal Algorithm is selected by comparing false color, noise and color moire fringe. After comparison, it is found that the three basic algorithms will have color moire fringes appear. But the gradient interpolation algorithm has the least moire fringe and noise. COK Algorithm, second of all. But the COK Algorithm has color overflows.

The rapid growth of my country's civil aviation passenger traffic has led to increasingly tight operating resources in the terminal. In order to optimize the operating efficiency of the baggage turntable for arriving passengers, the simulated annealing algorithm is used in the baggage claim carousel allocation. According to the set objective function model, the simulated annealing algorithm solution process is designed and the solution process is simulated. According to the comparison between the obtained results and the results of the first-come-first-served distribution method, it is shown that the simulated annealing algorithm has a better result than the first-come, first-served distribution method and meets the target requirements. It provides a way to alleviate the airport baggage claim carousel allocation problem and a feasible way to use efficiency.

In order to accurately determine the stopping state of a vehicle, a wheel location detection method combining enhanced images with CenterNet network was proposed.Image enhancement network was used to process the detected images to improve the contrast and sharpness of images. Then, CenterNet network was used to detect the wheel position in the enhanced image.The algorithm proposed in this paper can efficiently and accurately detect the vehicle wheel location, thus determining a variety of vehicle stopping states, so as to avoid the scene complexity limit, camera Angle and other factors that lead to the wrong judgment of stopping states, and greatly improve the efficiency of roadside parking management.

With the aging of the population, the situation of pension problems is getting more and more serious. Aiming at the problems of old people forgetting to take medicine, taking wrong medicine, etc., a smart reminder medicine box that can be divided into medicine is designed to assist the old people to take medicine on time and take medicine quantitatively. The use of mechanical design to realize the convenience of adding medicines, the use of single-chip microcomputers to realize time reminders and control of the amount of medicines, and at the same time to realize the collection of information on the amount of medicines, can assist doctors in diagnosis and help businesses understand the market.

Due to the uncertainties, constraints, nonlinearity and the correlation among variables in the control of hydropower industry, it is difficult to obtain the accurate mathematical model. In this paper, a generalized predictive identification algorithm based on the nonlinear system in the control field of hydropower industry is presented, and a model of the generalized predictive identification algorithm based on the nonlinear system in the control field of hydropower industry is established AB simulation. The simulation results show that the algorithm is robust, can effectively overcome the system delay, and can be applied to the open-loop unstable non minimum phase system. It has strong applicability in the field of electric industry control and can produce good economic benefits.

A method is proposed for constrained multi-objective optimization problems, which is based on Newton method and the linear weighted sum method. The automobile impact problem is studied in this paper. The multi-objective optimization model of the impact problem would be established. Then, an improved design of front rail is obtained by the algorithm proposed.

The seated comfort is one of the important parts in vehicle design. The research of the second row's head comfort and buttocks comfort of a hybrid electric vehicle is conducted by multi-object optimization method in this paper. The surrogate functions of the performances are obtained by design of experiments, and the Pareto optimal front is worked out employing a multi-object optimization algorithm. In the end it proved great improvement in the level of the seated comfort design, and the improved design performed significantly better than the initial one.

The automatic adjustment of the driving posture can effectively help improve the driver's experience, and thus it is one of the important reference indicators for the design of the vehicles. This paper focuses on the intelligent adaptive driving posture using the machine learning (ML). Firstly, laser scanning was used to obtain the point cloud data of the most sold vehicles in the market. Then, the driving posture's key parameters were screened and extracted through the big data processing method. Finally, the deep learning neural network (DNN) was applied to establish a supervised learning model to figure out the intelligent adjustment of driving posture for different vehicles. Numerical results can demonstrate the accuracy and effectiveness of the proposed method by comparing it with the actual driving experiments. The results show that the accuracy of the research is good and can provide a reference for the design of intelligent driving.

A harmonic injection method is proposed for reduce permanent magnet synchronous motor torque ripple and noise, which is based on harmonic model of PMSM. The effectiveness of the method is verified by simulation and experiment. It is proved that this method can significantly improve NVH performance of the motor.

In order to design a kind of intelligent mechanical claw for home and office scene. Based on the investigation, analysis and arrangement of the features of items in the family and the functional demands of the family on the mechanical claw, a design scheme of intelligent mechanical claw which can switch functions and grasping modes according to the different grasping objects is proposed. Mechanical structure design, including mechanism motion design, part structure design, and use Autodesk Inventor software for mechanical parts, component design and simulation assembly; The stress analysis module in Autodesk Inventor was used to simulate the key components. Finally, the design scheme of mechanical structure and control system for the mechanical claw are proposed.

