CN118313812A - Electric power big data acquisition and processing method based on machine learning - Google Patents
Electric power big data acquisition and processing method based on machine learning Download PDFInfo
- Publication number
- CN118313812A CN118313812A CN202410419887.1A CN202410419887A CN118313812A CN 118313812 A CN118313812 A CN 118313812A CN 202410419887 A CN202410419887 A CN 202410419887A CN 118313812 A CN118313812 A CN 118313812A
- Authority
- CN
- China
- Prior art keywords
- data
- equipment
- power
- fault
- power equipment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 22
- 238000003672 processing method Methods 0.000 title claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 80
- 238000011217 control strategy Methods 0.000 claims abstract description 59
- 238000004088 simulation Methods 0.000 claims abstract description 57
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 230000006854 communication Effects 0.000 claims abstract description 23
- 238000004891 communication Methods 0.000 claims abstract description 23
- 238000011084 recovery Methods 0.000 claims abstract description 21
- 238000007405 data analysis Methods 0.000 claims abstract description 19
- 238000012544 monitoring process Methods 0.000 claims abstract description 17
- 238000002955 isolation Methods 0.000 claims abstract description 12
- 238000007726 management method Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 38
- 238000013527 convolutional neural network Methods 0.000 claims description 22
- 230000000694 effects Effects 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 15
- 230000008901 benefit Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 13
- 239000012634 fragment Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000007689 inspection Methods 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 9
- 230000006378 damage Effects 0.000 claims description 8
- 238000003745 diagnosis Methods 0.000 claims description 8
- 230000015654 memory Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000013144 data compression Methods 0.000 claims description 5
- 238000013501 data transformation Methods 0.000 claims description 5
- 238000009826 distribution Methods 0.000 claims description 5
- 238000005065 mining Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 230000007175 bidirectional communication Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000012800 visualization Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 26
- 238000004364 calculation method Methods 0.000 description 8
- 238000005457 optimization Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000005856 abnormality Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 3
- 238000013079 data visualisation Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000007787 long-term memory Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007728 cost analysis Methods 0.000 description 1
- 238000011550 data transformation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010921 in-depth analysis Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013433 optimization analysis Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Biophysics (AREA)
- Marketing (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application provides a machine learning-based power big data acquisition and processing method, and belongs to the technical field of power engineering. The machine learning-based electric power big data acquisition and processing method comprises the following steps: step 1, operation data of an electric power system are collected in real time and transmitted to a data center through a communication module; step 2, preprocessing the collected data in a data center and storing the preprocessed data in a layering and unified manner; step 3, modeling, predicting and analyzing the state of the power equipment according to different data types; step 4, monitoring abnormal events possibly causing prediction deviation according to the data analysis result; and step 5, when an abnormal event is monitored, starting the power grid simulation platform, selecting an optimal control strategy, and issuing a control instruction to an actual power grid to realize fault isolation and power grid self-recovery. The intelligent monitoring, analysis and management of the power system are realized, and the operation efficiency and the safety of the power grid are improved.
Description
Technical Field
The application relates to the technical field of power engineering, in particular to a machine learning-based power big data acquisition and processing method.
Background
With the continuous development of power systems and the wide application of power equipment, the power industry is faced with increasingly complex management and operation challenges. The traditional power equipment monitoring and management method is often dependent on manual inspection and maintenance in a fixed period, has the problems of untimely data acquisition, low information processing efficiency and the like, and is difficult to meet the requirements of a modern power system on efficient and intelligent monitoring and management.
With the development of big data technology, the power industry is actively exploring how to fully utilize massive power data to improve the operation efficiency and safety of the system. Big data technology is applied in the field of electric power, including data acquisition, storage, processing, analysis, etc., however, how to extract valuable information from huge amounts of data is still a problem to be solved.
In view of the above, the application provides a machine learning-based power big data acquisition and processing method, which aims to solve a plurality of challenges in power system management, improve the efficiency, accuracy and intelligent level of power equipment monitoring and promote the development of the power industry towards more sustainable and intelligent directions.
Disclosure of Invention
In order to overcome a series of defects existing in the prior art, an object of the present invention is to provide a machine learning-based power big data acquisition and processing method, which comprises the following steps:
step 1, operation data of an electric power system are collected in real time and transmitted to a data center through a communication module;
Step 2, preprocessing the collected data in a data center and storing the preprocessed data in a layering and unified manner;
Step 3, modeling, predicting and analyzing the state of the power equipment according to different data types;
step 4, monitoring abnormal events possibly causing prediction deviation according to the data analysis result;
and 5, when an abnormal event is monitored, starting a power grid simulation platform, selecting an optimal control strategy by evaluating the effects of different control strategies, and issuing a control instruction to an actual power grid to realize fault isolation and power grid self-recovery.
Further, step 1 includes the steps of:
Step 1.1, installing corresponding sensors at each key node of the power system to monitor voltage, current, power, temperature, humidity, equipment images and fault alarm information;
Step 1.2, selecting a proper sensor array topological structure according to the distribution condition of the power equipment, so that effective data transmission and a cooperative network are formed among the sensors;
Step 1.3, connecting the sensor array and the communication module to form an intelligent node, configuring a communication protocol and an address, and realizing bidirectional communication with a data center so as to transmit monitoring data;
Step 1.4, setting the frequency and range of data acquisition, and selecting a data compression and encryption algorithm to reduce the delay of data transmission and ensure the safety of data;
Step 1.5, sending the acquired data to a data center in real time through a communication module;
Step 1.6, the data center receives and processes the data from the sensor array, and performs data cleaning, analysis, mining and visualization.
