CN117933701B

CN117933701B - Rail transit engineering construction safety risk monitoring method and system

Info

Publication number: CN117933701B
Application number: CN202410044755.5A
Authority: CN
Inventors: 袁梦; 肖云; 徐华; 刘昌�; 黄杰; 周家源
Original assignee: Guangzhou Metro Construction Management Co ltd; Guangzhou Metro Group Co Ltd; China Railway Construction South China Construction Co Ltd
Current assignee: Guangzhou Metro Construction Management Co ltd; Guangzhou Metro Group Co Ltd; China Railway Construction South China Construction Co Ltd
Priority date: 2024-01-11
Filing date: 2024-01-11
Publication date: 2024-07-26
Anticipated expiration: 2044-01-11
Also published as: CN117933701A

Abstract

The invention relates to the technical field of rail traffic engineering, in particular to a method and a system for monitoring the safety risk of rail traffic engineering construction. The method comprises the following steps: carrying out real-time image acquisition on the track engineering zone so as to acquire track engineering zone image data; carrying out region mode identification on the image data of the track engineering zone so as to obtain the topographic feature data; future weather prediction is carried out on the track engineering zone according to the topography feature data, so that weather prediction data are obtained; acquiring geological exploration report data of a track engineering zone; and carrying out flood control and drainage capacity evaluation on the track engineering zone according to the weather prediction data and the geological exploration report data, thereby obtaining flood control and drainage evaluation report data. The invention can comprehensively monitor and evaluate the safety risk in the track engineering construction process.

Description

Rail transit engineering construction safety risk monitoring method and system

Technical Field

The invention relates to the technical field of rail traffic engineering, in particular to a method and a system for monitoring the safety risk of rail traffic engineering construction.

Background

Track traffic engineering construction is a complex and dangerous task, including the design, construction and maintenance of track traffic systems such as railways, subways, trams, etc. In the construction process, various safety risks exist, such as construction accidents, equipment faults, casualties, natural disasters and the like. Therefore, monitoring and assessing security risks in engineering construction in real time is of paramount importance.

Currently, security risk monitoring for rail traffic engineering relies mainly on traditional empirical methods, which often fail to identify and evaluate risk comprehensively. Furthermore, monitoring systems are often decentralized and lack integration, making real-time decision making and risk mitigation difficult.

Disclosure of Invention

Based on this, the present invention needs to provide a method for monitoring the safety risk of track traffic engineering construction, so as to solve at least one of the above technical problems.

In order to achieve the above purpose, the track traffic engineering construction safety risk monitoring method comprises the following steps:

step S1: carrying out real-time image acquisition on the track engineering zone so as to acquire track engineering zone image data; carrying out region mode identification on the image data of the track engineering zone so as to obtain the topographic feature data;

Step S2: future weather prediction is carried out on the track engineering zone according to the topography feature data, so that weather prediction data are obtained; acquiring geological exploration report data of a track engineering zone; carrying out flood control and drainage capacity evaluation on the track engineering zone according to weather prediction data and geological exploration report data so as to obtain flood control and drainage evaluation report data;

step S3: carrying out intelligent risk early warning on the track engineering zone according to weather prediction data, geological exploration report data and flood control drainage evaluation report data so as to obtain risk early warning report data;

Step S4: when the risk level data are determined to be high risk level data, carrying out refuge planning on the track engineering zone by using a preset auxiliary decision module, so as to obtain refuge distribution data; the risk early warning report data and refuge site distribution data are sent to constructors and nearby residents through a preset cloud monitoring platform so as to guide the residents and constructors to evacuate;

In the invention, the image data of the track engineering zone is acquired through real-time image acquisition. Such data may provide real-time visual information to aid in monitoring and detecting potential security risks. The region pattern recognition is to analyze and process the track engineering zone image data to extract the topographic feature data. Such characteristic data may provide useful information about the regional environment, such as building distribution, road network, etc., to facilitate subsequent risk assessment and early warning. Future weather predictions may provide information regarding weather conditions that may occur in the track engineering zone, such as rainfall, wind, etc. This is critical for assessing weather related risks such as floods. Acquisition of geological survey report data may provide geological information about groundwater level, soil stability, etc. This is important for assessing problems such as geological risk and soil liquefaction. Flood control drainage capacity assessment in combination with weather forecast data and geological survey report data, it may be assessed whether the flood control drainage system of the track engineering zone is adequate to cope with possible flood risks. This helps to take necessary safety measures in advance, reducing the impact of floods on engineering construction. According to weather forecast data, geological exploration report data and flood control drainage evaluation report data, intelligent risk early warning can be conducted. By comprehensively analyzing the data, potential safety risks can be identified, and early warning signals can be sent out timely so as to take corresponding countermeasures. When the risk level is determined to be high, the refuge planning can be performed by using a preset auxiliary decision-making module. This allows for the determination of a suitable refuge site distribution for use by constructors and nearby residents in refuge in emergency situations. And the risk early warning report data and refuge site distribution data are sent to constructors and nearby residents through a preset cloud monitoring platform, so that relevant guidance and information can be provided. Through a preset cloud monitoring platform, risk early warning report data can be timely sent to constructors. This helps to guide constructors to take necessary countermeasures, reduces the influence of risks on construction work, and thus ensures the safety of constructors. In conclusion, the method also carries out intelligent risk early warning and emergency guidance on the track engineering zone by utilizing artificial intelligence, auxiliary decision making, early warning pushing and other technologies, thereby timely reminding and guiding constructors and residents to take corresponding precautionary measures and reducing risk loss and casualties.