With the development of the times and the progress of society, the popularization rate of computer network technology and information technology in China is accelerating, and the Internet of things technology also appears in people's vision, and is gradually known by people. In the context of this era of big data, it has brought great challenges to all walks of life. The development of everything must conform to the development theme of the times. In order to meet the challenge of the research and development of the data asset management system of the graph database in the new era, this paper puts forward the method of applying the Internet of things technology to the research and development of the data asset management system of the graph database. Combining with the foreign research and development plans of the data asset management system of the graph database, the data resources of the graph database are carried out from the platform system, the management structure and the organization arrangement Based on the research and analysis of production management system, a research scheme of data asset management system of graph database which can meet the development requirements of the new era is worked out. Through long-term research and analysis, we can find that the Internet of things technology analysis method proposed in this paper can effectively provide new development ideas for the research and development of data asset management system based on graph database under the Internet of things technology.

Based on the technical specifications of smart meters, the electrical performance evaluation methods of single-phase smart meters are studied, and the accuracy, stability and safety of smart meters are analyzed and studied through specific data verification. Analysis the basic characteristics of the electrical performance of the smart energy meter, and evaluate the electrical performance based on the basic characteristics. Collect the specific working characteristic data of the intelligent electric energy meter, accurately measure the parameters of the intelligent electric energy meter, and analysis the electrical performance. Analysis the influence of the harmonic components of the power system on the measurement of the electric energy meter, study the power consumption of the smart electric energy meter, and the accuracy and stability of the operation.

Cr8Mo2SiV cold work die parts cracked early during machining process. In order to find out the reason of cracking failure of the die parts, the macro crack analysis, chemical composition analysis, hardness test and microstructure analysis of the parts were carried out by stereo microscope, direct reading spectrometer, digital Vickers hardness tester, metallographic microscope and scanning electron microscope. Some results are obtained: the chemical composition of cracked parts was tested and found similar to Cr8Mo2SiV, compared with standard Cr8Mo2SiV, it contains 3.81%W, and its hardness is 229 HV1, means that the cracked parts are not hardened. By observing the micro morphology of the crack, there are oxide and decarburization layer on the crack surface of the parts. It proves that the crack was formed before spheroidizing annealing. It is a typical raw material crack.

OpenMP supports parallel incremental development, and has become a mainstream parallel programming standard for shared memory systems. A parallel compiler running on multi-core DSP is designed for OpenMP programs. The main achievement includes translator and runtime. The translator converts OpenMP instructions in source files into function calls in runtime where the specific implementation is provided. The core problem of parallel compiler is how to design parallel strategy to allocate computing tasks to each core, which corresponds to the concept of parallel domain in OpenMP standard. The compiler realizes the of master-slave core by transforming parallel instructions and supporting runtime, which has guiding significance for the design of parallel compiler.

With the advancement of the construction of the China southern regional frequency regulation auxiliary service market, the calculation method of the AGC performance index of the generator set needs to be further optimized. To this end, an optimization scheme for calculating the AGC performance index of generator sets that is suitable for the China southern regional frequency regulation auxiliary service market is proposed. First, the demand for AGC unit participation in the frequency regulation market is analyzed, and a model for AGC unit participation in the frequency regulation market is established. Through theoretical analysis and data comparison, it is shown that the traditional calculation method of AGC frequency modulation performance index is not suitable for the calculation of AGC performance of FM market units. Aiming at the key content of the frequency modulation market for the AGC performance requirements of the unit, a suitable calculation method is proposed. Finally, the actual case that illustrates the applicability of the index calculation method in this paper is analyzed.

In view of the fact that traditional substation maintenance operations need to carry a large amount of paper data and the on-site paperlessness is low, a mobile terminal supporting intelligent maintenance of substations has been developed. Introduce the development background, system integration scheme, various management functions of the device and its application in substation maintenance operations, recorder viewing and grid fault analysis. The application of the mobile terminal can improve the intelligence level and convenience of on-site maintenance operations, shorten maintenance time, improve maintenance quality, and improve the level of relay protection lean management.

With the rapid development of digital technology, the advantages of the multiple interactive teaching model of college English have become increasingly prominent. The multiple interactive teaching model of college English focuses on teacher-student interaction, student-student interaction, student-teaching material interaction, and human-computer interaction. The research analyzes the characteristics of the multiple interactive teaching mode compared with the traditional teaching mode, and also points out some of the problems and challenges it faces, and compares the experimental class with the control class in specific teaching practices, and draws the conclusion: The multi-interactive English teaching mode is conducive to improving students' English learning performance, and is a practical and effective teaching mode. At the same time, the thesis briefly introduces the creation of Web-based interactive teaching system, and according to the characteristics of Web network teaching, expounds the selection of the system architecture and methods to improve the security and stability of the Web network teaching system.

We have witnessed many firms use social network to facilitate their business transactions. In this study we discuss the impact of social network on new ventures' risk control. We test our hypotheses with survey data collected from 3149 from a questionnaire survey about private businesses in China. Our results show that social network is the protective umbrella to prevent risk during their growth.