Further, step 2 includes the steps of:
Step 2.1, in a data center, data cleaning is carried out on the collected data, irrelevant data, repeated data and noise data are deleted, and missing values and abnormal values are processed;
Step 2.2, carrying out data transformation on the cleaned data;
Step 2.3, using heterogeneous databases to store data in a layered and unified manner, namely dividing the data into a plurality of layers according to different types, loads and applications, and using different database systems for storage and management in each layer;
and 2.4, establishing a global data mode or view on the basis of the heterogeneous database, and realizing transparent access of data.
Further, step 3 includes the steps of:
Step 3.1, extracting and identifying appearance, structure and damage characteristics of the power equipment by using a convolutional neural network model for the image data of the power equipment so as to realize intelligent inspection and fault diagnosis of the equipment;
Step 3.2, for text data of the electric equipment, performing semantic coding and entity information identification on an overhaul report text and a fault record text of the equipment by using a pre-trained PowerBERT model so as to extract key information and state labels of the equipment;
And 3.3, modeling and predicting the operation parameters of the equipment by using a long-short-term memory network for the time sequence data of the power equipment so as to evaluate the performance and the service life of the equipment.
Further, in step 3.1, the construction of the convolutional neural network model includes the following steps:
Collecting normal images and abnormal images of the power equipment and preprocessing the images;
performing feature extraction on the image data by using a convolutional neural network;
According to different types and fault conditions of the power equipment, a convolutional neural network architecture is designed, and proper super parameters are selected;
Dividing image data into a training set, a verification set and a test set, training and optimizing a convolutional neural network model by using the training set and the verification set, and evaluating the convolutional neural network model by using the test set;
And classifying the new power equipment image by using the trained convolutional neural network model, judging the state and fault type of the equipment according to the output of the model, realizing intelligent inspection and fault diagnosis of the equipment, and visualizing the result of the model and the original image.
Further, step 3.2 comprises the steps of:
extracting relevant text fragments from an overhaul report text and a fault record text of the power equipment, wherein the text fragments comprise names, models, numbers, positions, overhaul time, overhaul personnel, fault types, fault reasons and fault treatment of the equipment;
Using a pre-trained PowerBERT model to carry out semantic coding on the extracted text fragment, namely converting the text fragment into a high-dimensional vector representation so as to capture semantic information and context relation of the text;
Using a pre-trained PowerBERT model to identify entity information of the extracted text segment, namely marking keywords or phrases in the text segment as corresponding entity categories including equipment names, equipment numbers and fault types;
And extracting key information and state labels of the equipment, including the running state, fault frequency, fault severity and fault influence range of the equipment, according to the semantic coding and entity information identification results, and storing the information and labels in a database so as to facilitate subsequent data analysis and application.
Further, step 3.3 includes the steps of:
Extracting relevant time sequence data from the operation data of the power equipment to form a time sequence;
modeling the time sequence data by using a long-short-term memory network, namely taking the time sequence data as the input of an LSTM network, and constructing the structure and parameters of the LSTM network by using a plurality of LSTM units and full connection layers;
Predicting the time sequence data by using an LSTM network, namely taking the historical time sequence data as the input of the LSTM network to obtain the output of the future time sequence data, and corresponding prediction errors and confidence intervals;
And evaluating the performance and the service life of the power equipment according to the prediction result of the LSTM, namely judging the running state and the fault risk of the power equipment and the residual service life and the replacement period of the equipment according to the change trend and the fluctuation range of the time sequence data so as to facilitate manual verification and analysis.
Further, step 4 includes the steps of:
step 4.1, identifying abnormal events which possibly affect the state prediction of the power equipment according to the result of the data analysis
Step 4.2, for each abnormal event, analyzing the reasons, the range of influence, the duration and the degree of harm of the occurrence of the abnormal event, and evaluating the deviation degree and the direction of the abnormal event to the state prediction of the power equipment;
Step 4.3, according to the emergency degree and importance of the abnormal event, making a corresponding emergency treatment scheme;
Step 4.4, implementing an emergency treatment scheme, monitoring the state change of the power equipment, recording the emergency treatment process and result, and evaluating the emergency treatment effect and efficiency;
And 4.5, after the abnormal event is ended, recovering the normal operation of the power equipment, modeling, analyzing and predicting the state of the power equipment again, and calibrating the parameters and errors of the prediction model.
Further, step5 includes the steps of:
Step 5.1, starting a power grid simulation platform, and inputting related information of an abnormal event, wherein the related information comprises a fault type, a fault position and fault time;
step 5.2, operating a simulation platform to obtain the influence of the abnormal event on the power grid;
step 5.3, selecting a plurality of possible control strategies according to the simulation result;
Step 5.4, for each control strategy, running the simulation platform again, and evaluating the recovery effect of the simulation platform on the power grid;
Step 5.5, comprehensively comparing the advantages and disadvantages of each control strategy, selecting the optimal control strategy, or integrating a plurality of control strategies to form an optimal combination;
and 5.6, sending a control instruction to a control center or control equipment of the actual power grid through the communication module, and executing an optimal control strategy to realize fault isolation and power grid self-recovery.
Compared with the prior art, the application has at least the following technical effects or advantages.
The intelligent monitoring, analysis and management of the power system are realized, and the operation efficiency and the safety of the power grid are improved.
Drawings
Fig. 1 is a schematic flow chart of a machine learning-based power big data acquisition and processing method according to an embodiment of the application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention become more apparent, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of the invention.
All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiments described below, together with the words of orientation, are exemplary and intended to explain the invention and should not be taken as limiting the invention.