Preferably, the present invention also provides a track traffic engineering construction safety risk monitoring system for executing the track traffic engineering construction safety risk monitoring method as described above, the track traffic engineering construction safety risk monitoring system comprising: :

The image acquisition and identification module is used for carrying out real-time image acquisition on the track engineering zone so as to acquire track engineering zone image data; carrying out region mode identification on the image data of the track engineering zone so as to obtain the topographic feature data;

The urban flood control evaluation module is used for carrying out future weather prediction on the track engineering zone according to the topography and topography characteristic data so as to obtain weather prediction data; acquiring geological exploration report data of a track engineering zone; carrying out flood control and drainage capacity evaluation on the track engineering zone according to weather prediction data and geological exploration report data so as to obtain flood control and drainage evaluation report data;

the urban intelligent risk early warning module is used for carrying out intelligent risk early warning on the track engineering zone according to weather prediction data, geological exploration report data and flood control drainage evaluation report data so as to acquire risk early warning report data;

the urban refuge planning module is used for planning refuge places of the track engineering zones by utilizing the preset auxiliary decision-making module when the risk level data are determined to be high risk level data, so as to acquire refuge place distribution data; the risk early warning report data and refuge site distribution data are sent to constructors and nearby residents through a preset cloud monitoring platform so as to guide the residents and constructors to evacuate;

the invention processes and analyzes the acquired image by real-time image acquisition and image recognition technology. The module can automatically identify and extract key information in the image, such as construction sites, equipment states, personnel activities and the like. Thus, real-time image data of the track engineering zone can be timely obtained, and basic data are provided for subsequent risk assessment and early warning. The urban flood control evaluation module can evaluate the flood control and drainage capacity of the track engineering zone by combining weather prediction data and geological exploration report data. The system can analyze the information of the topography and the topography of the urban area, the sewage pipe network, the drainage ditch and the like, comprehensively consider the factors of rainfall, topography, water flow and the like, and predict the possible flood risk and drainage problem of the urban area. The evaluation result will generate corresponding report data, providing important reference for engineering decision. And through an urban intelligent risk early warning module, weather prediction data, geological exploration report data and flood control drainage evaluation report data are comprehensively considered, and intelligent risk early warning is carried out on the track engineering zone. The system analyzes potential influences of various risk factors according to the early warning model and algorithm, and gives an alarm in time. For example, the system may pre-warn of possible construction accidents, equipment failures, floods, geological disasters, etc., so that the relevant personnel and residents take necessary guidance and decision support. When the risk level data reaches a high risk level, the system will utilize the urban refuge planning module to plan the refuge site for the track engineering zone. The system can analyze factors such as surrounding topography, building structures and the like, determine safe refuge site distribution, and send relevant information to relevant personnel through the cloud monitoring platform so as to guide evacuation actions. This ensures a safe shelter for constructors and nearby residents in the event of a risk. In summary, the invention can comprehensively monitor and evaluate the safety risk in the track engineering construction process, discover the potential risk in time, and provide necessary guidance and guarantee for related personnel through early warning and decision support, thereby improving the safety and efficiency of the track engineering construction and reducing accidents and losses.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of a non-limiting implementation, made with reference to the accompanying drawings in which:

Fig. 1 is a schematic flow chart of steps of a track traffic engineering construction security risk monitoring method according to an embodiment.

Fig. 2 shows a detailed step flow diagram of step S3 of an embodiment.

Fig. 3 shows a detailed step flow diagram of step S36 of an embodiment.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

To achieve the above objective, referring to fig. 1 to 3, the present invention provides a method for monitoring the safety risk of track traffic engineering construction, which comprises the following steps:

Specifically, for example, unmanned aerial vehicle or other aircraft capable of carrying imaging equipment can be used for aerial photography, and ground imaging equipment can also be used for image acquisition. Next, regional pattern recognition is performed on the track engineering zone image data, and topographical features may be identified and extracted using computer vision techniques and machine learning algorithms. The topographical feature data may include relief, river and lake, vegetation cover, and the like, for describing the geographic environment of the track engineering zone.

specifically, future weather predictions for the track engineering zone may be made from the topographical feature data using, for example, meteorological data and a numerical weather prediction model. This may include predicting changes in meteorological parameters such as rainfall, wind speed, temperature, etc. over the next days or weeks. And performing geological exploration, including geological exploration survey, geological sample acquisition and the like, so as to acquire geological features and geological structure information of the track engineering zone. Such data may include information of soil type, groundwater level, geologic horizons, etc. And evaluating the flood control and drainage capacity of the track engineering zone by combining weather prediction data and geological exploration report data. This may include simulating the evolution of floods in different rainfall scenarios, analyzing the impact of factors such as topography, water flow paths, drainage systems, etc. on floods, and evaluating the flood and drainage capacity of the track engineering zone. And generating flood control drainage evaluation report data according to the result of the flood control drainage capability evaluation. The report may include information of flood risk level, potential impact of flood disasters, suggested flood control and drainage measures, etc. for the track engineering zone to support subsequent engineering planning and decisions.

