Power grid equipment fault analysis method and device and electronic equipment
By extracting the operating parameters of power grid equipment and filtering the target feature vectors, the target network diagram is constructed, and the problem of low efficiency of power grid equipment failure analysis is solved, and fast and accurate fault diagnosis is achieved.
Patent Information
- Application Number
- CN202411052170.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-15
AI Technical Summary
The prior art is inefficient when analyzing power grid equipment faults, making it difficult to effectively diagnose and locate faults, resulting in inefficient analysis.
By extracting the operating parameters of each device in the power grid, the device feature vector is obtained, and the target feature vector is filtered according to the information gain value, a target network diagram is constructed to determine the fault type and probability, and alarm information is generated.
It improves the efficiency of grid equipment failure analysis, reduces the amount of subsequent calculated data, and realizes rapid diagnosis and positioning of grid equipment failures.
Smart Images

Figure CN118962334A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of electrical technology, and in particular to a method, device and electronic equipment for analyzing power grid equipment failure. Background Art
[0002] With the development of electrical technology, the types and quantities of devices that can be connected to the distribution network are increasing. Since different devices can affect each other's operating status, when a device in the distribution network fails, it may have a chain effect on other devices, thereby increasing the difficulty of diagnosing and locating the primary fault in the distribution network. As a result, the existing technology has low efficiency in analyzing device failures when analyzing the device information of each device in the distribution network one by one.
[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention
[0004] The present application provides a method, device and electronic device for analyzing power grid equipment failure, so as to at least solve the technical problem of low efficiency in analyzing equipment failure in the power grid in the prior art.
[0005] According to one aspect of the present application, a method for analyzing power grid equipment failures is provided, comprising: extracting features from first data of each of L devices included in the power grid to obtain M device feature vectors for each device, wherein L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating status of the device from different dimensions; determining S target feature vectors of the device based on the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector with an information gain value greater than or equal to a preset threshold; determining a target label and a target probability corresponding to the device based on the S target feature vectors of each device, wherein the target label is used to characterize the failure type of the device, and the target probability is used to characterize the probability of the device having the failure type corresponding to the target label.
[0006] Optionally, in the process of extracting features from the first data of each of the L devices included in the power grid, the method for analyzing power grid equipment failure also includes: filtering the first data of each device through a filter to obtain second data of each device, wherein the data filtering is used to remove interference data that does not carry operating parameters in the first data; performing data cleaning on the second data of each device based on a preset algorithm to obtain third data of each device, wherein the data cleaning is used to supplement the operating parameters missing in the second data due to communication failures; normalizing the operating parameters included in the third data of each device to obtain fourth data of each device, wherein the normalization is used to update the operating parameters included in the third data to within a preset range; performing feature extraction on the fourth data of each of the L devices to obtain M device feature vectors for each device.
[0007] Optionally, before determining the S target feature vectors of the device based on the M device feature vectors of each device, the method for analyzing power grid equipment failure further includes: obtaining Y preset labels corresponding to L devices, wherein Y is a positive integer, and each preset label in the Y preset labels is used to characterize a fault type of the device; obtaining a first parameter corresponding to each preset label in the Y preset labels, wherein the first parameter is used to characterize the entropy of the preset label; obtaining X second parameters corresponding to each device feature vector of the kth device, wherein k is a positive integer less than or equal to L, and the second parameter corresponding to each device feature vector is used to characterize the conditional entropy of the preset label when the kth device is in the value corresponding to the device feature vector, and the device feature vector corresponds to X values; based on the X second parameters corresponding to each device feature vector of the kth device and the first parameter corresponding to each preset label, determine the information gain value corresponding to each device feature vector of the kth device.
[0008] Optionally, in the process of determining the target label and target probability corresponding to each device based on the S target feature vectors of each device, the method for analyzing power grid equipment failure also includes: constructing a target network diagram based on the S target feature vectors corresponding to each device, wherein the target network diagram is used to characterize the dependency relationship between all target feature vectors corresponding to L devices; determining a parent node set for each device based on the target network diagram, wherein the parent node set includes at least one target feature vector, and any target feature vector included in the parent node set is directly connected to a target feature vector of the device; using a combined vector obtained by concatenating all target feature vectors included in the parent node set of each device as the target vector of the device; and determining a target label and target probability corresponding to the device based on the target vector of each device.
[0009] Optionally, in the process of determining the target label and target probability corresponding to each device based on the target vector of the device, the method for analyzing power grid equipment failure also includes: determining Y first probabilities corresponding to the device based on the target vector of the kth device among L devices, wherein the Y first probabilities correspond to Y preset labels one-to-one, and the first probability is used to characterize the posterior probability of the kth device having a fault type corresponding to the preset label when the kth device is in the operating state corresponding to the target vector; taking the largest first probability among the Y first probabilities corresponding to the kth device as the target probability of the kth device; and taking the preset label corresponding to the target probability of the kth device as the target label of the kth device.
[0010] Optionally, before determining the Y first probabilities corresponding to the kth device based on the target vector of the device among L devices, the method for analyzing power grid equipment failure also includes: determining the first sequence corresponding to the kth device based on the parent node set of the kth device, wherein the first sequence is a conditional probability distribution sequence under the values corresponding to each target feature vector of the kth device in the parent node set.