As shown in fig. 1, a machine learning-based power big data acquisition processing method includes the following steps:
step 1, operation data of an electric power system are collected in real time and transmitted to a data center through a communication module;
Step 2, preprocessing the collected data in a data center and storing the preprocessed data in a layering and unified manner;
Step 3, modeling, predicting and analyzing the state of the power equipment according to different data types;
step 4, monitoring abnormal events possibly causing prediction deviation according to the data analysis result;
and 5, when an abnormal event is monitored, starting a power grid simulation platform, selecting an optimal control strategy by evaluating the effects of different control strategies, and issuing a control instruction to an actual power grid to realize fault isolation and power grid self-recovery.
The method comprises the steps of 1, mainly utilizing equipment such as an intelligent ammeter, an intelligent sensor and an intelligent gateway to collect data such as voltage, current, power, frequency, temperature, humidity and weather in a power system in real time, and transmitting the data to a data center through a wired or wireless communication module. The data center may be a cloud or local server with high speed data processing capability and large capacity data storage space.
And 2, carrying out pretreatment operations such as quality inspection, cleaning, compression, encryption and the like on the collected data so as to improve the usability and safety of the data and save the storage space. Meanwhile, according to the characteristics of the source, the type, the use and the like of the data, the data is stored in a layered and unified mode, so that the subsequent data retrieval and analysis are facilitated. The hierarchical storage of the data can adopt large data platforms such as Hadoop, spark and the like, and the distributed storage and parallel computation of the data are realized.
And 3, performing data-driven modeling, prediction and analysis on the state of the power equipment by mainly using an artificial intelligence technology. According to the type of the data, different machine learning algorithms can be adopted, for example, for image data, a convolutional neural network can be adopted for image recognition and classification, so that intelligent inspection of the power equipment is realized; for text data, a natural language processing technology can be adopted to carry out text mining and analysis, so that fault diagnosis of the power equipment is realized; for time sequence data, a cyclic neural network or a long-term and short-term memory network can be adopted to conduct time sequence modeling and prediction, and state prediction and optimization of the power equipment are achieved.
Step 4, mainly by evaluating and comparing the result of data analysis, finding abnormal events which may cause prediction deviation, such as failure of power equipment, abrupt change of load, fluctuation of renewable energy sources, and the like. For abnormal events, an abnormal detection algorithm such as an isolated forest, a support vector machine and the like can be adopted to identify and position the abnormal events so as to take corresponding measures in time.
And 5, simulating and evaluating the influence of the abnormal event mainly by using a power grid simulation platform, selecting an optimal control strategy by comparing the effects of different control strategies, and issuing a control instruction to an actual power grid to realize fault isolation and power grid self-recovery. The power grid simulation platform can adopt a digital twin technology to construct a digital model of the power grid, so that real-time simulation and optimization of the power grid are realized. The control strategy can adopt intelligent optimization algorithms such as reinforcement learning and the like to realize self-adaption and dynamic adjustment of the control strategy.
Further, step 1 includes the steps of:
Step 1.1, installing corresponding sensors at each key node of the power system to monitor voltage, current, power, temperature, humidity, equipment images and fault alarm information;
Step 1.2, selecting a proper sensor array topological structure according to the distribution condition of the power equipment, so that effective data transmission and a cooperative network are formed among the sensors;
Step 1.3, connecting the sensor array and the communication module to form an intelligent node, configuring a communication protocol and an address, and realizing bidirectional communication with a data center so as to transmit monitoring data;
Step 1.4, setting the frequency and range of data acquisition, and selecting a data compression and encryption algorithm to reduce the delay of data transmission and ensure the safety of data;
Step 1.5, sending the acquired data to a data center in real time through a communication module;
Step 1.6, the data center receives and processes the data from the sensor array, and performs data cleaning, analysis, mining and visualization.
The purpose of the step 1.1 is to acquire the running state and the equipment condition of the power system in real time so as to discover and process abnormal conditions in time and improve the reliability and the safety of the power system.
The purpose of step 1.2 is to optimize the layout and connection mode of the sensor, improve the efficiency and quality of data transmission, reduce data loss and interference, and enhance the cooperation capability of the sensor. The sensor array topology refers to the spatial geometry of the sensors, and is commonly provided with circumferential annular arrangement, rectangular arrangement, abnormal arrangement and the like.
The purpose of step 1.3 is to enable the sensor array to have a communication function, and to transmit and receive data to and from the data center, and to receive control instructions from the data center. The communication module may be wireless or wired and the communication protocol may be TCP/IP, MQTT, zigBee or the like.
The purpose of step 1.4 is to determine appropriate data acquisition parameters according to different data types and requirements, and compress and encrypt the data to improve the speed and security of data transmission. The data compression is a technology for reducing the storage space and transmission bandwidth of data by removing redundant information in the data, and common data compression algorithms include Huffman coding, LZW algorithm, arithmetic coding and the like. Data encryption is a technology of converting data into ciphertext through an encryption algorithm, so that an unauthorized user cannot read the data, and common data encryption algorithms include AES, DES, RSA and the like.
The purpose of step 1.5 is to transfer the data collected by the sensor array to a data center in time for subsequent data processing and analysis. The data center is a platform integrating functions of data storage, processing, analysis, display and the like, and can carry out multi-dimensional mining and utilization on data.
The purpose of step 1.6 is to effectively manage and utilize the data transmitted by the sensor array to extract valuable information and knowledge, providing support for optimization and decision making of the power system. Data cleansing refers to quality inspection and error correction of data to improve accuracy and consistency of the data. The data analysis refers to statistics and calculation of data to obtain basic characteristics and rules of the data. Data mining refers to deep exploration and discovery of data using techniques such as artificial intelligence, machine learning, etc., to obtain implicit, valuable, predictable information and knowledge. The data visualization means that information and knowledge of data are displayed in a visual and easy-to-understand form by means of graphics, charts, animations and the like.