Specifically, an intelligent risk early warning model is established by using weather prediction data, geological exploration report data and flood control drainage evaluation report data. The model can monitor the risk condition of the track engineering zone in real time according to real-time weather data and geological information and by combining a preset risk index and a model algorithm. And generating risk early warning report data according to the output result of the intelligent risk early warning model. The report may include information of the current risk level, risk events that may occur, suggested countermeasures, etc.

Specifically, for example, the risk level data is determined to be high risk level data in the track engineering zone. And carrying out refuge planning on the track engineering zone by utilizing technologies such as a Geographic Information System (GIS), population density analysis and the like according to a preset auxiliary decision-making module. Taking into consideration factors such as the accommodation capacity, safety, convenience and the like of the refuge site, the refuge site distribution of each area is determined. And integrating the risk early warning report data and refuge site distribution data through a preset cloud monitoring platform, and sending the integrated risk early warning report data and the refuge site distribution data to constructors and nearby residents. The early warning information and the refuge distribution map can be sent by means of mobile phone application, short message notification or email and the like.

Preferably, step S2 comprises the steps of:

step S21: acquiring historical meteorological data of a track engineering zone;

specifically, the stored historical weather data may be obtained, for example, by contacting a local weather department or weather site. Or purchase or acquire historical weather data for the track engineering zone using a weather data subscription service or a weather data provider's platform. Historical weather data may be solicited from relevant institutions or researchers if other relevant engineering projects have similar weather data collection in the area.

Step S22: carrying out model training and verification on a preset special weather prediction model by using historical meteorological data so as to obtain the special weather prediction model;

Specifically, the historical meteorological data may be divided into training and verification sets using, for example, machine learning or deep learning algorithms. Based on a preset weather prediction model architecture, the model is trained by using historical meteorological data, and model parameters are adjusted to improve prediction accuracy. And evaluating and verifying the trained model by using a verification set, and determining the performance and reliability of the model by comparing the prediction result of the model with actual observation data.

Step S23: the method comprises the steps of observing meteorological stations and detecting troposphere of an meteorological unmanned aerial vehicle in a track engineering zone, so that real-time meteorological data are obtained;

Specifically, for example, a weather site may be provided in a track engineering zone, and weather-observing devices such as a thermometer, hygrometer, anemometer, etc. may be installed to periodically record and collect weather-observing data. And deploying a meteorological unmanned aerial vehicle or an aircraft, and carrying out unmanned aerial vehicle troposphere detection on the atmospheric parameters of the track engineering zone by carrying a meteorological sensor to acquire more detailed meteorological data. Integrating and analyzing meteorological site observation data and meteorological unmanned aerial vehicle detection data to obtain real-time meteorological conditions of the track engineering zone.

Step S24: inputting the real-time weather data into a special weather prediction model to perform future weather prediction, thereby obtaining weather prediction data;

Specifically, for example, real-time weather data including temperature, humidity, wind speed, precipitation, and other indicators may be input into a pre-trained dedicated weather prediction model. The model calculates and analyzes according to the characteristics of the real-time data and the modes of the historical data, and generates a future weather prediction result. And according to weather prediction data output by the model, such as temperature change, precipitation and the like in the future days, the weather prediction data is provided for a track engineering team as decision references for engineering scheduling, resource allocation, risk management and the like.

Step S25: acquiring geological exploration report data of a track engineering zone;

Specifically, for example, a professional geological exploration company may be commissioned to perform geological exploration operations, including operations such as geological exploration drilling, sampling, and the like. And obtaining a geological exploration report according to the result of geological exploration work, wherein the geological exploration report comprises information such as stratum conditions, groundwater level, soil type, rock strength and the like.

Step S26: the method comprises the steps of carrying out intelligent pipe network system pressure real-time monitoring on a track engineering zone so as to obtain real-time pipe network load data;

In particular, for example, intelligent sensors and monitoring devices can be installed in a pipe network system of a track engineering zone for monitoring the pressure and flow of the pipe network in real time. The data collected by the sensor comprises parameters such as pipeline pressure, flow velocity, liquid level and the like, and the data of pipe network load is obtained in real time.

Step S27: carrying out flood digital simulation and permeability evaluation on the track engineering zone according to the topography and topography characteristic data and the geological exploration report data, thereby obtaining permeability evaluation data;

Specifically, flood digital simulation can be performed based on the information of groundwater level, soil type and the like in the geological exploration report according to the topography and topography characteristic data, and the water flow and waterlogging conditions under different flood situations can be simulated. And evaluating the permeability of the track engineering zone according to permeability data and soil types in the geological exploration report, wherein the permeability comprises indexes such as soil permeability coefficient, horizontal permeability and the like.

Step S28: according to weather prediction data, pipe network load real-time data and permeability evaluation data, constructing a flood control drainage intelligent evaluation model of the track engineering zone, so as to obtain the flood control intelligent evaluation model; and carrying out intelligent flood control and drainage capacity assessment on the track engineering land by using the intelligent flood control assessment model so as to obtain flood control and drainage assessment report data.

