Electrical equipment fault analysis system based on data multi-source fusion

Abstract

The invention discloses an electrical equipment fault analysis system based on data multi-source fusion. The electrical equipment fault analysis system comprises an intelligent operation and maintenance module, an intelligent acquisition module, an intelligent health assessment module and an intelligent fault processing module, the intelligent operation and maintenance module is used for obtaining a distributed strategy according to distributed edge calculation and obtaining equipment self-inspection information according to historical fault information and pre-fault information; the intelligent acquisition module is used for acquiring equipment operation data in real time according to a distributed strategy, and acquiring environment data and weather data; the intelligent health assessment module is used for acquiring the equipment health level of the power primary equipment and predicting the equipment health trend by using a machine learning model; the intelligent fault processing module is used for establishing a fault diagnosis model and obtaining a fault reason; the machine learning technology is used for early warning faults, and maintenance cost is reduced; by adopting distributed edge calculation, the data transmission delay is reduced, and the burden of the server is relieved; and a more accurate decision is made for fault early warning through multi-source data.

Description

Technical Field

[0001] The present invention relates to the technical field of measuring electrical variables, and in particular to an electrical equipment fault analysis system based on multi-source data fusion. Background Art

[0002] In recent years, with the steady advancement of energy internet and new power systems, the requirements for monitoring and fault diagnosis of power equipment have gradually increased. The power system is an important basic system to support the stable development of the national economy. If power equipment has an accident, the economic losses and social impact will be immeasurable; therefore, strengthening the safety monitoring and management of the power monitoring system is of great significance to ensuring economic and social development and people's lives. The power secondary system refers to a series of equipment and components responsible for the protection, control, measurement and communication functions of the power system.

[0003] At present, the Chinese invention patent with application number 201110299750.X provides a method and device for monitoring the operating conditions of primary power equipment. Although it realizes abnormal predictive diagnosis of abnormal signs detected in the early stage of equipment use and aging diagnosis of equipment performance changes year by year, it makes decisions on whether the equipment has failed or needs maintenance through the fuzzy theory of self-learning and self-diagnosis, and analyzes whether the equipment is faulty and diagnoses the type of fault through vibration spectrum monitoring and infrared detection technology combined with an intelligent diagnosis system, effectively improving the operating efficiency of primary equipment and promoting the further development of smart grids. However, it can only perform abnormal predictive diagnosis through abnormal signs, ignoring the influence of other factors on abnormal predictive diagnosis, and does not consider the collection, transmission, storage and analysis of massive data, which is not conducive to the smooth and efficient operation of the central server. Summary of the invention

[0004] The technical problem solved by the present invention is that the existing technology can only perform abnormal predictive diagnosis through abnormal signs, ignores the influence of other factors on abnormal predictive diagnosis, and does not consider the collection, transmission, storage and analysis of massive data, which is not conducive to the smooth and efficient operation of the central server.