Further, step 2 includes the steps of:
Step 2.1, in a data center, data cleaning is carried out on the collected data, irrelevant data, repeated data and noise data are deleted, and missing values and abnormal values are processed;
Step 2.2, carrying out data transformation on the cleaned data;
Step 2.3, using heterogeneous databases to store data in a layered and unified manner, namely dividing the data into a plurality of layers according to different types, loads and applications, and using different database systems for storage and management in each layer;
and 2.4, establishing a global data mode or view on the basis of the heterogeneous database, and realizing transparent access of data.
In this embodiment, data cleansing is a process of rechecking and checking data, aiming at deleting duplicate information, correcting errors present, and providing data consistency. The method for cleaning the data comprises the steps of selecting a subset, renaming a column name, deleting repeated values, processing missing values, conforming, sorting the data, processing abnormal values and the like. Data transformation refers to converting or unifying data into a form more suitable for machine training or data analysis. The purpose of data transformation is to scale the data, distribute symmetrically, reduce dimensionality, extract linear features, etc. The data transformation method comprises feature binarization, feature normalization, continuous feature transformation, qualitative feature dummy coding and the like. Data integration refers to the merging of data from different data sources into one consistent data store. The data integration aims at realizing sharing and transparent access of data and eliminating redundancy and inconsistency of the data. Methods of data integration include data schema conversion, data value conversion, data warehouse building, and the like. Data reduction refers to reducing the amount or dimension of data while maintaining the integrity and quality of the data. The purpose of the data protocol is to reduce the complexity of the data and improve the efficiency and accuracy of data analysis. The method of data protocol includes feature selection, feature extraction, data sampling, data clustering, etc.
Further, step 3 includes the steps of:
Step 3.1, extracting and identifying appearance, structure and damage characteristics of the power equipment by using a convolutional neural network model for the image data of the power equipment so as to realize intelligent inspection and fault diagnosis of the equipment;
Step 3.2, for text data of the electric equipment, performing semantic coding and entity information identification on an overhaul report text and a fault record text of the equipment by using a pre-trained PowerBERT model so as to extract key information and state labels of the equipment;
And 3.3, modeling and predicting the operation parameters of the equipment by using a long-short-term memory network for the time sequence data of the power equipment so as to evaluate the performance and the service life of the equipment.
In this embodiment, the purpose of step 3.1 is to automatically detect and identify the surface condition and internal structure of the power device by using the spatial information and local features of the image data, so as to discover the abnormality and damage of the device, and improve the safety and reliability of the device.
The purpose of the step 3.2 is to utilize semantic information and context relation of text data to deeply understand and analyze overhaul history and fault experience of the electric equipment so as to acquire key information and state labels of the equipment and provide references for fault diagnosis and prevention of the equipment. PowerBERT is a pre-training language model for the electric power field, which can perform efficient semantic coding and entity information recognition on text data related to electric power, and improve the representation capability and usability of the text data.
The purpose of the step 3.3 is to dynamically model and predict the operation parameters of the power equipment by utilizing the time information and the dynamic characteristics of the time sequence data so as to evaluate the performance and the service life of the equipment and provide basis for optimizing and maintaining the equipment. The long-term and short-term memory network is a variant of a cyclic neural network, can be used for long-term memory and learning of time sequence data, and is commonly used for modeling and prediction of time sequence data, such as power load prediction, power price prediction and the like.
Further, in step 3.1, the construction of the convolutional neural network model includes the following steps:
Collecting normal images and abnormal images of the power equipment and preprocessing the images;
performing feature extraction on the image data by using a convolutional neural network;
According to different types and fault conditions of the power equipment, a convolutional neural network architecture is designed, and proper super parameters are selected;
Dividing image data into a training set, a verification set and a test set, training and optimizing a convolutional neural network model by using the training set and the verification set, and evaluating the convolutional neural network model by using the test set;
And classifying the new power equipment image by using the trained convolutional neural network model, judging the state and fault type of the equipment according to the output of the model, realizing intelligent inspection and fault diagnosis of the equipment, and visualizing the result of the model and the original image.
Further, step 3.2 comprises the steps of:
extracting relevant text fragments from an overhaul report text and a fault record text of the power equipment, wherein the text fragments comprise names, models, numbers, positions, overhaul time, overhaul personnel, fault types, fault reasons and fault treatment of the equipment;
Using a pre-trained PowerBERT model to carry out semantic coding on the extracted text fragment, namely converting the text fragment into a high-dimensional vector representation so as to capture semantic information and context relation of the text;
Using a pre-trained PowerBERT model to identify entity information of the extracted text segment, namely marking keywords or phrases in the text segment as corresponding entity categories including equipment names, equipment numbers and fault types;
And extracting key information and state labels of the equipment, including the running state, fault frequency, fault severity and fault influence range of the equipment, according to the semantic coding and entity information identification results, and storing the information and labels in a database so as to facilitate subsequent data analysis and application.
Further, step 3.3 includes the steps of:
Extracting relevant time sequence data from the operation data of the power equipment to form a time sequence;
modeling the time sequence data by using a long-short-term memory network, namely taking the time sequence data as the input of an LSTM network, and constructing the structure and parameters of the LSTM network by using a plurality of LSTM units and full connection layers;
Predicting the time sequence data by using an LSTM network, namely taking the historical time sequence data as the input of the LSTM network to obtain the output of the future time sequence data, and corresponding prediction errors and confidence intervals;
And evaluating the performance and the service life of the power equipment according to the prediction result of the LSTM, namely judging the running state and the fault risk of the power equipment and the residual service life and the replacement period of the equipment according to the change trend and the fluctuation range of the time sequence data so as to facilitate manual verification and analysis.