[0011] Optionally, after determining the target label and target probability corresponding to each device based on the S target feature vectors of the device, the method for analyzing power grid equipment failure also includes: treating the device among the L devices whose target probability is greater than a preset probability as a faulty device; generating alarm information based on the target label and first data of the faulty device, and sending the alarm information to the visualization screen of the terminal used by the user.
[0012] According to another aspect of the present application, there is also provided an analysis device for power grid equipment failure, including: a feature extraction unit, used to extract features from first data of each of the L devices included in the power grid, and obtain M device feature vectors for each device, wherein L and M are both positive integers, and the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating status of the device from different dimensions; a first determination unit, used to determine S target feature vectors of the device based on the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold; a second determination unit, used to determine a target label and a target probability corresponding to the device based on the S target feature vectors of each device, wherein the target label is used to characterize the failure type of the device, and the target probability is used to characterize the probability of the device having the failure type corresponding to the target label.
[0013] According to another aspect of the present application, a computer program product is provided, in which a computer program is stored, wherein when the computer program is run, the computer program product is controlled to execute any one of the above-mentioned methods for analyzing a fault of a power grid device.
[0014] According to another aspect of the present application, an electronic device is also provided, wherein the electronic device includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any one of the above-mentioned methods for analyzing power grid equipment failures.
[0015] In the present application, firstly, feature extraction is performed on the first data of each of the L devices included in the power grid to obtain M device feature vectors for each device, wherein L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating status of the device from different dimensions, and then, S target feature vectors of the device are determined based on the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold value, and finally, the target label and target probability corresponding to the device are determined based on the S target feature vectors of each device, wherein the target label is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target label.
[0016] It can be seen from the above content that the technical solution of the present application first extracts features from the operating parameters (i.e., the first data) of all devices in the power grid. Compared with the method of analyzing the device information of each device one by one in the prior art, the present application achieves the purpose of subsequently conducting a comprehensive analysis of the equipment failures in the distribution network based on the operating parameters of all devices. Moreover, after obtaining the M device feature vectors corresponding to each device, in order to further reduce the amount of data subsequently calculated by the present application, the present application uses the device feature vector whose information gain value in the M device feature vectors of each device is greater than or equal to the preset threshold as the target feature vector corresponding to the device, thereby reducing the amount of data that needs to be analyzed by subsequent solutions. Finally, the present application determines the fault type with the highest probability of occurrence of the device (i.e., the target label) and the probability of occurrence of this type of equipment failure (i.e., the target probability) based on the S target feature vectors corresponding to each device, thereby achieving the purpose of analyzing equipment failures in the power grid.
[0017] It can be seen that the present application achieves the purpose of analyzing equipment failures in the power grid by determining the fault type with the highest probability of occurring in the device and the probability of occurrence of this type of equipment failure based on the S target feature vectors corresponding to each device, thereby achieving the technical effect of improving the efficiency of analyzing equipment failures in the power grid, and further solving the technical problem of low efficiency in analyzing equipment failures in the power grid in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0019] Figure 1 is a flow chart of an optional method for analyzing a power grid equipment failure according to an embodiment of the present application;
[0020] Figure 2 is a schematic diagram of an optional power grid equipment fault analysis device according to an embodiment of the present application;
[0021] Figure 3 is a schematic diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.
[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0024] It should also be noted that the relevant information (including equipment operating parameter information) and data (including but not limited to data for display and analysis) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.
[0025] In addition, the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant information and data involved in this application comply with the relevant laws, regulations and standards of the relevant regions, and necessary confidentiality measures have been taken and do not violate public order and good morals. In addition, this application provides corresponding operation entrances for users to choose to agree to authorization or refuse authorization. If the user chooses to refuse authorization, he / she will enter the corresponding expert decision-making process.
[0026] According to an embodiment of the present application, an embodiment of a method for generating an application code is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0027] The present application provides a power grid equipment failure analysis system (hereinafter referred to as the analysis system) for executing the power grid equipment failure analysis method of the present application. Figure 1 is a flow chart of an optional method for analyzing a power grid device failure according to an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:
[0028] Step S101 : extracting features from first data of each of L devices included in a power grid to obtain M device feature vectors for each device.
[0029] In step S101, L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating state of the device from different dimensions.
[0030] Optionally, the L devices include at least a transformer, a circuit breaker, a disconnector and a power terminal in the power grid.
[0031] Optionally, the analysis system can deploy multiple sensors and monitoring devices in the power grid, such as transformers, circuit breakers, disconnectors, etc., and then collect operating data (i.e., first data) of the equipment in the power grid in real time through the sensors, such as current, voltage, vibration, temperature and other parameters.
[0032] Optionally, the frequency of collecting operating data is dynamically adjusted according to the importance level and performance requirements of the equipment, and the analysis system can achieve a sampling accuracy of seconds or milliseconds.
[0033] Optionally, the analysis system characterizes the operating status of the equipment from different dimensions through equipment feature vectors, which can reduce the complexity of the subsequent analysis system in storing and processing equipment-related information. The equipment feature vectors can reduce the amount of data that needs to be processed and the amount of resources consumed in the subsequent decision-making process of the analysis system, thereby improving the efficiency of subsequent equipment fault analysis. At the same time, it reduces the system resource cost of the analysis system for storing and analyzing equipment-related information.
[0034] Step S102: Determine S target feature vectors of each device according to the M device feature vectors of the device.
[0035] In step S102, S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold.