Specifically, for example, an intelligent flood control drainage assessment model can be constructed based on weather prediction data, pipe network load real-time data and permeability assessment data, and a machine learning or mathematical modeling method can be adopted. And carrying out intelligent assessment on flood control and drainage capacity of the track engineering land by using the constructed intelligent flood control assessment model, wherein assessment indexes comprise flood prediction, drainage capacity analysis and the like. And generating flood control drainage evaluation report data according to the output result of the flood control drainage evaluation model, wherein the flood control drainage evaluation report data comprises information such as flood risk level evaluation, suggested drainage measures, emergency plans and the like.

According to the invention, by acquiring the historical meteorological data, the standard of the meteorological data can be established, and the past weather conditions including temperature, precipitation, wind speed and other information can be known. This helps to understand the climate characteristics and trends of the track engineering zone, and thereby predict future weather conditions. By training and verifying a preset special weather prediction model by using historical meteorological data, an accurate weather prediction model can be established. The model can be used for predicting future weather conditions and providing reliable weather prediction data for planning, designing and constructing track engineering. By means of meteorological site observation and meteorological unmanned aerial vehicle troposphere detection, real-time meteorological data comprising temperature, humidity, wind direction, wind speed and other information can be obtained. The method is beneficial to monitoring and knowing the current weather conditions of the track engineering zone in real time, and provides accurate data support for engineering construction and operation. By inputting real-time weather data into a dedicated weather prediction model, future weather predictions can be made. The method is helpful for knowing weather conditions including precipitation conditions, air temperature changes and the like in a future period in advance, and provides a prediction basis for construction and operation arrangement of the track engineering. By acquiring the geological exploration report data, the geological condition of the track engineering zone can be known, including information such as soil type, stratum structure, geological structure and the like. The method is favorable for evaluating the influence of geological conditions on track engineering, provides basis for engineering design and construction, and reduces geological disaster risks. By carrying out intelligent pipe network system pressure real-time monitoring, the load real-time data of the track engineering land area pipe network can be obtained. The method is helpful for knowing the running state and the load condition of the pipe network system, finding potential faults or problems in time and reducing the occurrence of equipment faults and accidents. Flood conditions and groundwater penetration capacity that may occur in the track engineering zone may be predicted by flood numerical simulation and permeability evaluation based on geological survey report data. This helps to assess the flood risk and drainage capacity of the track engineering zone, providing a basis for engineering planning and design. By establishing an intelligent flood control and drainage assessment model and inputting flood digital simulation and permeability assessment data, flood control and drainage capacity assessment data of the track engineering zone can be obtained. The evaluation data can be used for judging whether the flood control drainage capacity of the engineering meets the requirements or not, and providing scientific basis for decision making. In summary, by acquiring meteorological data, geological exploration report data, real-time pipe network load data and permeability evaluation data of the track engineering zone and establishing a special weather prediction model and a flood control drainage intelligent evaluation model, accurate data support and evaluation report can be provided for planning, design and decision-making of the track engineering zone. This helps to improve flood control and drainage in track construction zones, reduce flood risk, and ensure safe and sustainable development of the construction.

Preferably, step S3 comprises the steps of:

Step S31: collecting nearby building image data of the track engineering zone, so as to obtain building image data;

Specifically, for example, an image of a building near an orbital engineering zone may be captured using an unmanned aerial vehicle or satellite, thereby acquiring building image data. The images may include different angles and resolutions to facilitate detailed observation and analysis of the building. Image processing systems, which may be hardware specifically designed for image analysis or computer software, are used to process and store image data, communicate with a drone or satellite via an interface, and receive data. Typically, the image processing system has storage, real-time monitoring and image processing functions, and the drone or satellite is connected to the image processing system using a suitable communication protocol. This may involve different communication modes of wireless signals, optical fibers, the internet, etc., depending on the type of drone or satellite and the image processing system, determining the frequency of image acquisition, i.e. the time interval of image capture. Depending on the area to be monitored and the requirements of the application. Some areas may require high frequency acquisition while other areas may require only low frequency acquisition, storing the acquired image data in a suitable storage medium, such as a hard disk drive, cloud storage, or database. The image data should be stored in a retrievable and manageable manner for future analysis and reference to obtain building image data.

Step S32: acquiring building year data corresponding to each building in building image data; performing crack detection and structure aging degree evaluation on nearby buildings according to building image data and corresponding building year data, so as to obtain building structure aging data;

Specifically, building year data corresponding to each building in the building image data may be acquired using, for example, building Information Model (BIM) technology. BIM technology is a building design, construction and management method based on a three-dimensional model, and can provide full life cycle information of a building. BIM software is used for creating and updating a three-dimensional model of each building, and matching and comparing the three-dimensional model with image data, so that the position, shape and characteristics of each building are identified, and corresponding building year data are acquired. BIM databases, which may be software specifically designed for BIM technology or general purpose database software, are used to store and manage BIM models and year of construction data, and communicate and exchange data with BIM software through interfaces. Typically, the BIM database has storage, query and data processing functions, and BIM models and year of construction data are stored in the database using appropriate data formats. This may involve XML, JSON, SQL or the like of different data formats, depending on the type of BIM software and database software, determining the frequency of data updates, i.e. the time interval for BIM model and year of construction data updates. Depending on the building and application requirements that need to be monitored. Some buildings may need to be updated at a high frequency, while other buildings may need to be updated at a low frequency, the acquired building year data and the image data are associated, and classified and ordered according to a certain rule, so as to obtain building year data corresponding to each building.