Further, step 4 includes the steps of:
Step 4.1, identifying an abnormal event which possibly affects the state prediction of the power equipment according to the result of data analysis;
Step 4.2, for each abnormal event, analyzing the reasons, the range of influence, the duration and the degree of harm of the occurrence of the abnormal event, and evaluating the deviation degree and the direction of the abnormal event to the state prediction of the power equipment;
Step 4.3, according to the emergency degree and importance of the abnormal event, making a corresponding emergency treatment scheme;
Step 4.4, implementing an emergency treatment scheme, monitoring the state change of the power equipment, recording the emergency treatment process and result, and evaluating the emergency treatment effect and efficiency;
And 4.5, after the abnormal event is ended, recovering the normal operation of the power equipment, modeling, analyzing and predicting the state of the power equipment again, and calibrating the parameters and errors of the prediction model.
In step 4.1, in order to improve the accuracy and reliability of the state prediction of the power equipment, it is necessary to identify the abnormal event from the result of the data analysis. An abnormal event refers to an event that causes a sudden change in the state of the power equipment or a deviation from a normal range during the operation of the power equipment due to various internal or external factors. For example, malfunctions, repairs, debugging, load changes, environmental disturbances, etc. of the power equipment may be abnormal events. According to the method for identifying the abnormal event, the operations such as abnormality detection, abnormality classification, abnormality positioning and the like can be performed on the state data of the power equipment by utilizing technologies such as statistics, machine learning, expert systems and the like according to the result of data analysis, so that the abnormal event which possibly influences the state prediction of the power equipment is found out.
In step 4.2, in order to effectively handle the abnormal events, an in-depth analysis is required for each abnormal event to understand its effect on the state prediction of the power equipment. The analysis method of the abnormal event can quantitatively or qualitatively evaluate the occurrence reason, the influence range, the duration, the hazard degree and the like of the abnormal event by utilizing the technologies of causal analysis, fault tree analysis, event tree analysis, risk analysis and the like according to the information of the type, the characteristics, the source and the like of the abnormal event, so as to determine the deviation degree and the direction of the abnormal event on the state prediction of the power equipment. For example, if an abnormal event is caused by a fault in the power equipment, it may cause the state of the power equipment to decrease, thereby making the prediction model generate too high a prediction value, resulting in prediction deviation.
In step 4.3, in order to timely restore the normal operation of the power equipment, a corresponding emergency treatment scheme is required to be formulated according to the emergency degree and importance of the abnormal event. The emergency degree and importance of the abnormal event can be comprehensively evaluated according to factors such as the hazard degree, the influence range, the duration and the like of the abnormal event, so that the threat degree of the emergency degree and the importance to the operation safety and stability of the power equipment is determined. The method for formulating the emergency treatment scheme can compare and select different emergency measures by utilizing technologies such as decision analysis, optimization analysis, simulation analysis and the like according to information such as types, characteristics, sources and the like of the abnormal events, so that the optimal emergency treatment scheme is determined. For example, if an abnormal event is caused by a load change of the power equipment, possible emergency treatment schemes are to adjust an operation parameter of the power equipment, switch an operation mode of the power equipment, increase or decrease the number of operations of the power equipment, etc.
In step 4.4, in order to effectively execute the emergency treatment scheme, real-time monitoring needs to be performed on the state change of the power equipment, the process and the result of the emergency treatment are recorded, and the effect and the efficiency of the emergency treatment are evaluated. The monitoring method for the state change of the power equipment can be used for carrying out operations such as real-time display, tracking, early warning and the like on the state of the power equipment according to the state data of the power equipment by utilizing technologies such as data visualization, data mining, data analysis and the like, so that state abnormality of the power equipment can be found and fed back in time. The recording method of the emergency treatment process and result can utilize the technologies of database, log, report and the like to carry out complete storage, management, inquiry and other operations on the emergency treatment process and result according to the information of the emergency treatment scheme, measure, time, cost and the like, thereby facilitating subsequent analysis and evaluation. The method for evaluating the effect and the efficiency of the emergency treatment can quantitatively or qualitatively evaluate the effect and the efficiency of the emergency treatment by utilizing technologies such as evaluation analysis, benefit analysis, cost analysis and the like according to information such as targets, indexes, standards and the like of the emergency treatment, so that the advantages and disadvantages of the emergency treatment and the improvement direction are determined.
In step 4.5, in order to ensure the accuracy and reliability of the state prediction of the power equipment, the normal operation of the power equipment needs to be restored after the abnormal event is ended, the state of the power equipment is modeled, analyzed and predicted again, and the parameters and errors of the prediction model are calibrated. The recovery method for the normal operation of the power equipment can adjust and optimize the operation parameters, the operation mode, the operation number and the like of the power equipment by utilizing the technologies of a control system, a scheduling system, a maintenance system and the like according to the operation requirement of the power equipment, so that the state of the power equipment is recovered to a normal range. The method for re-modeling, analyzing and predicting the state of the power equipment can preprocess the state data of the power equipment by utilizing technologies such as data cleaning, data conversion, data integration and the like according to the state data of the power equipment, so that the quality and the usability of the data are improved; modeling state data of the power equipment by utilizing technologies such as machine learning, deep learning, neural network and the like, so as to construct a proper prediction model; and analyzing and predicting the state data of the power equipment by utilizing technologies such as data analysis, data mining, data visualization and the like, so as to obtain an accurate and reliable prediction result. The method for calibrating the parameters and the errors of the prediction model can utilize the technologies of parameter estimation, error analysis, model optimization and the like to adjust and optimize the parameters and the errors of the prediction model according to the comparison of the prediction result and the actual result, thereby improving the performance and the accuracy of the prediction model.