[0036] Optionally, the analysis system can also use a greedy algorithm to gradually select the target feature vector, that is, the analysis system sorts the M device feature vectors corresponding to the device from large to small according to the information gain value of each device feature vector corresponding to the device, and then the analysis system uses the device feature vectors ranked in the top S in the sorting result as the S target feature vectors of the device.
[0037] Optionally, information gain is a method for measuring the purity of a data set. The information gain value can be used to evaluate the contribution of each device feature to the classification of the data set. The analysis system determines S target feature vectors of the device based on the information gain value, thereby achieving the purpose of retaining the device features that contribute most to subsequent device fault analysis. The S target feature vectors can provide strong support for subsequent fault diagnosis and location, while reducing the amount of data that needs to be processed by the subsequent analysis system, thereby further improving the efficiency of subsequent equipment fault analysis.
[0038] Step S103, determining the target label and target probability corresponding to each device according to the S target feature vectors of the device.
[0039] In step S103, the target tag is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target tag.
[0040] Optionally, the analysis system can determine the Y preset labels and Y first probabilities corresponding to the device based on the S target feature vectors of each device, wherein each of the Y preset labels corresponds to a device failure type, the Y first probabilities correspond one-to-one to the Y preset labels, and each of the Y first probabilities is used to characterize the probability of the corresponding failure type occurring in the device. Afterwards, the analysis system takes the maximum first probability among the Y first probabilities as the target probability, and the analysis system takes the preset label corresponding to the target probability as the target label.
[0041] Optionally, after the analysis system obtains the target label and target probability corresponding to each of the L devices, the analysis system determines the device corresponding to the maximum target probability among the L target probabilities as the faulty device, thereby achieving the purpose of analyzing and locating equipment failures in the power grid based on the target label and target probability corresponding to each device.
[0042] It can be seen from the above content that the technical solution of the present application first extracts features from the operating parameters (i.e., the first data) of all devices in the power grid. Compared with the method of analyzing the device information of each device one by one in the prior art, the present application achieves the purpose of subsequently conducting a comprehensive analysis of the equipment failures in the distribution network based on the operating parameters of all devices. Moreover, after obtaining the M device feature vectors corresponding to each device, in order to further reduce the amount of data subsequently calculated by the present application, the present application uses the device feature vector whose information gain value in the M device feature vectors of each device is greater than or equal to the preset threshold as the target feature vector corresponding to the device, thereby reducing the amount of data that needs to be analyzed by subsequent solutions. Finally, the present application determines the fault type with the highest probability of occurrence of the device (i.e., the target label) and the probability of occurrence of this type of equipment failure (i.e., the target probability) based on the S target feature vectors corresponding to each device, thereby achieving the purpose of analyzing equipment failures in the power grid.
[0043] It can be seen that the present application achieves the purpose of analyzing equipment failures in the power grid by determining the fault type with the highest probability of occurring in the device and the probability of occurrence of this type of equipment failure based on the S target feature vectors corresponding to each device, thereby achieving the technical effect of improving the efficiency of analyzing equipment failures in the power grid, and further solving the technical problem of low efficiency in analyzing equipment failures in the power grid in the prior art.
[0044] In an optional embodiment, in the process of extracting features from the first data of each of the L devices included in the power grid, the analysis system first filters the first data of each device through a filter to obtain the second data of each device, wherein the data filtering is used to remove interference data that does not carry operating parameters in the first data. Afterwards, the analysis system cleans the second data of each device based on a preset algorithm to obtain the third data of each device, wherein the data cleaning is used to supplement the operating parameters missing in the second data due to communication failures. Then, the analysis system normalizes the operating parameters included in the third data of each device to obtain the fourth data of each device, wherein the normalization is used to update the operating parameters included in the third data to within a preset range. Finally, the analysis system extracts features from the fourth data of each of the L devices to obtain M device feature vectors for each device.
[0045] Optionally, in the analysis system, firstly based on filtering technology, a low-pass filter or a median filter is used to remove noise signals and interference signals in the collected first data. Then, the analysis system detects whether there is a data missing problem due to sensor failure or communication interruption. If there is a data missing problem, a preset algorithm, such as a linear interpolation algorithm or a Lagrange interpolation algorithm, is used to fill the missing values in the collected data.
[0046] Then, the analysis system normalizes the data obtained after interpolation processing, so as to achieve the purpose of narrowing the numerical range corresponding to different parameters in the data, and avoid the problem of instability in the subsequent analysis system calculation process due to excessive differences in parameter values in the data.
[0047] Optionally, in the above embodiment, the analysis system scales the parameter values in the data obtained after the interpolation process to between 0 and 1, and the formula used in the scaling process is shown in the following formula (1):
[0048]
[0049] In the above formula (1), x is the original parameter value in the data obtained after interpolation processing, x′ is the parameter value obtained after normalization processing, min(x) is the minimum original parameter value in the data obtained after interpolation processing, and max(x) is the maximum original parameter value in the data obtained after interpolation processing.
[0050] Optionally, after the normalization process is completed, the analysis system can also perform a second data cleaning on the data obtained by the normalization process to achieve the purpose of deleting abnormal parameter values in the data. The abnormal parameter values are used to characterize data that exceeds a preset range. The analysis system improves the quality of the preprocessed data through the above-mentioned preprocessing operations, thereby improving the efficiency of feature extraction of the data by the subsequent analysis system.