Step S33: carrying out virtual simulation on a tramcar subgrade construction stress field in a track engineering zone so as to obtain subgrade stress analysis data;

Specifically, the tramway subgrade construction stress field virtual simulation may be performed on the track engineering zone using, for example, finite Element Analysis (FEA) techniques to obtain subgrade stress analysis data. The FEA technology is an engineering analysis method based on a numerical method, and can provide information such as deformation, stress, strain and the like of a structure under the condition of stress. The FEA software is used for establishing and simulating a three-dimensional model of the tramway subgrade, and proper boundary conditions, load conditions and material properties are given according to a construction scheme and engineering conditions, so that the stress field virtual simulation is performed. Simulation results are executed and output using FEA computing systems, which may be hardware specifically designed for FEA technology, or computer software, which communicates with the FEA software through an interface and receives data. Typically, the FEA computing system has computing, real-time monitoring and result output functions, and the FEA model is discretized and solved using appropriate computing methods to obtain stress field distribution and related parameters. This may involve different calculation methods such as finite difference method, finite volume method, finite element method, etc., depending on the type of FEA software and the calculation system, the frequency of the simulation, i.e., the time interval for which the simulation is executed is determined. Depending on the roadbed to be analyzed and the requirements of the application. Some foundations may require high frequency simulation, while others may require only low frequency simulation, with the output simulation results stored in a suitable storage medium, such as a hard disk drive, cloud storage, or database. The simulation results should be stored in a retrievable and manageable manner for future analysis and reference to obtain subgrade stress analysis data.

Step S34: performing rock-soil body internal microcrack expansion risk assessment on the track engineering zone according to the geological exploration report data so as to obtain geological disaster risk assessment data;

Specifically, for example, a GPR apparatus may be used to transmit and receive electromagnetic waves, and appropriate frequencies and depths may be selected based on geological survey report data to perform a risk assessment of microcrack propagation within the rock-soil body. GPR data is processed and displayed using GPR analysis systems, which may be hardware specifically designed for GPR technology, or computer software, that communicate with and receive data from GPR devices via interfaces. Generally, the GPR analysis system has functions of storage, real-time monitoring and data processing, and uses a proper signal processing method to filter, enhance and interpret GPR data, so as to obtain parameters such as the position, the size, the morphology and the like of micro-cracks in a rock-soil body. This may involve different signal processing methods, such as fourier transforms, wavelet transforms, neural networks, etc., depending on the type of GPR apparatus and analysis system, determining the frequency of the detection, i.e. the time interval during which the detection is performed. Depending on the rock-soil body to be evaluated and the requirements of the application. Some rock-soil bodies may need high-frequency detection, while other rock-soil bodies may need low-frequency detection, and the processed GPR data and geological exploration report data are compared and analyzed, and risk assessment is performed according to a certain standard, so that geological disaster risk assessment data are obtained.

Step S35: performing collapse risk degree evaluation on nearby buildings according to roadbed stress analysis data, building structure aging evaluation data and geological disaster risk evaluation data, so as to obtain building collapse risk evaluation data;

Specifically, for example, MCDM software may be used to build and solve a collapse risk level assessment problem, and determine appropriate objective functions, scheme sets, and index weights according to roadbed stress analysis data, building structure aging assessment data, and geological disaster risk assessment data, so as to perform collapse risk level assessment. The MCDM technique is a decision support method based on a mathematical model, and can provide optimal selection among a plurality of targets, a plurality of schemes and a plurality of indexes. Modeling and solving the collapse risk degree assessment problem by using a proper optimization method, so as to obtain parameters such as the collapse risk degree of each building. This may involve different optimization methods such as analytic hierarchy process, fuzzy comprehensive evaluation method, genetic algorithm, etc., depending on the type of MCDM software and computing system, determining the frequency of the evaluation, i.e., the time interval in which the evaluation is performed. Depending on the building and application requirements that need to be evaluated. Some buildings may require high frequency evaluations, while other buildings may require only low frequency evaluations, with the output evaluation results stored in a suitable storage medium, such as a hard disk drive, cloud storage, or database. The assessment results should be stored in a retrievable and manageable manner for future analysis and reference to obtain building collapse risk assessment data.

Step S36: and carrying out intelligent risk early warning on the track engineering zone according to the building collapse risk assessment data, the weather forecast data, the geological exploration report data and the flood control drainage assessment report data, so as to obtain risk early warning report data.

Specifically, for example, an intelligent early warning system (IWS) technology may be used to perform intelligent risk early warning on the track engineering zone according to building collapse risk assessment data, weather prediction data, geological exploration report data, and flood control drainage assessment report data, so as to obtain risk early warning report data. The IWS technology is an artificial intelligence based early warning method, and can provide comprehensive analysis and dynamic prediction among various risk factors. An intelligent risk early warning model is established and operated by using IWS software, and proper input variables, output variables and model parameters are determined according to building collapse risk assessment data, weather prediction data, geological exploration report data and flood control drainage assessment report data, so that intelligent risk early warning is performed.

The invention can acquire the building appearance information of the track engineering zone by collecting the image data of the nearby buildings. This helps to understand the structural features, morphology and status of the building, providing the underlying data for subsequent structural assessment and risk analysis. The structural aging data of the nearby building can be obtained by acquiring the building year data in the building image data and carrying out crack detection and structural aging degree evaluation by combining the image data. This helps to understand the structural health of the building, discover existing cracking and aging problems in advance, and evaluate the stability and safety of the building. By performing virtual simulation of the tramway subgrade construction stress field, stress analysis data of the subgrade can be obtained. The method is favorable for knowing the bearing capacity and stability of the roadbed of the track engineering zone, evaluating the structural safety of the roadbed and providing basis for engineering design and construction. And (3) evaluating the risk of the expansion of the microcracks in the rock-soil body of the track engineering zone according to the geological exploration report data, so that the risk condition of the geological disaster can be evaluated. This helps to understand potential threats to geological disasters, such as ground subsidence, landslide, fractures, etc., and provides a reference for engineering planning and risk management. By combining roadbed stress analysis data, building structure aging evaluation data and geological disaster risk evaluation data, the collapse risk degree of the nearby building can be evaluated. This helps determine the structural stability and load bearing capacity of the building, evaluates the risk of collapse of the building, and provides targeted risk management measures. By performing intelligent risk early warning according to building collapse risk assessment data, weather prediction data, geological exploration report data and flood control drainage assessment report data, potential risk conditions can be timely identified and a risk early warning report can be generated. This helps take necessary measures in advance to mitigate the risk of possible disasters, ensuring the safety of the track engineering zone. The early warning report data can provide detailed risk assessment results and suggestions so that decision makers and related personnel can take appropriate actions to ensure the smooth progress of engineering and the safety of surrounding residents.

Preferably, step S36 comprises the steps of:

step S361: dividing the nearby buildings into high, medium and low risk grades according to the building collapse risk assessment data, thereby obtaining hierarchical early warning building data;

Specifically, building collapse risk assessment data may be obtained, for example, through a database, which may include building structure information, geographic location information, building material information, and the like. Then, based on these data, a risk assessment model or expertise is employed to perform risk assessment on each building. The evaluation results can be classified into high, medium, and low risk levels, and these level information is stored in a database for later use. Finally, according to the grading, hierarchical early warning building data are obtained, wherein the data comprise unique identifiers of the buildings and corresponding risk grades.

Step S362: carrying out total asset loss evaluation on Gao Weilou blocks according to the hierarchical early-warning building data so as to obtain asset loss prediction data;

Specifically, for example, gao Weilou pieces of information can be screened from the hierarchical early warning building data, and then data related to the high-risk buildings are collected, including building estimation, insurance information, rent and the like. Using the asset loss assessment model, gao Weilou pieces of total asset loss assessment were performed on these data. The evaluation results include the predicted asset loss amount for each tall building, which may also be stored in a database for later use.

Step S363: acquiring a historical risk factor data set, corresponding historical risk category data, corresponding historical risk level data and corresponding asset loss data;

In particular, a public database, such as a database maintained by a government agency, local building administration, or related research agency, for example, may be queried to obtain a historical risk factor dataset. These databases may contain records of data concerning year of construction, geologic conditions, natural disaster frequency, etc. The historical risk category data and the historical risk level data may be annotated according to the historical records or expert evaluations of the building. Some public databases may also contain asset loss data about a building, such as loss reports or insurance claim records in the event of a disaster. Or query-related research literature, scientific reports, or risk assessment reports may contain data about historical risk factors, risk categories, risk levels, and asset losses. These documents and reports may be issued by government agencies, academic institutions, professional consultation companies or construction industry associations.

Step S364: performing feature engineering on the historical risk factor dataset, the corresponding historical risk category data, the corresponding historical risk level data and the corresponding asset loss data, thereby obtaining a structured risk feature dataset;

In particular, statistical features, such as maxima, minima, averages, standard deviations, etc., may be extracted from the historical risk factor dataset, for example, to describe the distribution of risk factors. In combination with historical risk category data, one can create a single-hot coded or binary coded feature representing the risk category to which each sample belongs. The risk level can be digitized by combining the historical risk level data, and a magnitude relation of the numerical characteristic representing the risk level is created. Statistical features associated with risk factors, such as total loss amount, average loss amount, etc., are extracted from the asset loss data. Geographic Information System (GIS) data, such as land utilization, topographical features, etc., may be used as additional features to increase the richness of the risk feature dataset.

Step S365: model training and generalizing a preset deep neural network structure by using a structured risk characteristic data set so as to obtain an initial risk prediction model;

Specifically, a deep neural network model may be designed for training and generalization to predict risk, for example, using a structured risk feature dataset as input. For example, modeling may be performed using a deep learning model such as a multi-layer perceptron (MLP), convolutional Neural Network (CNN), or Recurrent Neural Network (RNN). Through iterative training and verification processes, model parameters are optimized, so that risks can be accurately predicted.

Step S366: performing multi-task learning on the initial risk prediction model so as to obtain a multi-task risk prediction model;

Step S367: performing model compression on the multitask risk prediction model so as to obtain a lightweight risk prediction model;

in particular, the multitasking risk prediction model may be model compressed, for example, to reduce complexity and storage space requirements of the model. Common model compression methods include pruning (e.g., structural pruning or parametric pruning), quantization (e.g., weight quantization or activation quantization), and model distillation (using smaller models to learn knowledge of large models). By means of model compression, a lightweight risk prediction model can be obtained, and the method is suitable for environments with limited resources or limited computing capacity.

Step S368: and inputting asset loss prediction data, building collapse risk assessment data, weather prediction data, geological exploration report data and flood control drainage assessment report data into a lightweight risk prediction model to perform intelligent risk early warning on the track engineering zone, so as to obtain risk early warning report data.

The invention can divide nearby buildings into high, medium and low risk grades by grading the building collapse risk assessment data, thereby obtaining grading early warning building data. This helps to quickly identify high risk buildings, take risk management measures preferentially, and provide targets for critical monitoring and maintenance. By evaluating asset loss for high-risk buildings in the hierarchical early-warning building data, potential total asset loss can be predicted. The method is beneficial to evaluating economic risks and influence ranges brought by high-risk buildings, provides basis for risk management decision, reasonably distributes resources and adopts corresponding risk control measures. By acquiring historical risk factor data sets, such as historical building collapse statistics data, historical geological exploration report data, historical flood control drainage report data and historical weather data, and corresponding historical risk class data and corresponding asset loss data, a risk prediction model can be established. The method is helpful for knowing the risk event and loss situation occurring in the past, providing reference for risk prediction and early warning, and improving the accuracy and reliability of risk management. By feature engineering the historical risk factor dataset, the historical risk category data, the historical risk level data, and the asset loss data, a structured risk feature dataset may be extracted and constructed. This helps to find key risk features and patterns, providing input data for training and prediction of risk prediction models. By using the structured risk feature data set to model training and generalize the preset deep neural network structure, an initial risk prediction model can be obtained. This helps build a data-driven based model, learn risk patterns and trends from historical data, providing a basis for predictive power and risk assessment. By multitasking the initial risk prediction model, the model can be extended to a model that can predict multiple risk tasks simultaneously. The method is helpful for comprehensively considering different risk factors and risk categories, and improves the comprehensiveness and accuracy of risk prediction. By compressing the model of the multitasking risk prediction model, the complexity and the computing resource requirement of the model can be reduced, and the lightweight risk prediction model can be obtained. The method is beneficial to improving the deployment efficiency and instantaneity of the model, and is suitable for risk early warning requirements in actual application scenes. The intelligent risk early warning of the track engineering zone can be realized by inputting asset loss prediction data, building collapse risk assessment data, weather prediction data, geological exploration report data and flood control drainage assessment report data into a lightweight risk prediction model. This helps to discover potential risks and predict possible losses in time, provides risk early warning report data, and supports decision makers to take corresponding risk management and emergency measures.

Preferably, step S4 comprises the steps of:

step S41: when the risk level data are determined to be high risk level data, road traffic flow detection is carried out on the zones nearby the track; thereby acquiring traffic flow data;

Specifically, traffic flow detection may be performed on roads near the track using, for example, traffic monitoring devices (e.g., cameras, sensors, etc.). Traffic flow data can be obtained by monitoring the number and flow of vehicles on the road in real time. For example, the condition of traffic flow may be evaluated by counting the number of vehicles passing through the road per hour or calculating the average speed of the vehicles on the road.

Step S42: carrying out intelligent analysis on traffic flow data on the traffic jam degree so as to obtain a road jam state data set;

Specifically, for example, the acquired traffic flow data may be intelligently analyzed to evaluate the congestion degree of the road. By analyzing the indexes such as the flow density, the vehicle speed, the road capacity and the like, the congestion state of the road can be determined. For example, a machine learning algorithm or a rules engine may be used to classify traffic flow data to classify roads into various states of clear, creep, or congested, to obtain a road congestion state dataset.

Step S43: acquiring road traffic network data of a zone nearby a track;

Specifically, for example, road traffic network data of a zone near a track may be acquired, including geometric information of roads, connection relationships between roads, road attributes, and the like. Such data may be obtained from a transportation sector, geographic Information System (GIS), or related institutions. The road traffic network data may be represented graphically or in the form of a network for subsequent path planning and decision analysis.

Step S44: carrying out safety evacuation path planning according to the road congestion state data set and the road traffic network data, thereby obtaining a candidate evacuation path data set;

In particular, the safety evacuation path planning may be performed using, for example, a road congestion status data set and road traffic network data. By taking into account road congestion and road attributes, a path planning algorithm (e.g., shortest path algorithm, a-algorithm, etc.) may be used to calculate a plurality of candidate evacuation paths. These paths should take into account factors such as avoiding congested roads, selecting safe roads, and minimizing evacuation time.

Step S45: carrying out intelligent visual monitoring personnel quantity statistics on Gao Weilou blocks and construction sites so as to obtain personnel distribution data;

Specifically, for example, the number of people in Gao Weilou and construction sites can be counted by using intelligent visual monitoring technology such as a monitoring camera or an image recognition algorithm. Personnel distribution data can be obtained by monitoring personnel in and out in real time or performing personnel detection and counting by using a computer vision technology.

Step S46: carrying out refuge planning on the track engineering zone by utilizing a preset auxiliary decision module according to the candidate evacuation path data set and the personnel distribution data, thereby acquiring refuge distribution data;

Specifically, the evacuation site planning may be performed using a preset auxiliary decision module, for example, using the candidate evacuation path dataset and the personnel distribution data. The evacuation site is selected to be a suitable location according to factors such as distance on the evacuation path, personnel density, and capacity of the evacuation site. Algorithms or rule engines may be used to make decisions and optimizations to obtain refuge distribution data.

Step S47: screening the optimal evacuation path of the candidate evacuation path data set according to the road traffic network data and the refuge site distribution data, so as to obtain optimal evacuation path data;

Specifically, the candidate evacuation path dataset may be screened, for example, in combination with road traffic network data and refuge site distribution data, to obtain an optimal evacuation path. By taking into account road conditions, refuge capacity, evacuation time, etc., a path planning algorithm or decision model may be used to evaluate and select the optimal evacuation path.

Step S48: and sending the risk early warning report data, the refuge site distribution data and the optimal evacuation path data to constructors and nearby residents through a preset cloud monitoring platform so as to guide the residents and constructors to evacuate.

Specifically, for example, the risk early warning report data, refuge site distribution data, and optimal evacuation path data may be transmitted to constructors and nearby residents using a preset cloud monitoring platform. This may be accomplished through a cell phone application, short message, email, or other communication channel. By providing accurate risk information, evacuation paths, and refuge locations, residents and constructors can be guided for safe evacuation, as well as providing necessary guidance and support.

The invention can acquire real-time traffic flow data by detecting the road traffic flow in the zone near the track. This helps to evaluate road traffic conditions, determine if a traffic congestion condition exists, and provide basic data for subsequent safe evacuation path planning. By intelligently analyzing the traffic flow data, the traffic jam degree of the road can be identified and estimated. This helps to obtain a road congestion status data set, provide a detailed description of road conditions, and provide accurate traffic information for subsequent safe evacuation path planning and decision-making. By acquiring road traffic network data for the zone near the track, a road network topology may be established. This helps to analyze the connection between roads, provide basic data for the safety evacuation path planning, and ensure the effectiveness and connectivity of the evacuation path. By combining the road congestion status data set with the road traffic network data, a safe evacuation path planning can be performed. This helps to determine candidate evacuation paths, providing alternatives for subsequent optimal evacuation path screening, taking into account traffic conditions and road connectivity. The intelligent visual monitoring technology is used for counting the personnel number of the high-risk building and the construction site, and real-time personnel distribution data can be obtained. This helps to understand the personnel intensive areas and distribution, providing a reference basis for personnel distribution for refuge planning and evacuation path selection. By utilizing the candidate evacuation path data set and the personnel distribution data in combination with a preset auxiliary decision-making module, refuge planning can be performed. This helps to determine the appropriate refuge site distribution, taking into account evacuation paths, personnel distribution and safety factors, providing a site for safe refuge for residents and constructors. By combining road traffic network data and refuge site distribution data, the candidate evacuation path data set can be screened, and an optimal evacuation path can be selected. This helps to ensure that the best evacuation path is selected, taking into account road conditions, refuge distribution and traffic connectivity, improving evacuation efficiency and safety. And the cloud monitoring platform is used for sending the risk early warning report data, the refuge site distribution data and the optimal evacuation path data to constructors and nearby residents, so that real-time guidance and information can be provided. This helps guide residents and constructors to take correct evacuation actions, enhances emergency response capability, improves safety awareness and action efficiency, and reduces potential risks and injuries.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The track traffic engineering construction safety risk monitoring method is characterized by comprising the following steps of:

Step S3: carrying out intelligent risk early warning on the track engineering zone according to weather prediction data, geological exploration report data and flood control drainage evaluation report data so as to obtain risk early warning report data; wherein, step S3 includes:

step S36: performing intelligent risk early warning on the track engineering zone according to the building collapse risk assessment data, weather forecast data, geological exploration report data and flood control drainage assessment report data, so as to obtain risk early warning report data; wherein step S36 includes:

Step S364: performing feature processing on the historical risk factor dataset, the corresponding historical risk category data, the corresponding historical risk level data and the corresponding asset loss data, thereby obtaining a structured risk feature dataset;

Step S368: inputting asset loss prediction data, building collapse risk assessment data, weather prediction data, geological exploration report data and flood control drainage assessment report data into a lightweight risk prediction model to perform intelligent risk early warning on a track engineering zone, so as to obtain risk early warning report data;

Step S4: when the risk level data are determined to be high risk level data, carrying out refuge planning on the track engineering zone by using a preset auxiliary decision module, so as to obtain refuge distribution data; and sending the risk early warning report data and the refuge site distribution data to constructors and nearby residents through a preset cloud monitoring platform so as to guide the residents and constructors to evacuate.

2. The method for monitoring the risk of track traffic engineering construction safety according to claim 1, wherein the step S2 comprises the steps of:

step S21: acquiring historical meteorological data of a track engineering zone;

3. The method for monitoring the risk of construction safety of rail transit engineering according to claim 1, wherein the step S4 comprises the steps of:

Step S43: acquiring road traffic network data of a zone nearby a track;

4. A track traffic engineering construction safety risk monitoring system for performing the track traffic engineering construction safety risk monitoring method according to claim 1, the track traffic engineering construction safety risk monitoring system comprising:

The urban refuge planning module is used for planning refuge places of the track engineering zones by utilizing the preset auxiliary decision-making module when the risk level data are determined to be high risk level data, so as to acquire refuge place distribution data; and sending the risk early warning report data and the refuge site distribution data to constructors and nearby residents through a preset cloud monitoring platform so as to guide the residents and constructors to evacuate.