Further, step5 includes the steps of:
Step 5.1, starting a power grid simulation platform, and inputting related information of an abnormal event, wherein the related information comprises a fault type, a fault position and fault time;
step 5.2, operating a simulation platform to obtain the influence of the abnormal event on the power grid;
step 5.3, selecting a plurality of possible control strategies according to the simulation result;
Step 5.4, for each control strategy, running the simulation platform again, and evaluating the recovery effect of the simulation platform on the power grid;
Step 5.5, comprehensively comparing the advantages and disadvantages of each control strategy, selecting the optimal control strategy, or integrating a plurality of control strategies to form an optimal combination;
and 5.6, sending a control instruction to a control center or control equipment of the actual power grid through the communication module, and executing an optimal control strategy to realize fault isolation and power grid self-recovery.
In step 5.1, in order to simulate the influence of the abnormal event on the power grid, a power grid simulation platform needs to be started, and relevant information of the abnormal event is input. The power grid simulation platform is a computer-based software system and can simulate and analyze the running process of the power grid according to the information of the structure, parameters, running states and the like of the power grid. The related information of the abnormal event includes a fault type, a fault location, and a fault time. The fault type refers to the cause of an abnormal event such as a short circuit, a broken circuit, an overload, an overvoltage, etc. The fault location refers to a grid element or area where an abnormal event occurs, such as a transformer, line, busbar, load, etc. The failure time refers to the time at which the abnormal event occurs. The method for inputting the related information of the abnormal event can input the related information of the abnormal event into the power grid simulation platform in a manual input mode, a file import mode, a database reading mode and the like, so that input data is provided for subsequent simulation operation.
In step 5.2, in order to obtain the influence of the abnormal event on the power grid, a simulation platform needs to be operated to perform simulation calculation. The method for operating the simulation platform can configure and start the power grid simulation platform by utilizing the technologies of simulation control, simulation setting, simulation starting and the like according to the information of functions, interfaces, operations and the like of the power grid simulation platform, so that simulation calculation is started. The simulation calculation method can simulate and analyze the operation process of the power grid by utilizing the technologies of mathematical models, algorithms, programs and the like according to the information of the structure, parameters, operation states and the like of the power grid and the related information of the abnormal events, so that a simulation result is obtained. The simulation result refers to data or graphics reflecting the influence of an abnormal event on the power grid, such as the change condition of indexes of voltage, current, power, frequency, stability and the like, and the display condition of phenomena of topological structure, tide distribution, fault propagation and the like of the power grid. The simulation result can be output or displayed from the power grid simulation platform in the modes of data output, graphic display, report generation and the like, so that a basis is provided for the selection of a subsequent control strategy.
In step 5.3, in order to effectively process the abnormal event, several possible control strategies need to be selected according to the simulation result. The control strategy is to indicate the control measures or schemes adopted for the abnormal event, and aims to restore the normal operation of the power grid or reduce the damage of the abnormal event. The method for selecting the control strategy can evaluate and compare different control measures or schemes by using the technologies of experience knowledge, rule base, optimization algorithm and the like according to the simulation result, so as to determine a plurality of possible control strategies. For example, if an abnormal event is caused by a short circuit of a line, possible control strategies are to shut down the faulty line, adjust the operating mode of the grid, start up the backup power supply, etc.
In step 5.4, in order to evaluate the recovery effect of each control strategy, the simulation platform needs to be operated again for each control strategy to perform simulation calculation. The method for running the simulation platform again can utilize the technologies of simulation control, simulation setting, simulation starting and the like to configure and start the power grid simulation platform according to the content of each control strategy, so that simulation calculation is started. The simulation calculation method can simulate and analyze the operation process of the power grid by utilizing the technologies of a mathematical model, an algorithm, a program and the like according to the information of the structure, the parameters, the operation state and the like of the power grid and the content of each control strategy, so that a simulation result is obtained. The simulation result refers to data or graphics reflecting the recovery effect of each control strategy, such as the change condition of indexes of voltage, current, power, frequency, stability and the like, and the display condition of phenomena such as topological structure, tide distribution, fault elimination and the like of the power grid. The method for evaluating the recovery effect of each control strategy can evaluate and compare the simulation result of each control strategy by means of data analysis, graph comparison, index calculation and the like, so that the advantages, disadvantages and applicability of each control strategy are determined.
In step 5.5, in order to achieve the best fault handling effect, the advantages and disadvantages of the control strategies need to be comprehensively compared, and the optimal control strategy is selected, or a plurality of control strategies are comprehensively combined to form an optimal combination. The method for comprehensively comparing the advantages and disadvantages of the control strategies can be used for comprehensively evaluating according to the recovery effect, implementation difficulty, cost benefit and other factors of each control strategy, so that the advantages and disadvantages and the ranks of the control strategies are determined. The method for selecting the optimal control strategy can select the control strategy with the best recovery effect, lowest implementation difficulty and highest cost benefit as the optimal control strategy according to the comprehensive comparison result. The method for forming the optimal combination by integrating the plurality of control strategies can select a plurality of control strategies with better recovery effect, lower implementation difficulty and higher cost benefit according to the result of the integrated comparison, and the control strategies are combined into an optimal combination according to a certain sequence or priority to serve as the optimal control strategy.