[0005] In order to solve the above technical problems, the present invention provides an electrical equipment fault analysis system based on multi-source data fusion, including an intelligent operation and maintenance module, an intelligent acquisition module, an intelligent health assessment module and an intelligent fault processing module; The intelligent operation and maintenance module is used to obtain a distributed strategy according to distributed edge computing, and obtain corresponding historical fault information and corresponding pre-fault information through edge computing nodes according to the distributed strategy, obtain device self-test information according to the historical fault information and the pre-fault information, and each edge computing node issues a self-test command according to the device self-test information; The intelligent collection module is used to collect the equipment operation data of the primary power equipment in real time according to the distributed strategy using the secondary power equipment, and obtain the environmental data and weather data from the relevant platform; The intelligent health assessment module is used for the edge computing node to evaluate the health status of the primary power equipment within a unit time according to the equipment operation data and the primary equipment operation standard, obtain the equipment health level of each primary power equipment, and obtain the historical operation data of each primary power equipment, predict the equipment health trend of each primary power equipment according to the historical operation data, the environmental data and the weather data, and screen out abnormal primary power equipment according to the equipment health level and equipment health trend; The intelligent fault processing module is used for the edge computing node to obtain the abnormal equipment data set, and establish a fault diagnosis model based on the abnormal equipment data set using voiceprint recognition technology and machine learning model, and use the fault diagnosis model to obtain the cause of the fault based on the abnormal equipment data of the abnormal primary power equipment, and perform maintenance according to the cause of the fault; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The intelligent operation and maintenance module includes an intelligent self-checking unit and an intelligent strategy unit; The intelligent self-check unit is used for edge computing nodes to obtain historical fault information and pre-fault information, obtain historical fault indexes of historical fault devices according to the historical fault information, calculate the historical fault indexes of the historical fault devices to obtain a historical fault table, obtain pre-fault indexes of pre-abnormal power primary devices according to the pre-fault information, calculate the fault indexes of the pre-abnormal power primary devices to obtain a pre-fault table, cross-merge the historical fault table and the pre-fault table, perform weighted calculation on historical fault indexes and pre-fault indexes with the same device number to obtain a fault index, and the historical fault indexes and pre-fault indexes corresponding to other different device numbers are also defined as fault indexes; Define the historical fault equipment and pre-abnormal power primary equipment whose fault index is less than or equal to the first expected threshold as low-risk equipment, and obtain low-risk self-test information for the low-risk equipment; Define the historical fault equipment and pre-abnormal power primary equipment with a fault index greater than the first expected threshold and less than or equal to the second expected threshold as medium-risk equipment, and obtain medium-risk self-test information for the medium-risk equipment; The historical fault equipment and the pre-abnormal power primary equipment whose fault index is greater than the second expected threshold are defined as high-risk equipment, and high-risk self-test information is obtained for the high-risk equipment, and various equipment at different risk levels correspond to corresponding self-test information; Each edge computing node issues a self-check instruction according to the device self-check information; The historical fault information includes the equipment number, historical fault cause, historical fault times, historical maintenance operations and historical maintenance results, and the pre-failure information includes the pre-failure equipment number and pre-failure cause; The intelligent strategy unit is used to obtain a distributed strategy by using distributed edge computing, and send the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information to the corresponding edge computing node according to the distributed strategy; the edge computing node issues a self-test instruction to the corresponding primary power equipment according to the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information, obtains a self-test result, and sends the self-test result to the fault processing module for storage; The equipment self-inspection information includes low-risk self-inspection information, medium-risk self-inspection information and high-risk self-inspection information; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The intelligent acquisition module includes a device acquisition unit and a multi-source acquisition unit; The device collection unit is used to collect the device operation data of the primary power device by using the secondary power device, and send the device operation data to the corresponding edge computing nodes respectively according to the distributed strategy; The multi-source acquisition unit is used to connect the environment monitoring platform and the weather monitoring platform using an open interface, and to obtain environmental data from the environment monitoring platform and weather data from the weather monitoring platform.

[0006] As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The intelligent health assessment module includes a health assessment unit and a health prediction unit; The health assessment unit is used for the edge computing node to evaluate each operation data parameter in the device operation data according to the device operation standard, obtain the parameter level of each operation parameter, and obtain the device health level of the primary power device according to the parameter level. If the device health level is greater than or equal to the expected threshold, the corresponding primary power device is defined as a normal primary power device; if the device health level is less than the expected threshold, the corresponding primary power device is defined as an abnormal primary power device; The health prediction unit is used for the edge computing node to retrieve the historical operation data of the primary power equipment, use the historical operation data to train the machine learning model to predict the equipment health trend of the primary power equipment, and obtain the next period of equipment health level according to the equipment health trend, and define the primary power equipment whose next period of equipment health level is less than the expected threshold as the pre-abnormal primary power equipment; The historical operation data includes historical equipment data, historical environment data, historical weather data and historical health level of the primary power equipment; The historical equipment data includes equipment number, historical equipment main parameters, historical equipment working hours, historical equipment temperature, historical vibration amplitude and historical voiceprint information; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The intelligent fault processing module includes a fault sample unit and a fault model unit; The fault sample unit is used to obtain the abnormal operation data set of the primary power equipment from the defect fault simulation platform by using an open interface; The fault model unit is used to obtain voiceprint information and equipment fault information in the abnormal operation data set, extract equipment voiceprint information using a reverberation cancellation method, and use signal enhancement technology and background noise removal methods to achieve background noise reduction, obtain denoised equipment voiceprint information, input the equipment voiceprint information and the fault cause in the equipment fault information into a deep convolutional neural network model for training, use the equipment voiceprint information as input and the fault cause as output to establish a fault diagnosis model, input the abnormal power data of the abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model to obtain the fault cause of the abnormal primary power equipment, input the pre-abnormal power data of the pre-abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model, obtain the pre-fault cause of the pre-abnormal primary power equipment, and issue a fault repair instruction according to the fault cause and the pre-fault cause; The equipment failure information includes equipment number, main equipment parameters and failure cause; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The calculation expression for obtaining the equipment health level of the primary power equipment according to the parameter level is as follows:

[0007] Among them, S is the equipment health level, For each parameter level, is the weight of each parameter level, n is a natural number greater than zero, i is a natural number greater than zero, is the fault index of the primary power equipment, T is the unit time, and P is the usage time of the primary power equipment; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: Using the historical operation data to train a machine learning model to predict the equipment health trend of the primary power equipment includes: The historical environmental data, historical weather data and historical equipment data are used as inputs of the machine learning model, the equipment data of the next period of the historical equipment data is defined as the historical next period equipment data, and the historical next period equipment data is used as the output of the machine learning model to train the machine learning model until the model fit reaches an expected threshold, then the training is terminated, and an equipment health prediction model is obtained; Inputting the real-time collected environmental data, weather data and equipment operation data into the equipment health prediction model, obtaining the next period of primary equipment operation data of the primary power equipment, and defining the next period of primary equipment operation data as the next period of equipment data; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: Acquiring the fault index of the historical fault device according to the historical fault information includes: Obtaining historical fault causes in the historical fault information, obtaining each sub-fault index according to each historical fault cause, performing weighted calculation on each sub-fault index to obtain a historical fault index of the historical fault device, wherein the fault cause and the sub-fault index are in a one-to-one correspondence, and the weight of each sub-fault index in the weighted calculation is obtained according to the sub-fault index; Acquiring a pre-fault index of a pre-abnormal power primary device according to the pre-fault information includes: Acquire a pre-failure cause of the pre-failure information, and acquire a pre-failure index according to the pre-failure cause, wherein the pre-failure cause and the pre-failure index are in a one-to-one correspondence; As a preferred solution of the intelligent system of secondary equipment of power system based on multi-source fusion described in the present invention, wherein: The method of obtaining a distributed strategy by using distributed edge computing includes: Acquire regional distribution information of the primary power equipment and the secondary power equipment, acquire the edge computing node closest to the secondary power equipment according to the regional distribution information, establish a matching mapping table for the secondary power equipment and the corresponding edge computing node, and establish a data receiving platform and a data processing platform for each edge computing node according to the matching mapping table; The distributed strategy includes a pre-collection strategy, a health assessment strategy and a fault handling strategy.

[0008] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an electrical equipment fault analysis system based on multi-source data fusion as described above is implemented.

[0009] Beneficial effects of the present invention: The present invention uses machine learning technology to perform real-time analysis and prediction of the operating data and equipment status of primary power equipment, detect abnormal behavior, and warn of possible failures in advance, thereby realizing intelligent monitoring and predictive maintenance of equipment status, and reducing the failure risk and maintenance cost of primary power equipment; Distributed edge computing is used to delegate terminal data processing tasks to edge computing nodes closer to power equipment for processing, which reduces data transmission delays, reduces the burden on central servers, and increases system scalability and robustness; It not only uses the operating data of primary power equipment and the sensor data of the online monitoring system, but also integrates other relevant data sources, such as environmental parameters and weather data. By integrating multi-source data, it can more comprehensively understand the operating conditions of primary power equipment and make more accurate decisions and optimizations for equipment failure warnings. The probability of possible failure of primary power equipment is evaluated based on historical fault information and pre-failure information, and the fault index is obtained. Self-inspection information is obtained based on the fault index. The primary power equipment completes the self-inspection operation, reduces the failure risk of primary power equipment through pre-maintenance, and ensures the stable operation of primary power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A basic flow chart of an electrical equipment fault analysis system based on multi-source data fusion is provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0012] Example, see Figure 1 , which is an embodiment of the present invention, provides an electrical equipment fault analysis system based on multi-source data fusion, including an intelligent operation and maintenance module, an intelligent acquisition module, an intelligent health assessment module and an intelligent fault processing module; The intelligent operation and maintenance module is used to obtain a distributed strategy according to distributed edge computing, and obtain corresponding historical fault information and corresponding pre-fault information through edge computing nodes according to the distributed strategy, obtain device self-test information according to the historical fault information and the pre-fault information, and each edge computing node issues a self-test command according to the device self-test information; The intelligent collection module is used to collect the equipment operation data of the primary power equipment in real time according to the distributed strategy using the secondary power equipment, and obtain the environmental data and weather data from the relevant platform; The intelligent health assessment module is used for the edge computing node to evaluate the health status of the primary power equipment within a unit time according to the equipment operation data and the primary equipment operation standard, obtain the equipment health level of each primary power equipment, and obtain the historical operation data of each primary power equipment, predict the equipment health trend of each primary power equipment according to the historical operation data, the environmental data and the weather data, and screen out abnormal primary power equipment according to the equipment health level and equipment health trend; The intelligent fault processing module is used for the edge computing node to obtain the abnormal equipment data set, and establish a fault diagnosis model based on the abnormal equipment data set using voiceprint recognition technology and machine learning models, and use the fault diagnosis model to obtain the cause of the fault based on the abnormal equipment data of the abnormal primary power equipment, and perform maintenance based on the cause of the fault.

[0013] Distributed edge computing is a distributed open platform that integrates network, computing, storage, and application core capabilities to provide edge intelligence services nearby at the edge of the network near the source of objects or data. The data collected from the terminal is directly analyzed in the local device or network near the data generation, without the need to transmit the data to the cloud data processing center. The distributed strategy includes the matching of edge computing nodes and power equipment and various operations performed by edge computing nodes on power equipment; The voiceprint recognition technology uses a neural network model to identify voice features from voice signals and classify the voice signals; The machine learning model is a method that can learn from data to automatically discover laws and patterns in the data, and use these laws and patterns to make predictions and decisions.

[0014] In this embodiment, machine learning technology is used to perform real-time analysis and prediction of the operating data and equipment status of primary power equipment, detect abnormal behavior, and warn of possible failures in advance, thereby realizing intelligent monitoring and predictive maintenance of equipment status and reducing the failure risk and maintenance cost of primary power equipment. Distributed edge computing is used to delegate terminal data processing tasks to edge computing nodes closer to power equipment for processing, which reduces data transmission delays, reduces the burden on central servers, and increases system scalability and robustness; It not only uses the operating data of primary power equipment and the sensor data of the online monitoring system, but also integrates other relevant data sources, such as environmental parameters and weather data. By integrating multi-source data, it can more comprehensively understand the operating conditions of primary power equipment and make more accurate decisions and optimizations for equipment failure warnings. The probability of possible failure of primary power equipment is evaluated based on historical fault information and pre-failure information, and the fault index is obtained. Self-inspection information is obtained based on the fault index. The primary power equipment completes the self-inspection operation, reduces the failure risk of primary power equipment through pre-maintenance, and ensures the stable operation of primary power equipment.

[0015] The intelligent operation and maintenance module includes an intelligent self-checking unit and an intelligent strategy unit; The intelligent self-check unit is used for edge computing nodes to obtain historical fault information and pre-fault information, obtain historical fault indexes of historical fault devices according to the historical fault information, calculate the historical fault indexes of the historical fault devices to obtain a historical fault table, obtain pre-fault indexes of pre-abnormal power primary devices according to the pre-fault information, calculate the fault indexes of the pre-abnormal power primary devices to obtain a pre-fault table, cross-merge the historical fault table and the pre-fault table, perform weighted calculation on historical fault indexes and pre-fault indexes with the same device number to obtain a fault index, and the historical fault indexes and pre-fault indexes corresponding to other different device numbers are also defined as fault indexes; Define the historical fault equipment and pre-abnormal power primary equipment whose fault index is less than or equal to the first expected threshold as low-risk equipment, and obtain low-risk self-test information for the low-risk equipment; Define the historical fault equipment and pre-abnormal power primary equipment with a fault index greater than the first expected threshold and less than or equal to the second expected threshold as medium-risk equipment, and obtain medium-risk self-test information for the medium-risk equipment; The historical fault equipment and the pre-abnormal power primary equipment whose fault index is greater than the second expected threshold are defined as high-risk equipment, and high-risk self-test information is obtained for the high-risk equipment, and various equipment at different risk levels correspond to corresponding self-test information; Each edge computing node issues a self-check instruction according to the device self-check information; The historical fault information includes the equipment number, historical fault cause, historical fault times, historical maintenance operations and historical maintenance results, and the pre-failure information includes the pre-failure equipment number and pre-failure cause; The intelligent strategy unit is used to obtain a distributed strategy by using distributed edge computing, and send the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information to the corresponding edge computing node according to the distributed strategy; the edge computing node issues a self-test instruction to the corresponding primary power equipment according to the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information, obtains a self-test result, and sends the self-test result to the fault processing module for storage; The device self-test information includes low-risk self-test information, medium-risk self-test information and high-risk self-test information.