[0051] In an optional embodiment, before determining the S target feature vectors of the device based on the M device feature vectors of each device, the analysis system first obtains Y preset labels corresponding to the L devices, where Y is a positive integer, and each of the Y preset labels is used to characterize a fault type of the device. Then, the analysis system obtains a first parameter corresponding to each of the Y preset labels, where the first parameter is used to characterize the entropy of the preset label, and the analysis system also obtains X second parameters corresponding to each device feature vector of the kth device, where k is a positive integer less than or equal to L, and the second parameter corresponding to each device feature vector is used to characterize the conditional entropy of the preset label when the kth device is in the value corresponding to the device feature vector, and the device feature vector corresponds to X values. Finally, the analysis system determines the information gain value corresponding to each device feature vector of the kth device based on the X second parameters corresponding to each device feature vector of the kth device and the first parameter corresponding to each preset label.
[0052] Optionally, in the process of determining the information gain value corresponding to each device feature vector of the kth device by the analysis system, the formula used is as shown in the following formula (2):
[0053] Gain(Z)=H(Y)-H(Y|Z)(2)
[0054] In the above formula (2), Gain(Z) represents the information gain value corresponding to the z-th device feature vector of the k-th device, H(Y) represents the entropy of the y-th preset label among the Y preset labels (corresponding to the first parameter), and H(Y|Z) represents the conditional entropy of the y-th preset label when the device operating state Z corresponds to the z-th device feature vector (corresponding to the second parameter).
[0055] Optionally, the analysis system uses multiple device feature vectors with information gain values greater than a preset threshold as feature subsets corresponding to the devices, and then performs fault analysis on the devices in the power grid based on the feature subsets corresponding to each device, wherein the feature subsets corresponding to each device include S target feature vectors corresponding to the device.
[0056] Optionally, the above embodiment achieves the purpose of reducing the dimension of the feature vector that needs to be processed by the subsequent analysis system by selecting a suitable feature subset for each device, and the analysis system uses the device feature vector with an information gain value greater than a preset threshold as the target feature vector corresponding to the device, thereby achieving the purpose of retaining the device feature that contributes the most to the subsequent device fault analysis. The S target feature vectors can provide strong support for subsequent fault diagnosis and positioning, while reducing the amount of data that needs to be processed by the subsequent analysis system, thereby further improving the efficiency of subsequent equipment fault analysis.
[0057] In an optional embodiment, in the process of determining the target label and target probability corresponding to each device based on the S target feature vectors of each device, the analysis system first constructs a target network diagram based on the S target feature vectors corresponding to each device, wherein the target network diagram is used to characterize the dependency relationship between all target feature vectors corresponding to L devices. Then, the analysis system determines the parent node set of each device based on the target network diagram, wherein the parent node set includes at least one target feature vector, and any target feature vector included in the parent node set is directly connected to a target feature vector of the device. Then, the analysis system uses the combined vector obtained by splicing all the target feature vectors included in the parent node set of each device as the target vector of the device. Finally, the analysis system determines the target label and target probability corresponding to each device based on the target vector of the device.
[0058] Optionally, the target network graph is a Bayesian network structure. The Bayesian network is a probabilistic graphical model that can be used to characterize the probabilistic dependency relationship between variables. Specifically, the Bayesian network describes the joint probability distribution of random variables by constructing a directed acyclic graph.
[0059] Optionally, in a Bayesian network (i.e., target network diagram) used for substation defect diagnosis, a node in the Bayesian network can represent a device in the power grid, and can also represent a device characteristic or operating status of a device, that is, the specific definition of the node in the Bayesian network can be adjusted according to the precise granularity of the power grid fault diagnosis problem. If a finer-grained diagnosis is required, each node in the Bayesian network represents a device characteristic or operating status of the device (such as current characteristics, voltage characteristics, and temperature characteristics, etc.); if a coarser-grained diagnosis is required, each node in the Bayesian network represents the overall operating status of a device (such as normal status, fault status, etc.).
[0060] Optionally, the analysis system can determine the association relationship between devices based on the S target feature vectors corresponding to each device, and then construct a Bayesian network based on the association relationship between the devices, wherein the association relationship between the devices can be modeled based on the physical connection, electrical connection, operation logic and other characteristics between the devices. It can be deduced that the analysis system takes into account the factor of the association relationship between different devices in the subsequent process of analyzing the power grid equipment failure based on the target network diagram, thereby improving the accuracy of subsequent equipment failure analysis and positioning.
[0061] Optionally, the analysis system can construct a joint state vector S based on all target feature vectors included in the parent node set corresponding to the kth device k (i.e. the target vector).
[0062] In an optional embodiment, in the process of determining the target label and target probability corresponding to each device based on the target vector of the device, the analysis system first determines Y first probabilities corresponding to the device based on the target vector of the kth device among L devices, wherein the Y first probabilities correspond one-to-one to the Y preset labels, and the first probability is used to characterize the posterior probability of the kth device having a fault type corresponding to the preset label when the kth device is in the operating state corresponding to the target vector. Afterwards, the analysis system uses the largest first probability among the Y first probabilities corresponding to the kth device as the target probability of the kth device, and then the analysis system uses the preset label corresponding to the target probability of the kth device as the target label of the kth device.