With the rapid development of artificial intelligence, face detection technology is widely used in our daily lives, such as mobile payment, video conferencing and personal identification. However, face the scenario while the face been blocking or crowding, the face detection accuracy would be greatly reduced. Therefore, in this paper, a high-precision face detection algorithm based on deep residual network has been proposed to solve this issue. Firstly, adding neural framework branches based on the Resnet-50 to improve the detection accuracy. Then import the soft-NMS method to enhance the robustness of algorithm. Experimental test on the public data set FDDB, the results indicate that the accuracy of this algorithm can reach 94.2% with good robustness, both accuracy and speed are better than the previous algorithm.

Organizing nonstop train is one of the effective ways to organize traffic flow. This article introduces the conditions, principles and schemes of the nonstop transport organization. On this basis, this paper establishes a mathematical model with the goal of minimizing the total consumption of traffic in the loading and unloading areas and during operation. This model is constrained by the uniqueness of the plan and the restriction of direct traffic flow. This paper constructs an example to test the actual application of the model, verify its effectiveness, and explain the advantages of organizing nonstop train.

Energy conservation in ship operation has become an important measure to reduce global greenhouse gas emissions. Therefore, it is necessary to carry out effective measurement of the energy efficiency of operating ships, so as to lay a foundation for their energy consumption. Based on this, this study designed a measurement method of energy efficiency of operating ships based on data mining. On the basis of analyzing the energy efficiency characteristics of operating ships, the paper analyzes the vessel resistance and the transfer of energy demand. Based on this, the energy efficiency measurement algorithm of operating ships is designed. By standardizing the energy consumption value of ships during navigation, the energy efficiency conversion of operating ships can be suppressed and utilized to avoid excessive energy consumption of ships during navigation and effectively control the Greenhouse gas emissions during navigation.

A novel robust sparse recovery (SR) space-time adaptive processing (STAP) algorithm based on particle swarm optimization (PSO) for non-stationary clutter suppression is presented in this paper. A cost function for PSO in the presence of parameter errors is theoretically derived. An improved estimation process of clutter spectrum based on this cost function which is called PSO-SR is proposed and analyzed. A more accurate estimation result of clutter spectrum could be provided by this algorithm than the previous proposed algorithms in the presence of considerable parameter errors. Simulation results demonstrate the robust performance of this algorithm.

IOP Publishing Limited ("IOPP") is retracting this paper following an investigation in accordance with COPE guidelines. which has confirmed translated plagiarism.

This work was originally published in Chinese [1] and has been translated and published without permission or acknowledgement to the original authors. IOP Publishing Limited request any citations to this article be redirected to the original work [1].

IOPP have made multiple attempts to contact the author but has not received a response. The author is encouraged to contact jpconf@ioppublishing.org.

[1] LIN Min-hong, MENG Zu-qiang. Multimodal Sentiment Analysis Based on Attention Neural Network[J].Computer Science, 2020, 47(11A): 508-514.

Retraction published: 23 April 2021

This article has been retracted by IOP Publishing following an allegation that raises concerns this article may have been created, manipulated, and/or sold by a commercial entity. In addition, IOP Publishing has seen no evidence that reliable peer review was conducted on this article, despite the clear standards expected of and communicated to conference organisers.

The authors of the article have been given opportunity to present evidence that they were the original and genuine creators of the work, however at the time of publication of this notice, IOP Publishing has not received any response. IOP Publishing has analysed the article and agrees there are enough indicators to cause serious doubts over the legitimacy of the work and agree this article should be retracted. The authors are encouraged to contact IOP Publishing Limited if they have any comments on this retraction.

Retraction published: 9 September 2022

This article has been retracted by IOP Publishing following an allegation that raises concerns this article may have been created, manipulated, and/or sold by a commercial entity. In addition, IOP Publishing has seen no evidence that reliable peer review was conducted on this article, despite the clear standards expected of and communicated to conference organisers.

The authors of the article have been given opportunity to present evidence that they were the original and genuine creators of the work, however at the time of publication of this notice, IOP Publishing has not received any response. IOP Publishing has analysed the article and agrees there are enough indicators to cause serious doubts over the legitimacy of the work and agree this article should be retracted. The authors are encouraged to contact IOP Publishing Limited if they have any comments on this retraction.

Retraction published: 9 September 2022

This article has been retracted by IOP Publishing following an allegation that raises concerns this article may have been created, manipulated, and/or sold by a commercial entity. In addition, IOP Publishing has seen no evidence that reliable peer review was conducted on this article, despite the clear standards expected of and communicated to conference organisers.

The authors of the article have been given opportunity to present evidence that they were the original and genuine creators of the work, however at the time of publication of this notice, IOP Publishing has not received any response. IOP Publishing has analysed the article and agrees there are enough indicators to cause serious doubts over the legitimacy of the work and agree this article should be retracted. The authors are encouraged to contact IOP Publishing Limited if they have any comments on this retraction.

Retraction published: 9 September 2022