[0016] In this embodiment, a distributed strategy is obtained based on edge distributed computing, and edge computing nodes are used to monitor and analyze data on primary power equipment, which reduces data transmission delays, reduces the burden on the central server, and increases system scalability and robustness. A fault index is obtained based on historical fault information and pre-fault information, and self-test information is obtained based on the fault index. The failure risk of primary power equipment is reduced through pre-maintenance, thereby ensuring the stable operation of primary power equipment.

[0017] The intelligent acquisition module includes a device acquisition unit and a multi-source acquisition unit; The device collection unit is used to collect the device operation data of the primary power device by using the secondary power device, and send the device operation data to the corresponding edge computing nodes respectively according to the distributed strategy; The multi-source acquisition unit is used to connect the environment monitoring platform and the weather monitoring platform using an open interface, and to obtain environmental data from the environment monitoring platform and weather data from the weather monitoring platform.

[0018] In this embodiment, according to the distributed strategy, the equipment operation data of the primary power equipment is collected in real time by the secondary power equipment, which provides accurate and timely data support for the real-time monitoring of the primary power equipment and the fault warning and fault diagnosis of the primary power equipment. At the same time, the multi-source data is obtained from other platforms by using open interfaces, so as to have a more comprehensive understanding of the operation status of the primary power equipment and make more accurate decisions and optimizations.

[0019] The intelligent health assessment module includes a health assessment unit and a health prediction unit; The health assessment unit is used for the edge computing node to evaluate each operation data parameter in the device operation data according to the device operation standard, obtain the parameter level of each operation parameter, and obtain the device health level of the primary power device according to the parameter level. If the device health level is greater than or equal to the expected threshold, the corresponding primary power device is defined as a normal primary power device; if the device health level is less than the expected threshold, the corresponding primary power device is defined as an abnormal primary power device; The health prediction unit is used for the edge computing node to retrieve the historical operation data of the primary power equipment, use the historical operation data to train the machine learning model to predict the equipment health trend of the primary power equipment, and obtain the next period of equipment health level according to the equipment health trend, and define the primary power equipment whose next period of equipment health level is less than the expected threshold as the pre-abnormal primary power equipment; The historical operation data includes historical equipment data, historical environment data, historical weather data and historical health level of the primary power equipment; The historical equipment data includes equipment number, historical equipment main parameters, historical equipment working hours, historical equipment temperature, historical vibration amplitude and historical voiceprint information.

[0020] In this embodiment, the equipment health level of the primary power equipment is evaluated based on the equipment operation data, ensuring real-time understanding of the equipment operation status and real-time discovery of abnormal primary power equipment. A machine learning model is established based on historical operation data to predict the health trend of the primary power equipment, which is conducive to making more accurate decisions and optimizations for equipment failure warnings.

[0021] The intelligent fault processing module includes a fault sample unit and a fault model unit; The fault sample unit is used to obtain the abnormal operation data set of the primary power equipment from the defect fault simulation platform by using an open interface; The fault model unit is used to obtain voiceprint information and equipment fault information in the abnormal operation data set, extract equipment voiceprint information using a reverberation cancellation method, and use signal enhancement technology and background noise removal methods to achieve background noise reduction, obtain denoised equipment voiceprint information, input the equipment voiceprint information and the fault cause in the equipment fault information into a deep convolutional neural network model for training, use the equipment voiceprint information as input and the fault cause as output to establish a fault diagnosis model, input the abnormal power data of the abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model to obtain the fault cause of the abnormal primary power equipment, input the pre-abnormal power data of the pre-abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model, obtain the pre-fault cause of the pre-abnormal primary power equipment, and issue a fault repair instruction according to the fault cause and the pre-fault cause; The device failure information includes the device number, main parameters of the device and the cause of the failure.

[0022] The defect fault simulation platform is a platform that simulates the fault conditions of power secondary equipment and obtains fault data of power secondary equipment; The reverberation cancellation method is a method for extracting device voiceprint information by deconvolving the reverberation signal using an inverse filter; The signal enhancement technology is a technology for processing signals, eliminating noise interference, and improving signal quality; The background noise removal method includes a differential denoising method, a feature extraction method and a pattern recognition method.

[0023] In this embodiment, a fault diagnosis model is established based on the abnormal operation data set, and the abnormal power data and pre-abnormal power data are input into the fault diagnosis model to obtain the fault cause and the pre-fault cause. Maintenance instructions are issued based on the fault cause and the pre-fault cause, saving a lot of manpower and material resources for maintaining the healthy operation of primary power equipment.

[0024] The calculation expression for obtaining the equipment health level of the primary power equipment according to the parameter level is as follows:

[0025] Among them, S is the equipment health level, For each parameter level, is the weight of each parameter level, n is a natural number greater than zero, i is a natural number greater than zero, is the fault index of the primary power equipment, T is the unit time, and P is the usage time of the primary power equipment.

[0026] In this embodiment, the equipment health level of the primary power equipment is obtained according to each parameter level of the equipment operation data, which provides a comprehensive and complete evaluation index for accurately evaluating the equipment health level of the primary power equipment and ensures the accuracy of the equipment health level evaluation.

[0027] Using the historical operation data to train a machine learning model to predict the equipment health trend of the primary power equipment includes: The historical environmental data, historical weather data and historical equipment data are used as inputs of the machine learning model, the equipment data of the next period of the historical equipment data is defined as the historical next period equipment data, and the historical next period equipment data is used as the output of the machine learning model to train the machine learning model until the model fit reaches an expected threshold, then the training is terminated, and an equipment health prediction model is obtained; The environmental data, weather data and equipment operation data collected in real time are input into the equipment health prediction model to obtain the next period of primary equipment operation data of the primary power equipment, and the next period of primary equipment operation data is defined as the next period of equipment data.

[0028] In this embodiment, a machine learning model is established based on historical operating data to predict the next period's equipment data through environmental data, weather data, and equipment operation data, providing a comprehensive data reference for obtaining the next period's equipment health trend, ensuring the source of early warning data for primary power equipment failure warnings, and reducing the failure risk of primary power equipment.

[0029] Acquiring the fault index of the historical fault device according to the historical fault information includes: Obtaining historical fault causes in the historical fault information, obtaining each sub-fault index according to each historical fault cause, performing weighted calculation on each sub-fault index to obtain a historical fault index of the historical fault device, wherein the fault cause and the sub-fault index are in a one-to-one correspondence, and the weight of each sub-fault index in the weighted calculation is obtained according to the sub-fault index; Acquiring a pre-fault index of a pre-abnormal power primary device according to the pre-fault information includes: A pre-failure cause of the pre-failure information is obtained, and a pre-failure index is obtained according to the pre-failure cause, wherein the pre-failure cause and the pre-failure index are in a one-to-one correspondence.

[0030] In this embodiment, a historical fault index is obtained according to historical fault causes, and a pre-failure index is obtained according to pre-failure causes, which provides a specific and comprehensive data basis for obtaining the fault index and obtaining equipment self-check information according to the fault index.

[0031] The method of obtaining a distributed strategy by using distributed edge computing includes: Acquire regional distribution information of the primary power equipment and the secondary power equipment, acquire the edge computing node closest to the secondary power equipment according to the regional distribution information, establish a matching mapping table for the secondary power equipment and the corresponding edge computing node, and establish a data receiving platform and a data processing platform for each edge computing node according to the matching mapping table; The distributed strategy includes a pre-collection strategy, a health assessment strategy and a fault handling strategy.

[0032] In this embodiment, distributed edge computing is used to obtain a matching mapping table, which ensures the orderliness and strategy of using edge computing nodes to obtain equipment operation data of primary power equipment in real time, reduces data transmission delays, reduces the burden on the central server, and increases system scalability and robustness.

[0033] In this embodiment, machine learning technology is used to perform real-time analysis and prediction of the operating data and equipment status of primary power equipment, detect abnormal behavior, and warn of possible failures in advance, thereby realizing intelligent monitoring and predictive maintenance of equipment status and reducing the failure risk and maintenance cost of primary power equipment. Distributed edge computing is used to delegate terminal data processing tasks to edge computing nodes closer to power equipment for processing, which reduces data transmission delays, reduces the burden on central servers, and increases system scalability and robustness; It not only uses the operating data of primary power equipment and the sensor data of the online monitoring system, but also integrates other relevant data sources, such as environmental parameters and weather data. By integrating multi-source data, it can more comprehensively understand the operating conditions of primary power equipment and make more accurate decisions and optimizations for equipment failure warnings. The probability of possible failure of primary power equipment is evaluated based on historical fault information and pre-failure information, and the fault index is obtained. Self-inspection information is obtained based on the fault index. The primary power equipment completes the self-inspection operation, reduces the failure risk of primary power equipment through pre-maintenance, and ensures the stable operation of primary power equipment.