[0063] Optionally, the jth first probability P(j|S k ), the formula used by the analysis system is shown in the following formula (3):
[0064]
[0065] Optionally, in the above formula (3), P(S k |j) is the joint state vector S under the condition that the fault type corresponding to the jth preset label among the Y preset labels is given. k The probability of P(j) is the prior probability of the fault type corresponding to the jth preset label, and P(S k ) is the joint state vector S k The marginal probability of .
[0066] In an optional embodiment, before determining the Y first probabilities corresponding to the kth device based on the target vector of the device among L devices, the analysis system first determines the first sequence corresponding to the kth device based on the parent node set of the kth device, wherein the first sequence is a conditional probability distribution sequence under the values corresponding to each target feature vector of the kth device in the parent node set.
[0067] Optionally, after constructing the Bayesian network, the analysis system needs to determine the node parameters in the Bayesian network, that is, the conditional probability distribution sequence corresponding to each node, wherein the conditional probability distribution sequence corresponding to each device can be trained based on the historical operating data of the equipment in the power grid, that is, based on the existing historical operating data, the conditional probability distribution sequence corresponding to each device is determined by maximum likelihood estimation or Bayesian estimation.
[0068] Alternatively, assume that X i represents the i-th operating state corresponding to the k-th device (i.e., the operating state corresponding to the i-th target feature vector in the S target feature vectors), P(X i |X Pa(i) ) represents the parent node X of a given k-th device Pa(i) The conditional probability distribution under the condition, then the joint probability distribution sequence of the Bayesian network (i.e. the first sequence) is as shown in the following formula (4):
[0069]
[0070] In the above formula (4), X Pa(i) Represents the parent node set of the k-th device.
[0071] In an optional embodiment, after determining the target label and target probability corresponding to each device based on the S target feature vectors of the device, the analysis system regards the device among the L devices whose target probability is greater than the preset probability as a faulty device. Thereafter, the analysis system generates an alarm message based on the target label and the first data of the faulty device, and sends the alarm message to the visualization screen of the terminal used by the user.
[0072] Optionally, the analysis system can achieve the purpose of determining the location where the equipment failure may occur based on the equipment association model (i.e., the Bayesian network model corresponding to the target network diagram), that is, by analyzing the equipment association relationship in the equipment association model, the cross-influence between different devices can be determined, and the operating status of each device is comprehensively considered, so that the equipment fault diagnosis results are more accurate and reliable.
[0073] It can be seen from the above content that the technical solution of the present application first extracts features from the operating parameters (i.e., the first data) of all devices in the power grid. Compared with the method of analyzing the device information of each device one by one in the prior art, the present application achieves the purpose of subsequently conducting a comprehensive analysis of the equipment failures in the distribution network based on the operating parameters of all devices. Moreover, after obtaining the M device feature vectors corresponding to each device, in order to further reduce the amount of data subsequently calculated by the present application, the present application uses the device feature vector whose information gain value in the M device feature vectors of each device is greater than or equal to the preset threshold as the target feature vector corresponding to the device, thereby reducing the amount of data that needs to be analyzed by subsequent solutions. Finally, the present application determines the fault type with the highest probability of occurrence of the device (i.e., the target label) and the probability of occurrence of this type of equipment failure (i.e., the target probability) based on the S target feature vectors corresponding to each device, thereby achieving the purpose of analyzing equipment failures in the power grid.
[0074] It can be seen that the present application achieves the purpose of analyzing equipment failures in the power grid by determining the fault type with the highest probability of occurring in the device and the probability of occurrence of this type of equipment failure based on the S target feature vectors corresponding to each device, thereby achieving the technical effect of improving the efficiency of analyzing equipment failures in the power grid, and further solving the technical problem of low efficiency in analyzing equipment failures in the power grid in the prior art.
[0075] According to another aspect of the embodiment of the present application, a device for analyzing a power grid device failure is also provided. Figure 2 is a schematic diagram of an optional power grid equipment failure analysis device according to an embodiment of the present application, such as Figure 2 As shown, the power grid equipment fault analysis device includes: a feature extraction unit 201 , a first determination unit 202 and a second determination unit 203 .
[0076] Optionally, a feature extraction unit is used to extract features from the first data of each of the L devices included in the power grid to obtain M device feature vectors for each device, wherein L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating status of the device from different dimensions; a first determination unit is used to determine S target feature vectors of the device based on the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold; a second determination unit is used to determine a target label and a target probability corresponding to the device based on the S target feature vectors of each device, wherein the target label is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target label.
[0077] In an optional embodiment, the feature extraction unit includes: a filtering subunit, a cleaning subunit, a normalization processing subunit and a feature extraction subunit.
[0078] Optionally, a filtering subunit is used to filter the first data of each device through a filter to obtain the second data of each device, wherein the data filtering is used to remove interference data that does not carry operating parameters in the first data; a cleaning subunit is used to clean the second data of each device based on a preset algorithm to obtain the third data of each device, wherein the data cleaning is used to supplement the operating parameters missing in the second data due to communication failure; a normalization processing subunit is used to normalize the operating parameters included in the third data of each device to obtain the fourth data of each device, wherein the normalization processing is used to update the operating parameters included in the third data to within a preset range; a feature extraction subunit is used to extract features from the fourth data of each of the L devices to obtain M device feature vectors for each device.
[0079] In an optional embodiment, the device for analyzing power grid equipment failure further includes: a first acquisition unit, a second acquisition unit, a third acquisition unit, and a third determination unit.
[0080] Optionally, a first acquisition unit is used to acquire Y preset labels corresponding to L devices, where Y is a positive integer, and each of the Y preset labels is used to characterize a fault type of the device; a second acquisition unit is used to acquire a first parameter corresponding to each of the Y preset labels, where the first parameter is used to characterize the entropy of the preset label; a third acquisition unit is used to acquire X second parameters corresponding to each device feature vector of the kth device, where k is a positive integer less than or equal to L, and the second parameter corresponding to each device feature vector is used to characterize the conditional entropy of the preset label when the kth device is in the value corresponding to the device feature vector, and the device feature vector corresponds to X values; a third determination unit is used to determine the information gain value corresponding to each device feature vector of the kth device based on the X second parameters corresponding to each device feature vector of the kth device and the first parameter corresponding to each preset label.
[0081] In an optional embodiment, the second determining unit includes: a constructing subunit, a first determining subunit, a second determining subunit, and a third determining subunit.
[0082] Optionally, a construction subunit is used to construct a target network diagram based on the S target feature vectors corresponding to each device, wherein the target network diagram is used to characterize the dependency relationship between all target feature vectors corresponding to L devices; a first determination subunit is used to determine the parent node set of each device based on the target network diagram, wherein the parent node set includes at least one target feature vector, and any target feature vector included in the parent node set is directly connected to a target feature vector of the device; a second determination subunit is used to use a combined vector obtained by concatenating all target feature vectors included in the parent node set of each device as the target vector of the device; and a third determination subunit is used to determine the target label and target probability corresponding to each device based on the target vector of the device.
[0083] In an optional embodiment, the third determining subunit further includes: a first determining module, a second determining module and a third determining module.
[0084] Optionally, a first determination module is used to determine Y first probabilities corresponding to the device based on the target vector of the kth device among L devices, wherein the Y first probabilities correspond one-to-one to Y preset labels, and the first probability is used to characterize the posterior probability that the kth device has a fault type corresponding to the preset label when the kth device is in the operating state corresponding to the target vector; the second determination module is used to take the largest first probability among the Y first probabilities corresponding to the kth device as the target probability of the kth device; and the third determination module is used to take the preset label corresponding to the target probability of the kth device as the target label of the kth device.
[0085] In an optional embodiment, the device for analyzing power grid equipment failure further includes: a fourth determination unit.
[0086] Optionally, the fourth determination unit is used to determine a first sequence corresponding to the kth device based on the parent node set of the kth device, wherein the first sequence is a conditional probability distribution sequence under the values corresponding to each target feature vector of the kth device in the parent node set.
[0087] In an optional embodiment, the device for analyzing power grid equipment failure further includes: a fifth determination unit and a generation unit.
[0088] Optionally, the fifth determination unit is used to treat the device among the L devices whose target probability is greater than the preset probability as a faulty device; the generation unit is used to generate alarm information based on the target label and the first data of the faulty device, and send the alarm information to the visualization screen of the terminal used by the user.
[0089] It can be seen from the above content that the analysis device for power grid equipment failure first extracts features from the operating parameters (i.e., the first data) of all devices in the power grid. Compared with the method of analyzing the device information of each device one by one in the prior art, the analysis device for power grid equipment failure achieves the purpose of subsequently performing a comprehensive analysis on the equipment failures in the distribution network based on the operating parameters of all devices. Moreover, after obtaining the M device feature vectors corresponding to each device, in order to further reduce the amount of data subsequently calculated in this application, the analysis device for power grid equipment failure uses the device feature vector whose information gain value in the M device feature vectors of each device is greater than or equal to the preset threshold as the target feature vector corresponding to the device, thereby reducing the amount of data that needs to be analyzed by subsequent schemes. Finally, the analysis device for power grid equipment failure determines the fault type with the highest probability of occurrence of the device (i.e., the target label) and the probability of occurrence of this type of equipment failure (i.e., the target probability) based on the S target feature vectors corresponding to each device, thereby achieving the purpose of analyzing equipment failures in the power grid.
[0090] It can be seen that the power grid equipment failure analysis device achieves the purpose of analyzing equipment failures in the power grid by determining the fault type with the highest probability of occurrence of the device and the probability of occurrence of this type of equipment failure based on the S target feature vectors corresponding to each device, thereby achieving the technical effect of improving the efficiency of analyzing equipment failures in the power grid, and further solving the technical problem of low efficiency in analyzing equipment failures in the power grid in the prior art.
[0091] According to another aspect of an embodiment of the present application, a computer program product is further provided. The computer program product includes a stored computer program, wherein when the computer program is running, the computer program product is controlled to execute any one of the above-mentioned methods for analyzing a fault of a power grid device.
[0092] According to another aspect of an embodiment of the present application, there is also provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned methods for analyzing a fault in a power grid device by executing the executable instructions.
[0093] Optionally, Figure 3 is a schematic diagram of an optional electronic device according to an embodiment of the present application, such as Figure 3 As shown, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, any one of the above-mentioned methods for analyzing power grid equipment failures is implemented.
[0094] Optionally, when the processor executes the program, it can implement the following steps: extract features from the first data of each of the L devices included in the power grid to obtain M device feature vectors for each device, wherein L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating status of the device from different dimensions; determine S target feature vectors of the device based on the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold; determine the target label and target probability corresponding to the device based on the S target feature vectors of each device, wherein the target label is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target label.