In step 5.6, in order to realize fault isolation and grid self-recovery, a control command is sent to a control center or control equipment of an actual grid through a communication module, and an optimal control strategy is executed. The communication module is a network-based hardware device or a network-based software system, and can realize information exchange between the power grid simulation platform and a control center or control equipment of an actual power grid. The control command refers to a command generated to control the running state of the power grid according to the content of the optimal control strategy, such as a switch state, a voltage value, a frequency value, a protection device setting and the like. The method for sending the control instruction can send the control instruction to a control center or control equipment of an actual power grid from the power grid simulation platform by utilizing technologies such as a network protocol, a data format, an encryption technology and the like through a communication module, so that information transmission and exchange are realized. The method for executing the optimal control strategy can adjust and optimize the running state of the actual power grid by utilizing technologies such as a control system, a dispatching system, a protection system and the like according to the received control instruction, thereby realizing fault isolation and power grid self-recovery. Fault isolation refers to isolating a grid element or region in which an abnormal event occurs from other normal grid elements or regions, thereby preventing the propagation and propagation of faults. The self-recovery of the power grid means that after fault isolation, the self-regulation capacity or standby resources of the power grid are utilized to recover the normal operation of the power grid or to reach a new stable state.
Claims (9)
1. The electric power big data acquisition and processing method based on machine learning is characterized by comprising the following steps of:
step 1, operation data of an electric power system are collected in real time and transmitted to a data center through a communication module;
Step 2, preprocessing the collected data in a data center and storing the preprocessed data in a layering and unified manner;
Step 3, modeling, predicting and analyzing the state of the power equipment according to different data types;
step 4, monitoring abnormal events possibly causing prediction deviation according to the data analysis result;
and 5, when an abnormal event is monitored, starting a power grid simulation platform, selecting an optimal control strategy by evaluating the effects of different control strategies, and issuing a control instruction to an actual power grid to realize fault isolation and power grid self-recovery.
2. The machine learning-based power big data acquisition processing method as claimed in claim 1, wherein the step 1 comprises the steps of:
Step 1.1, installing corresponding sensors at each key node of the power system to monitor voltage, current, power, temperature, humidity, equipment images and fault alarm information;
Step 1.2, selecting a proper sensor array topological structure according to the distribution condition of the power equipment, so that effective data transmission and a cooperative network are formed among the sensors;
Step 1.3, connecting the sensor array and the communication module to form an intelligent node, configuring a communication protocol and an address, and realizing bidirectional communication with a data center so as to transmit monitoring data;
Step 1.4, setting the frequency and range of data acquisition, and selecting a data compression and encryption algorithm to reduce the delay of data transmission and ensure the safety of data;
Step 1.5, sending the acquired data to a data center in real time through a communication module;
Step 1.6, the data center receives and processes the data from the sensor array, and performs data cleaning, analysis, mining and visualization.
3. The machine learning-based power big data acquisition processing method according to claim 1, wherein the step 2 comprises the steps of:
Step 2.1, in a data center, data cleaning is carried out on the collected data, irrelevant data, repeated data and noise data are deleted, and missing values and abnormal values are processed;
Step 2.2, carrying out data transformation on the cleaned data;
Step 2.3, using heterogeneous databases to store data in a layered and unified manner, namely dividing the data into a plurality of layers according to different types, loads and applications, and using different database systems for storage and management in each layer;
and 2.4, establishing a global data mode or view on the basis of the heterogeneous database, and realizing transparent access of data.
4. The machine learning-based power big data acquisition processing method according to claim 1, wherein the step 3 comprises the steps of:
Step 3.1, extracting and identifying appearance, structure and damage characteristics of the power equipment by using a convolutional neural network model for the image data of the power equipment so as to realize intelligent inspection and fault diagnosis of the equipment;
Step 3.2, for text data of the electric equipment, performing semantic coding and entity information identification on an overhaul report text and a fault record text of the equipment by using a pre-trained PowerBERT model so as to extract key information and state labels of the equipment;
And 3.3, modeling and predicting the operation parameters of the equipment by using a long-short-term memory network for the time sequence data of the power equipment so as to evaluate the performance and the service life of the equipment.
5. The machine learning-based power big data acquisition processing method according to claim 4, wherein in step 3.1, the construction of the convolutional neural network model comprises the following steps:
Collecting normal images and abnormal images of the power equipment and preprocessing the images;
performing feature extraction on the image data by using a convolutional neural network;
According to different types and fault conditions of the power equipment, a convolutional neural network architecture is designed, and proper super parameters are selected;
Dividing image data into a training set, a verification set and a test set, training and optimizing a convolutional neural network model by using the training set and the verification set, and evaluating the convolutional neural network model by using the test set;
And classifying the new power equipment image by using the trained convolutional neural network model, judging the state and fault type of the equipment according to the output of the model, realizing intelligent inspection and fault diagnosis of the equipment, and visualizing the result of the model and the original image.
6. The machine learning-based power big data acquisition processing method according to claim 4, wherein the step 3.2 includes the steps of:
extracting relevant text fragments from an overhaul report text and a fault record text of the power equipment, wherein the text fragments comprise names, models, numbers, positions, overhaul time, overhaul personnel, fault types, fault reasons and fault treatment of the equipment;
Using a pre-trained PowerBERT model to carry out semantic coding on the extracted text fragment, namely converting the text fragment into a high-dimensional vector representation so as to capture semantic information and context relation of the text;
Using a pre-trained PowerBERT model to identify entity information of the extracted text segment, namely marking keywords or phrases in the text segment as corresponding entity categories including equipment names, equipment numbers and fault types;
And extracting key information and state labels of the equipment, including the running state, fault frequency, fault severity and fault influence range of the equipment, according to the semantic coding and entity information identification results, and storing the information and labels in a database so as to facilitate subsequent data analysis and application.
7. The machine learning-based power big data acquisition processing method of claim 4, wherein the step 3.3 comprises the steps of:
Extracting relevant time sequence data from the operation data of the power equipment to form a time sequence;
modeling the time sequence data by using a long-short-term memory network, namely taking the time sequence data as the input of an LSTM network, and constructing the structure and parameters of the LSTM network by using a plurality of LSTM units and full connection layers;
Predicting the time sequence data by using an LSTM network, namely taking the historical time sequence data as the input of the LSTM network to obtain the output of the future time sequence data, and corresponding prediction errors and confidence intervals;
And evaluating the performance and the service life of the power equipment according to the prediction result of the LSTM, namely judging the running state and the fault risk of the power equipment and the residual service life and the replacement period of the equipment according to the change trend and the fluctuation range of the time sequence data so as to facilitate manual verification and analysis.