[0034] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0035] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An electrical equipment fault analysis system based on multi-source data fusion, characterized in that: It includes intelligent operation and maintenance module, intelligent collection module, intelligent health assessment module and intelligent fault handling module; The intelligent operation and maintenance module is used to obtain a distributed strategy according to distributed edge computing, and obtain corresponding historical fault information and corresponding pre-fault information through edge computing nodes according to the distributed strategy, obtain device self-test information according to the historical fault information and the pre-fault information, and each edge computing node issues a self-test command according to the device self-test information; The intelligent collection module is used to collect the equipment operation data of the primary power equipment in real time according to the distributed strategy using the secondary power equipment, and obtain the environmental data and weather data from the relevant platform; The intelligent health assessment module is used for the edge computing node to evaluate the health status of the primary power equipment within a unit time according to the equipment operation data and the primary equipment operation standard, obtain the equipment health level of each primary power equipment, and obtain the historical operation data of each primary power equipment, predict the equipment health trend of each primary power equipment according to the historical operation data, the environmental data and the weather data, and screen out abnormal primary power equipment according to the equipment health level and equipment health trend; The intelligent fault processing module is used for the edge computing node to obtain the abnormal equipment data set, and establish a fault diagnosis model based on the abnormal equipment data set using voiceprint recognition technology and machine learning models, and use the fault diagnosis model to obtain the cause of the fault based on the abnormal equipment data of the abnormal primary power equipment, and perform maintenance based on the cause of the fault.

2. The electrical equipment fault analysis system based on multi-source data fusion according to claim 1, characterized in that: The intelligent operation and maintenance module includes an intelligent self-checking unit and an intelligent strategy unit; The intelligent self-check unit is used for edge computing nodes to obtain historical fault information and pre-fault information, obtain historical fault indexes of historical fault devices according to the historical fault information, calculate the historical fault indexes of the historical fault devices to obtain a historical fault table, obtain pre-fault indexes of pre-abnormal power primary devices according to the pre-fault information, calculate the fault indexes of the pre-abnormal power primary devices to obtain a pre-fault table, cross-merge the historical fault table and the pre-fault table, perform weighted calculation on historical fault indexes and pre-fault indexes with the same device number to obtain a fault index, and the historical fault indexes and pre-fault indexes corresponding to other different device numbers are also defined as fault indexes; Define the historical fault equipment and pre-abnormal power primary equipment whose fault index is less than or equal to the first expected threshold as low-risk equipment, and obtain low-risk self-test information for the low-risk equipment; Define the historical fault equipment and pre-abnormal power primary equipment with a fault index greater than the first expected threshold and less than or equal to the second expected threshold as medium-risk equipment, and obtain medium-risk self-test information for the medium-risk equipment; The historical fault equipment and the pre-abnormal power primary equipment whose fault index is greater than the second expected threshold are defined as high-risk equipment, and high-risk self-test information is obtained for the high-risk equipment, and various equipment at different risk levels correspond to corresponding self-test information; Each edge computing node issues a self-check instruction according to the device self-check information; The historical fault information includes the equipment number, historical fault cause, historical fault times, historical maintenance operations and historical maintenance results, and the pre-failure information includes the pre-failure equipment number and pre-failure cause; The intelligent strategy unit is used to obtain a distributed strategy by using distributed edge computing, and send the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information to the corresponding edge computing node according to the distributed strategy; the edge computing node issues a self-test instruction to the corresponding primary power equipment according to the low-risk self-test information, the medium-risk self-test information and the high-risk self-test information, obtains a self-test result, and sends the self-test result to the fault processing module for storage; The device self-test information includes low-risk self-test information, medium-risk self-test information and high-risk self-test information.

3. The electrical equipment fault analysis system based on multi-source data fusion according to claim 1, characterized in that: The intelligent acquisition module includes a device acquisition unit and a multi-source acquisition unit; The device collection unit is used to collect the device operation data of the primary power device by using the secondary power device, and send the device operation data to the corresponding edge computing nodes respectively according to the distributed strategy; The multi-source acquisition unit is used to connect the environment monitoring platform and the weather monitoring platform using an open interface, and to obtain environmental data from the environment monitoring platform and weather data from the weather monitoring platform.

4. The electrical equipment fault analysis system based on multi-source data fusion according to claim 1, characterized in that: The intelligent health assessment module includes a health assessment unit and a health prediction unit; The health assessment unit is used for the edge computing node to evaluate each operation data parameter in the device operation data according to the device operation standard, obtain the parameter level of each operation parameter, and obtain the device health level of the primary power device according to the parameter level. If the device health level is greater than or equal to the expected threshold, the corresponding primary power device is defined as a normal primary power device; if the device health level is less than the expected threshold, the corresponding primary power device is defined as an abnormal primary power device; The health prediction unit is used for the edge computing node to retrieve the historical operation data of the primary power equipment, use the historical operation data to train the machine learning model to predict the equipment health trend of the primary power equipment, and obtain the next period of equipment health level according to the equipment health trend, and define the primary power equipment whose next period of equipment health level is less than the expected threshold as the pre-abnormal primary power equipment; The historical operation data includes historical equipment data, historical environment data, historical weather data and historical health level of the primary power equipment; The historical equipment data includes equipment number, historical equipment main parameters, historical equipment working hours, historical equipment temperature, historical vibration amplitude and historical voiceprint information.

5. The electrical equipment fault analysis system based on multi-source data fusion according to claim 1, characterized in that: The intelligent fault processing module includes a fault sample unit and a fault model unit; The fault sample unit is used to obtain the abnormal operation data set of the primary power equipment from the defect fault simulation platform by using an open interface; The fault model unit is used to obtain voiceprint information and equipment fault information in the abnormal operation data set, extract equipment voiceprint information using a reverberation cancellation method, and use signal enhancement technology and background noise removal methods to achieve background noise reduction, obtain denoised equipment voiceprint information, input the equipment voiceprint information and the fault cause in the equipment fault information into a deep convolutional neural network model for training, use the equipment voiceprint information as input and the fault cause as output to establish a fault diagnosis model, input the abnormal power data of the abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model to obtain the fault cause of the abnormal primary power equipment, input the pre-abnormal power data of the pre-abnormal primary power equipment obtained by the health assessment module into the fault diagnosis model, obtain the pre-fault cause of the pre-abnormal primary power equipment, and issue a fault repair instruction according to the fault cause and the pre-fault cause; The device failure information includes the device number, main parameters of the device and the cause of the failure.

6. The electrical equipment fault analysis system based on multi-source data fusion according to claim 4, characterized in that: The calculation expression for obtaining the equipment health level of the primary power equipment according to the parameter level is as follows: ； Among them, S is the equipment health level, For each parameter level, is the weight of each parameter level, n is a natural number greater than zero, i is a natural number greater than zero, is the fault index of the primary power equipment, T is the unit time, and P is the usage time of the primary power equipment.

7. The electrical equipment fault analysis system based on multi-source data fusion according to claim 4, characterized in that: Using the historical operation data to train a machine learning model to predict the equipment health trend of the primary power equipment includes: The historical environmental data, historical weather data and historical equipment data are used as inputs of the machine learning model, the equipment data of the next period of the historical equipment data is defined as the historical next period equipment data, and the historical next period equipment data is used as the output of the machine learning model to train the machine learning model until the model fit reaches an expected threshold, then the training is terminated, and an equipment health prediction model is obtained; The environmental data, weather data and equipment operation data collected in real time are input into the equipment health prediction model to obtain the next period of primary equipment operation data of the primary power equipment, and the next period of primary equipment operation data is defined as the next period of equipment data.

8. The electrical equipment fault analysis system based on multi-source data fusion according to claim 2, characterized in that: Acquiring the fault index of the historical fault device according to the historical fault information includes: Obtaining historical fault causes in the historical fault information, obtaining each sub-fault index according to each historical fault cause, performing weighted calculation on each sub-fault index to obtain a historical fault index of the historical fault device, wherein the fault cause and the sub-fault index are in a one-to-one correspondence, and the weight of each sub-fault index in the weighted calculation is obtained according to the sub-fault index; Acquiring a pre-fault index of a pre-abnormal power primary device according to the pre-fault information includes: A pre-failure cause of the pre-failure information is obtained, and a pre-failure index is obtained according to the pre-failure cause, wherein the pre-failure cause and the pre-failure index are in a one-to-one correspondence.

9. The electrical equipment fault analysis system based on multi-source data fusion according to claim 2, characterized in that: The method of obtaining a distributed strategy by using distributed edge computing includes: Acquire regional distribution information of the primary power equipment and the secondary power equipment, acquire the edge computing node closest to the secondary power equipment according to the regional distribution information, establish a matching mapping table for the secondary power equipment and the corresponding edge computing node, and establish a data receiving platform and a data processing platform for each edge computing node according to the matching mapping table; The distributed strategy includes a pre-collection strategy, a health assessment strategy and a fault handling strategy.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, an electrical equipment fault analysis system based on multi-source data fusion as described in any one of claims 1 to 9 is implemented.

Images