[0095] Optionally, the processor can implement the following steps when executing the program: filtering the first data of each device through a filter to obtain the second data of each device, wherein the data filtering is used to remove interference data that does not carry operating parameters in the first data; cleaning the second data of each device based on a preset algorithm to obtain the third data of each device, wherein the data cleaning is used to supplement the operating parameters missing in the second data due to communication failure; normalizing the operating parameters included in the third data of each device to obtain the fourth data of each device, wherein the normalization is used to update the operating parameters included in the third data to within a preset range; performing feature extraction on the fourth data of each of the L devices to obtain M device feature vectors for each device.
[0096] Optionally, when the processor executes the program, the following steps can be implemented: obtaining Y preset labels corresponding to L devices, where Y is a positive integer, and each preset label in the Y preset labels is used to characterize a fault type of the device; obtaining a first parameter corresponding to each preset label in the Y preset labels, where the first parameter is used to characterize the entropy of the preset label; obtaining X second parameters corresponding to each device feature vector of the kth device, where k is a positive integer less than or equal to L, and the second parameter corresponding to each device feature vector is used to characterize the conditional entropy of the preset label when the kth device is in the value corresponding to the device feature vector, and the device feature vector corresponds to X values; determining the information gain value corresponding to each device feature vector of the kth device based on the X second parameters corresponding to each device feature vector of the kth device and the first parameter corresponding to each preset label.
[0097] Optionally, when the processor executes the program, it can implement the following steps: construct a target network diagram based on the S target feature vectors corresponding to each device, wherein the target network diagram is used to characterize the dependency relationship between all target feature vectors corresponding to L devices; determine the parent node set of each device based on the target network diagram, wherein the parent node set includes at least one target feature vector, and any target feature vector included in the parent node set is directly connected to a target feature vector of the device; use the combined vector obtained by concatenating all target feature vectors included in the parent node set of each device as the target vector of the device; and determine the target label and target probability corresponding to each device based on the target vector of the device.
[0098] Optionally, when the processor executes the program, it can implement the following steps: determine Y first probabilities corresponding to the device based on the target vector of the kth device among L devices, wherein the Y first probabilities correspond one-to-one to Y preset labels, and the first probability is used to characterize the posterior probability that the kth device has a fault type corresponding to the preset label when the kth device is in the operating state corresponding to the target vector; use the largest first probability among the Y first probabilities corresponding to the kth device as the target probability of the kth device; and use the preset label corresponding to the target probability of the kth device as the target label of the kth device.
[0099] Optionally, when the processor executes the program, it can implement the following steps: determine the first sequence corresponding to the kth device based on the parent node set of the kth device, wherein the first sequence is a conditional probability distribution sequence under the values corresponding to each target feature vector of the kth device in the parent node set.
[0100] Optionally, when the processor executes the program, it can implement the following steps: treat the device among L devices whose target probability is greater than the preset probability as a faulty device; generate alarm information based on the target tag and the first data of the faulty device, and send the alarm information to the visualization screen of the terminal used by the user.
[0101] It can be seen from the above content that the technical solution of the present application first extracts features from the operating parameters (i.e., the first data) of all devices in the power grid. Compared with the method of analyzing the device information of each device one by one in the prior art, the present application achieves the purpose of subsequently conducting a comprehensive analysis of the equipment failures in the distribution network based on the operating parameters of all devices. Moreover, after obtaining the M device feature vectors corresponding to each device, in order to further reduce the amount of data subsequently calculated by the present application, the present application uses the device feature vector whose information gain value in the M device feature vectors of each device is greater than or equal to the preset threshold as the target feature vector corresponding to the device, thereby reducing the amount of data that needs to be analyzed by subsequent solutions. Finally, the present application determines the fault type with the highest probability of occurrence of the device (i.e., the target label) and the probability of occurrence of this type of equipment failure (i.e., the target probability) based on the S target feature vectors corresponding to each device, thereby achieving the purpose of analyzing equipment failures in the power grid.
[0102] It can be seen that the present application achieves the purpose of analyzing equipment failures in the power grid by determining the fault type with the highest probability of occurring in the device and the probability of occurrence of this type of equipment failure based on the S target feature vectors corresponding to each device, thereby achieving the technical effect of improving the efficiency of analyzing equipment failures in the power grid, and further solving the technical problem of low efficiency in analyzing equipment failures in the power grid in the prior art.
[0103] The above-mentioned embodiments or examples disclosed in the present application are not exhaustive, but are only illustrative of some embodiments or examples, and are not intended to be specific limitations on the scope of protection disclosed in the present application. In the absence of contradiction, each step in a certain embodiment or example in the present application can be implemented as an independent example, and the steps can be combined arbitrarily. For example, the scheme after removing some steps in a certain embodiment or example can also be implemented as an independent example, and the order of the steps in a certain embodiment or example can be arbitrarily exchanged. In addition, the optional methods or optional examples in a certain embodiment or example can be combined arbitrarily; in addition, the various embodiments or examples can be combined arbitrarily, for example, some or all steps of different embodiments or examples can be combined arbitrarily, and a certain embodiment or example can be combined arbitrarily with the optional methods or optional examples of other embodiments or examples.
[0104] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.
[0105] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0106] 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.
[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0108] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
[0109] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transSitory media), such as modulated data signals and carrier waves.
[0110] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0111] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0112] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A method for analyzing power grid equipment failure, characterized in that: include: Performing feature extraction on first data of each device among L devices included in the power grid to obtain M device feature vectors of each device, wherein L and M are both positive integers, the first data is used to characterize the operating parameters of the device collected by the sensor, and the device feature vector is used to characterize the operating state of the device from different dimensions; Determine S target feature vectors of the device according to the M device feature vectors of each device, where S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold; The target label and target probability corresponding to each device are determined based on the S target feature vectors of the device, wherein the target label is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target label.
2. The method for analyzing power grid equipment failure according to claim 1, characterized in that: Feature extraction is performed on the first data of each device in the L devices included in the power grid to obtain M device feature vectors of each device, including: Performing data filtering on the first data of each device through a filter to obtain second data of each device, wherein the data filtering is used to remove interference data that does not carry operating parameters in the first data; Performing data cleaning on the second data of each device based on a preset algorithm to obtain third data of each device, wherein the data cleaning is used to supplement the operating parameters missing from the second data due to the communication failure; Normalizing the operating parameters included in the third data of each device to obtain fourth data of each device, wherein the normalization is used to update the operating parameters included in the third data to within a preset range; Feature extraction is performed on the fourth data of each of the L devices to obtain M device feature vectors of each device.
3. The method for analyzing power grid equipment failure according to claim 1, characterized in that: Before determining the S target feature vectors of each device according to the M device feature vectors of the device, the power grid device fault analysis method further includes: Obtaining Y preset labels corresponding to the L devices, where Y is a positive integer, and each of the Y preset labels is used to characterize a fault type of the device; Obtaining a first parameter corresponding to each preset tag among the Y preset tags, wherein the first parameter is used to characterize the entropy of the preset tag; Obtain X second parameters corresponding to each device feature vector of the k-th device, where k is a positive integer less than or equal to L, and the second parameter corresponding to each device feature vector is used to characterize the conditional entropy of the preset label when the k-th device is in the value corresponding to the device feature vector, and the device feature vector corresponds to X values; An information gain value corresponding to each device feature vector of the k-th device is determined according to the X second parameters corresponding to each device feature vector of the k-th device and the first parameter corresponding to each preset label.
4. The method for analyzing power grid equipment failure according to claim 1, characterized in that: Determining a target label and a target probability corresponding to each device according to the S target feature vectors of the device includes: Constructing a target network graph according to the S target feature vectors corresponding to each device, wherein the target network graph is used to characterize the dependency relationship between all target feature vectors corresponding to the L devices; Determining a parent node set of each device according to the target network graph, wherein the parent node set includes at least one target feature vector, and any target feature vector included in the parent node set is directly connected to a target feature vector of the device; A combined vector obtained by concatenating all target feature vectors included in the parent node set of each device is used as the target vector of the device; The target label and target probability corresponding to each device are determined according to the target vector of the device.
5. The method for analyzing power grid equipment failure according to claim 4, characterized in that: In the process of determining the target label and target probability corresponding to each device according to the target vector of each device, the method for analyzing the fault of the power grid device further includes: Determine Y first probabilities corresponding to the device according to the target vector of the kth device among the L devices, wherein the Y first probabilities correspond one-to-one to the Y preset labels, and the first probabilities are used to characterize the posterior probability that the kth device has a fault type corresponding to the preset label when the kth device is in an operating state corresponding to the target vector; Taking the largest first probability among the Y first probabilities corresponding to the k-th device as the target probability of the k-th device; The preset label corresponding to the target probability of the k-th device is used as the target label of the k-th device.
6. The method for analyzing power grid equipment failure according to claim 5, characterized in that: Before determining Y first probabilities corresponding to the kth device among the L devices according to the target vector of the device, the power grid device failure analysis method further includes: A first sequence corresponding to the kth device is determined according to the parent node set of the kth device, wherein the first sequence is a conditional probability distribution sequence under the value corresponding to each target feature vector of the kth device in the parent node set.
7. The method for analyzing power grid equipment failure according to claim 1, characterized in that: After determining the target label and target probability corresponding to each device according to the S target feature vectors of each device, the power grid device failure analysis method further includes: The device with a target probability greater than a preset probability among the L devices is regarded as a faulty device; An alarm message is generated according to the target tag of the faulty device and the first data, and the alarm message is sent to a visualization screen of a terminal used by a user.
8. A device for analyzing power grid equipment failure, characterized in that: include: a feature extraction unit, configured to extract features from first data of each device among L devices included in the power grid, to obtain M device feature vectors of each device, wherein L and M are both positive integers, the first data are used to characterize operating parameters of the device collected by the sensor, and the device feature vectors are used to characterize operating states of the device from different dimensions; A first determining unit, configured to determine S target feature vectors of the device according to the M device feature vectors of each device, wherein S is a positive integer less than or equal to M, and the target feature vector is a device feature vector whose information gain value is greater than or equal to a preset threshold; The second determination unit is used to determine the target label and target probability corresponding to the device based on the S target feature vectors of each device, wherein the target label is used to characterize the fault type of the device, and the target probability is used to characterize the probability of the device having the fault type corresponding to the target label.
9. A computer program product, characterized in that The computer program product comprises a computer program, wherein when the computer program is run, the computer program product is controlled to execute the method for analyzing a fault of a power grid device according to any one of claims 1 to 7.
10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the power grid equipment fault analysis method described in any one of claims 1 to 7.