8. The machine learning-based power big data acquisition processing method as claimed in claim 1, wherein the step 4 includes the steps of:
step 4.1, identifying abnormal events which possibly affect the state prediction of the power equipment according to the result of the data analysis
Step 4.2, for each abnormal event, analyzing the reasons, the range of influence, the duration and the degree of harm of the occurrence of the abnormal event, and evaluating the deviation degree and the direction of the abnormal event to the state prediction of the power equipment;
Step 4.3, according to the emergency degree and importance of the abnormal event, making a corresponding emergency treatment scheme;
Step 4.4, implementing an emergency treatment scheme, monitoring the state change of the power equipment, recording the emergency treatment process and result, and evaluating the emergency treatment effect and efficiency;
And 4.5, after the abnormal event is ended, recovering the normal operation of the power equipment, modeling, analyzing and predicting the state of the power equipment again, and calibrating the parameters and errors of the prediction model.
9. The machine learning-based power big data acquisition processing method according to claim 1, wherein the step 5 comprises the steps of:
Step 5.1, starting a power grid simulation platform, and inputting related information of an abnormal event, wherein the related information comprises a fault type, a fault position and fault time;
step 5.2, operating a simulation platform to obtain the influence of the abnormal event on the power grid;
step 5.3, selecting a plurality of possible control strategies according to the simulation result;
Step 5.4, for each control strategy, running the simulation platform again, and evaluating the recovery effect of the simulation platform on the power grid;
Step 5.5, comprehensively comparing the advantages and disadvantages of each control strategy, selecting the optimal control strategy, or integrating a plurality of control strategies to form an optimal combination;
and 5.6, sending a control instruction to a control center or control equipment of the actual power grid through the communication module, and executing an optimal control strategy to realize fault isolation and power grid self-recovery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410419887.1A CN118313812A (en) | 2024-04-09 | 2024-04-09 | Electric power big data acquisition and processing method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410419887.1A CN118313812A (en) | 2024-04-09 | 2024-04-09 | Electric power big data acquisition and processing method based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118313812A true CN118313812A (en) | 2024-07-09 |
Family
ID=91729617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410419887.1A Pending CN118313812A (en) | 2024-04-09 | 2024-04-09 | Electric power big data acquisition and processing method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118313812A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118644233A (en) * | 2024-08-15 | 2024-09-13 | 福建水口发电集团有限公司 | Hydropower station intelligent inspection method and system based on image recognition technology |
| CN118643452A (en) * | 2024-08-15 | 2024-09-13 | 深圳市鱼儿科技有限公司 | Intelligent electric energy regulation and control system based on central control node |
-
2024
- 2024-04-09 CN CN202410419887.1A patent/CN118313812A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118644233A (en) * | 2024-08-15 | 2024-09-13 | 福建水口发电集团有限公司 | Hydropower station intelligent inspection method and system based on image recognition technology |
| CN118643452A (en) * | 2024-08-15 | 2024-09-13 | 深圳市鱼儿科技有限公司 | Intelligent electric energy regulation and control system based on central control node |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107358366B (en) | Distribution transformer fault risk monitoring method and system | |
| CN118313812A (en) | Electric power big data acquisition and processing method based on machine learning | |
| CN117408162B (en) | Power grid fault control method based on digital twin | |
| CN111191872B (en) | Multi-dimensional toughness evaluation system and method for power grid | |
| CN116308304B (en) | New energy intelligent operation and maintenance method and system based on meta learning concept drift detection | |
| CN117674249A (en) | Distributed photovoltaic-containing power distribution network fault self-healing control and evaluation method | |
| CN117172753A (en) | Intelligent hydropower station inspection system and method based on mobile internet | |
| CN118035814A (en) | State fault early warning method and system based on digital twin field data analysis equipment | |
| KR20220089853A (en) | Method for Failure prediction and prognostics and health management of renewable energy generation facilities using machine learning technology | |
| Zhang et al. | Distributed Architecture of Power Grid Asset Management and Future Research Directions | |
| CN118445731A (en) | Transformer fault determination method and device and transformer fault detection system | |
| CN117611135A (en) | Operation and maintenance control method and system for visual equipment | |
| CN117077532A (en) | Multi-model fusion method for life prediction of wind turbine generator | |
| CN111539642B (en) | Object-oriented power plant data acquisition and processing system and method thereof | |
| CN113076217A (en) | Disk fault prediction method based on domestic platform | |
| Su et al. | Research on digital twin architecture of rule-based intelligent diagnosis of reactor system | |
| CN117892252B (en) | Intelligent park operation management platform based on big data | |
| Wang et al. | Data Cleaning Algorithms for Power Information Communication Assets Data Based on Self-coding | |
| Sun et al. | Component Anomaly Warning and Diagnosis based on Knowledge Graph | |
| CN118863326A (en) | Operation data processing system, method, medium, device, and program for power plant | |
| CN116822788A (en) | Method for analyzing overall operation efficiency of dispatching command business | |
| Jiang et al. | A Hierarchical Maintenance Method for Power System Transmission and Substation Multi-Equipment Operation Status Considering Equipment Service Age | |
| CN118625028A (en) | Automatic checking method and device for running state of power grid equipment start | |
| Lian | Research on health analysis and prediction algorithm of coal mine underground system | |
| CN118760564A (en) | Power grid fault prediction method and system based on machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |