Abnormal operation fault diagnosis method and system for photovoltaic inverter and charging pile

Abstract

The invention provides a photovoltaic inverter and charging pile abnormal operation fault diagnosis method and system, and relates to the field of power terminal operation fault diagnosis. The method comprises the following steps: collecting operation data of a photovoltaic inverter and a charging pile in normal operation, and screening out a normal service data set and an abnormal operation data set from the operation data according to actual service requirements; and analyzing the occurrence reason of the abnormal operation data, and writing the abnormal operation data into the abnormal attribute corresponding to the abnormal operation data. Carrying out discretization processing on the abnormal operation data, and calculating an information gain rate of an abnormal attribute; and selecting the abnormal attribute with the maximum information gain rate as a basis for dividing root nodes, and forming an abnormal state decision tree. And analyzing the abnormal state decision tree to obtain an abnormal type and abnormal details of the current terminal equipment. According to the invention, effective fault diagnosis of abnormal operation behaviors of the photovoltaic inverter and the charging pile can be realized.

Description

Technical Field

[0001] The present invention relates to the field of power terminal operation fault diagnosis, and in particular to a method for diagnosing abnormal operation faults of photovoltaic inverters and charging piles. Background Art

[0002] With the continuous advancement and improvement of the construction of the power Internet of Things system, the types and numbers of terminal devices and applications are increasing, and communication business interaction, data networking transmission and other tasks are becoming more and more frequent. The high proportion of new energy photovoltaic access has led to the emergence of a large number of various new intelligent application terminals. For example, distributed photovoltaics and charging piles, in actual business, since the integrated energy operation and management platform can control tens of thousands or even hundreds of thousands of distributed photovoltaics or charging piles, the concept of virtual power plants has gradually emerged. That is, in the case of insufficient power grid load, the integrated energy operation and management platform can be given control instructions, and the integrated energy operation and management platform can adjust the power of the photovoltaics and charging piles under its jurisdiction to ensure the stability of the power grid.

[0003] With the access of more and more new energy terminals, the real-time security protection requirements of IoT terminals are also gradually increasing. The frequent interaction between new energy terminals and the integrated energy operation and management platform is prone to some malicious situations, including (1) the integrated energy operation and management platform maliciously tampering with data, such as falsely reporting the response to the grid load adjustment instructions, forging the power of charging users, etc.; (2) the integrated energy operation and management platform is controlled by hackers, and the charging pile power is increased when the grid is overloaded, increasing the grid load; malicious continuous start and stop operations such as photovoltaics and charging piles affect the power quality, etc. Therefore, it is urgent to improve the fault diagnosis of the group behavior of power IoT terminals and improve the reliability of interactive data. Summary of the invention

[0004] Purpose of the invention: A fault diagnosis method for abnormal operation of a photovoltaic inverter and a charging pile is proposed, and a system for implementing the above fault diagnosis method is proposed to effectively solve the above problems existing in the prior art.

[0005] The present invention provides a method for diagnosing abnormal operation faults of a photovoltaic inverter and a charging pile, the method comprising the following steps:

[0006] S1. Collecting the operating data of the photovoltaic inverter and the charging pile under normal operation, and filtering out the normal business data set and the abnormal operation data set from the operating data according to actual business needs;

[0007] S2. Analyze the causes of abnormal operation data, divide them into three types of abnormalities: environmental reasons, system abnormalities, and comprehensive energy operation management safety reasons, and write them into the abnormal attributes corresponding to the abnormal operation data;

[0008] S3, discretize the abnormal operation data and calculate the information gain rate of the abnormal attributes;

[0009] S4, selecting the abnormal attribute with the largest information gain rate as the basis for dividing the root node; using the dichotomy method to divide all the abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree;

[0010] S5. Analyze the abnormal state decision tree to obtain the terminal device, abnormal type, and abnormal details corresponding to the current abnormal operation data.

[0011] In a further embodiment, the operating data in step S1 includes real-time data collected under daily business conditions and historical data under past normal operations; the real-time data is collected from the comprehensive energy operation management platform; and the historical data is collected from the cloud master station business monitoring system.

[0012] In a further embodiment, after collecting the operating data of the photovoltaic inverter and the charging pile under normal operation, it also includes: summarizing the historical data, drawing a comparison curve, and using the comparison curve to distinguish the normal business data fluctuation range and the category of abnormal operating data.

[0013] In a further embodiment, the process of calculating the information gain rate of the abnormal attribute specifically includes:

[0014] Calculate the split information degree of the abnormal attribute M:

[0015]

[0016] In the formula, S represents the abnormal operation status data set; after the abnormal attributes are divided, it is i sub-datasets, and the sample of the jth sub-data is S j express;

[0017] The information gain calculated by dividing the abnormal attribute M into subsets is:

[0018] InfoGain(S,M)=H(S)-H M (S)

[0019] Where H(S) represents the information entropy of the abnormal operation status data set S, which represents the uncertainty and confusion degree in the data set; H M (S) represents the information entropy of the abnormal operation state data set S under the abnormal attribute M state;

[0020] The information entropy H(S) of the abnormal operation status data set S is calculated as follows:

[0021]

[0022] The information entropy H of the abnormal operation state data set S under the abnormal attribute M state M (S) is calculated as follows:

[0023]

[0024] Where P i represents the probability of the occurrence of the i-th abnormal attribute; m is the number of abnormal attribute categories; M j represents the jth abnormal attribute state; P(M j ) represents the probability of the jth abnormal attribute state appearing;

[0025] The information gain rate obtained according to the information gain InfoGain(S,M) is:

[0026]

[0027] Where InfoGain(S,M) represents the information gain calculated by dividing the abnormal attribute M into subsets; SplitInfo M (S) represents the split information degree of the abnormal attribute M.

[0028] In a further embodiment, after calculating the information gain rate of the abnormal attribute, the obtained information gain rate RainRatio(S,M) is used as the criterion for node division:

[0029] If the information gain rate of the current node is the largest, this node is selected as the split attribute.

[0030] In a further embodiment, a binary method is used to divide all abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree, which specifically includes:

[0031] Arrange all abnormal operation status data sets S in the node according to the size of the values, and obtain the value sequence of the abnormal operation status data set S of S;

[0032] There are n-1 binary methods in the value sequence of S. Each method will get a threshold, a total of n-1 thresholds, and then calculate the information gain rate under the binary result;

[0033] Record n-1 information gain rates in ascending order to form an abnormal state decision tree, extract the method with the largest information gain rate, and select this method as the attribute for dividing the abnormal operation state data set S.

[0034] The second aspect of the present invention proposes a photovoltaic inverter and charging pile abnormal operation fault diagnosis system. The fault diagnosis system is connected to the existing integrated energy operation management platform and the power Internet of Things terminal equipment in the power communication system to realize operation fault diagnosis, and make fault diagnosis of different states under abnormal operation state; in addition, the fault diagnosis system can automatically execute the photovoltaic inverter and charging pile abnormal operation fault diagnosis method as described in the first aspect.

[0035] Specifically, the fault diagnosis system includes a data acquisition unit, an abnormality type judgment unit, a first calculation unit, a decision tree generation unit, and a parsing unit.

[0036] The data acquisition unit is used to collect the operating data of the photovoltaic inverter and the charging pile under normal operation, and filter out the normal business data set and the abnormal operation data set from the operating data according to actual business needs.

[0037] The abnormal type judgment unit is used to analyze the causes of abnormal operation data, divide them into three abnormal types: self-environment reasons, system abnormality reasons and comprehensive energy operation management safety reasons, and write them into the abnormal attributes corresponding to the abnormal operation data.

[0038] The first calculation unit is used to discretize the abnormal operation data and calculate the information gain rate of the abnormal attribute.

[0039] The decision tree generation unit is used to select the abnormal attribute with the largest information gain rate as the basis for dividing the root node; use the dichotomy method to divide all the abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree.

[0040] The parsing unit is used to parse the abnormal state decision tree to obtain the terminal device, abnormal type, and abnormal details corresponding to the current abnormal operation data.

[0041] According to a third aspect of the present invention, an electronic device is provided, comprising a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the abnormal operation fault diagnosis method of the photovoltaic inverter and the charging pile as described in the first aspect is implemented.

[0042] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, in which at least one executable instruction is stored. When the executable instruction is executed on an electronic device, the electronic device executes the abnormal operation fault diagnosis method of the photovoltaic inverter and the charging pile as described in the first aspect.

[0043] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0044] Implement operational fault diagnosis for the power IoT terminal equipment connected to the integrated energy operation management platform and the power communication system, make fault diagnosis of different states under abnormal operation states, effectively conduct timely preventive fault diagnosis for abnormal behaviors of the power IoT terminal, and provide good advance guidance for the issuance of defense measures. Based on the generation of abnormal state decision trees, discrete data and numerical data can be effectively processed, solving the problem of weak single attribute interpretation ability of data types in communication transmission, and at the same time adding pruning operations to solve the problem of overfitting when the data transmission volume of the IoT terminal is large. The present invention provides reliable support for promoting the development of optimization of power IoT terminal fault diagnosis systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the data collection, transmission and aggregation framework of the power Internet of Things terminal business in an embodiment of the present invention.

[0046] Figure 2 This is a classification diagram of abnormal operation of the power Internet of Things terminal in an embodiment of the present invention.

[0047] Figure 3 Schematic diagram of the decision tree for abnormal state of the power Internet of Things terminal in an embodiment of the present invention.

[0048] Figure 4 It is a flow chart of abnormal state decision tree generation based on C4.5 algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described.

[0050] This embodiment discloses a method for diagnosing abnormal operation faults of photovoltaic inverters and charging piles based on decision tree technology. The method includes two main parts: abnormal data collection and decision tree C4.5 construction.

[0051] Abnormal data collection and processing

[0052] exist Figure 1During the operation of the distributed photovoltaic and charging piles shown in the figure, the photovoltaic inverters and charging piles are controlled by the grid access, and collect data such as current, voltage and power through the electric meter (generally once every 1-15 minutes, see the data set for details); on the other hand, they are also subject to the command control of the integrated energy operation and management platform, and can perform remote control operations such as start and stop, power adjustment, etc. At the same time, the integrated energy operation and management platform also collects data such as current, voltage and power of photovoltaic and charging piles (generally once every 1 minute).

[0053] When an abnormality is detected in the power IoT terminal in the fault diagnosis system, some charging piles and photovoltaic inverters in the historical database of the procurement system and the integrated energy operation management platform are analyzed for common abnormality types and other problems, and data classification is performed. Figure 2 .

[0054] Decision tree properties and C4.5 applications

[0055] By collecting and sorting the abnormal data of the IoT terminal, the collected data is continuous and the attributes of the data are numerical type attributes. The value of each data point can be used as the attribute of the abnormal condition. Through analysis, the characteristics of different abnormal conditions can be understood. Among them, the type of abnormal situation can be used as the attribute of the abnormal judgment of the IoT terminal. The various types of abnormalities that may occur when the IoT terminal equipment is running fault diagnosis are distinguished. At this time, the data is different from the continuous collection data, and is discrete data, and the data attributes are nominal attributes.

[0056] Different IoT terminal devices are defined as a root node, different abnormal operating states are defined as separated non-leaf nodes, and finally the occurrence state is defined as the result of the non-leaf node. The attributes connected between the two are expressed as fault diagnosis analysis.

[0057] The information gain of a certain attribute plays a decisive role in the classification of nodes in the decision tree. The larger the information gain value, the more suitable the attribute is for classification. In the C4.5 algorithm, the concept of information gain rate is proposed. The size of the information gain rate is used as the specification for selecting attributes in the present invention, and it can be used to process numerical data and discretized data at the same time to generate a more complete decision tree. After the completed abnormal state decision tree is formed, a pruning operation is added to solve the problem of data overfitting that may occur.

[0058] The specific implementation process of this plan is given below.

[0059] See also Figure 1The figure shows the data collection and transmission between the power IoT terminal of the present invention and the power communication system and the integrated energy operation management platform under the integrated energy operation management. Among them, the data interaction between the integrated energy operation management platform and the power system acquisition system will be recorded in the historical database. Figure 2 The abnormal state classification basis in the invention; the fusion terminal connected to the photovoltaic inverter and the charging pile can be used for the fault diagnosis construction of the intelligent algorithm, including the Figure 3 Abnormal state decision tree generation based on C4.5.

[0060] The classification and causes of abnormal states of the power IoT terminal are as follows:

[0061] ① Objective influences are reflected in the unsatisfactory data collection environment around photovoltaic inverters and charging piles. On the one hand, when the quality of 4G / 5G wireless network signals decreases, data transmission will be unstable, and packet loss or delays will occur. On the other hand, measurement errors caused by failure of the collection equipment or uncontrollable interference to the collection equipment will affect the accuracy of key parameters such as current, voltage and power, causing the fault diagnosis system to make wrong judgments and decisions based on wrong data.

[0062] ② Collection system abnormality, which is manifested in abnormal behavior of meters and sensors during the data collection process. On the one hand, meter failure will lead to the inability to accurately collect key parameters in power data, affecting the cloud's monitoring of photovoltaic inverters and charging piles. Meter failure may also lead to data loss and the inability to record and transmit real-time continuous power data. On the other hand, sensor failure will cause the photovoltaic inverter and charging pile to be unable to transmit real-time status information, causing the equipment to operate under inappropriate conditions, increasing the risk of failure. Secondly, the collected data may affect the management of the equipment. For example, temperature sensor failure will eventually affect the heat dissipation management of the equipment.

[0063] ③ The integrated energy operation and management security platform is abnormal, which is reflected in the situation when the integrated energy operation and management platform receives malicious and illegal intrusion. On the one hand, malicious code is injected into the integrated energy operation and management platform during the interaction process, which tampers with the data, including false reporting and misrepresentation of the response to the adjustment instructions issued by the power grid and falsification of the user electricity consumption of photovoltaic and charging piles; on the other hand, malicious control instructions are implanted in the integrated energy operation and management platform to increase the power of charging piles when the power grid is overloaded, thereby increasing the load on the power grid; malicious continuous start and stop operations are performed on photovoltaic and charging piles, affecting the quality of electricity.

[0064] And in Figure 1In the communication protocol between IoT terminal devices and converged terminals, the Modbus communication protocol is used. After the connection between the device and the host computer is established, the data query command of the device is sent and the data type returned by the device is a hexadecimal string. Therefore, it is still necessary to write code to convert the hexadecimal string into ASCII code. A static method convertHexToString written in Java is used as follows:

[0065] First, create a StringBuilder object stringBuilder to build the output string; traverse the input hexadecimal string hex, take adjacent characters each time, convert the two characters into an integer value decimal, and use the lnteger.parselent(output, 16) method, where 16 represents hexadecimal parsing; cast the integer value decimal to char type and append it to stringBuilder; after traversing all character pairs, return stringBuilder.toString(), which is the final constructed string.

[0066] See also Figure 4 , a flowchart is generated for the abnormal state decision tree. The specific method is as follows: first, the abnormal data set of the power Internet of Things terminal is imported and discretized, and then the information gain rate of all attributes is calculated, and the attribute with the largest gain rate is selected as the basis for dividing the root node. The most appropriate attribute is selected and the data set is divided according to the attribute value. Since there may be multiple characteristic values, it will lead to the situation of dividing multiple data sets. In this case, after the first division, the remaining data needs to be divided into another branch node for subsequent data division. At this node, the attribute corresponding to the maximum result is selected by calculating the information gain rate, and the data is further divided using this attribute. Then, this recursive method is used to process the data set. The recursive calculation ends when all the attributes of the data set are divided, or all the data under the branch belong to the same category.

[0067] The C4.5 algorithm is used to establish the fault condition attributes. The information gain rate of each attribute is calculated through the formula, and the attribute with the maximum gain rate value is selected to divide the given set. The specific calculation process is as follows:

[0068] First, the split information is calculated. Assuming that the abnormal operation status data set of a photovoltaic inverter (hereinafter referred to as a sample) is S, and it is divided into i sub-data sets after the abnormal attribute (defined as attribute M) occurs safely through the integrated energy operation management platform. The sample of the jth sub-data is represented by Sj, then the split information degree of attribute M is:

[0069]

[0070] The information gain calculated by dividing the attribute M into subsets is:

[0071] InfoGain(S,M)=H(S)-H M (S)

[0072] The information gain rate obtained according to the information gain is:

[0073]

[0074] After the calculation is completed, the information gain rate is used as the criterion for node division. If the information gain rate of the node is the largest, this node is selected as the split attribute.

[0075] In addition, the judgment condition for obtaining the decision tree for abnormal state decision in the present invention also uses the C4.5 algorithm, and the specific process is as follows:

[0076] Arrange all sample data in the node according to the size of the value to obtain the attribute sequence of the abnormal operation status data set S.

[0077] There are n-1 binary division methods in the value sequence of the abnormal operation status data set S. Each method will get a threshold, and there are n-1 thresholds in total. Then, the information gain rate under the binary division result is calculated.

[0078] The information gain rate of all methods with the best result is recorded in ascending order, and the method with the largest information gain rate is obtained, and this method is selected as the attribute for dividing S.

[0079] That is, the fault condition attributes can be established in the end, and the judgment conditions for the decision tree to make fault decisions can be obtained.

[0080] After the program traverses all the values ​​of the features that divide the data set, and determines whether the child data set is empty, if it is not empty, it recursively calls the createTree() function to add a child node, and adds the return value to the variable node. Therefore, after the recursion ends, the variable node will be nested with many dictionary data representing the abnormal state decision tree node information, as shown below:

[0081] {'PV inverter': {'Objective impact': {'Unstable signal': 'Low data continuity'},

[0082] {'Integrated energy operation and management security anomaly': 'Malicious data modification': 'Abnormal voltage / power fluctuation'}}}

[0083] Since the abnormal state decision tree stored in the dictionary type seems difficult to understand and not very direct and beautiful, it is expressed in the form of a tree diagram generated by the GitMind drawing tool code. The final form is Figure 3 shown.

[0084] In summary, the present invention is conceived on the basis of the power IoT terminal fault diagnosis system and the decision tree generation of the C4.5 algorithm, and then develops a power IoT terminal group fault diagnosis method. Through the design of this method, an abnormal diagnosis decision can be made for different similar abnormal states and their cause attributes, improving the efficiency of behavioral diagnosis of abnormal states, timely analyzing the abnormalities of IoT terminals and proposing relevant governance measures, thereby providing a scientific reference basis for power companies to safely and reliably diagnose faults of new energy terminals.

[0085] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for diagnosing abnormal operation faults of photovoltaic inverters and charging piles, characterized in that: The steps include: S1. Collecting the operating data of the photovoltaic inverter and the charging pile under normal operation, and filtering out the normal business data set and the abnormal operation data set from the operating data according to actual business needs; S2. Analyze the causes of abnormal operation data, divide them into three types of abnormalities: environmental reasons, system abnormalities, and comprehensive energy operation management safety reasons, and write them into the abnormal attributes corresponding to the abnormal operation data; S3, discretize the abnormal operation data and calculate the information gain rate of the abnormal attributes; S4, selecting the abnormal attribute with the largest information gain rate as the basis for dividing the root node; using the dichotomy method to divide all the abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree; S5. Analyze the abnormal state decision tree to obtain the terminal device, abnormal type, and abnormal details corresponding to the current abnormal operation data.

2. The photovoltaic inverter and charging pile abnormal operation fault diagnosis method according to claim 1, characterized in that: The operation data in step S1 includes real-time data collected in a daily business environment and historical data under past normal operation; The real-time data is collected from the comprehensive energy operation management platform; the historical data is collected from the cloud master station business monitoring system.

3. The photovoltaic inverter and charging pile abnormal operation fault diagnosis method according to claim 2 is characterized in that: After collecting the operating data of the photovoltaic inverter and charging pile under normal operation, it also includes: The historical data are summarized and a comparison curve is drawn to distinguish the normal business data fluctuation range and the category of abnormal operation data by the comparison curve.

4. The photovoltaic inverter and charging pile abnormal operation fault diagnosis method according to claim 1, characterized in that: Calculate the information gain rate of abnormal attributes, including: Calculate the split information degree of the abnormal attribute M: In the formula, S represents the abnormal operation status data set, which is divided into i sub-data sets after the abnormal attributes are divided, and the sample of the j-th sub-data is S j express; The information gain calculated by dividing the abnormal attribute M into subsets is: InfoGain(S,M)=H(S)-H M (S) Where H(S) represents the information entropy of the abnormal operation status data set S; H M (S) represents the information entropy of the data set S under the abnormal attribute M state; The information entropy H(S) of the abnormal operation status data set S is calculated as follows: The information entropy H of the abnormal operation state data set S under the abnormal attribute M state M (S) is calculated as follows: Where Pi represents the probability of the occurrence of the i-th abnormal attribute; m is the number of abnormal attribute categories; M j represents the jth abnormal attribute state; P(M j ) represents the probability of the jth abnormal attribute state appearing; The information gain rate obtained according to the information gain InfoGain(S, M) is: Where InfoGain(S, M) represents the information gain calculated by dividing the abnormal attribute M into subsets; SplitInfo M (S) represents the split information degree of the abnormal attribute M.

5. The photovoltaic inverter and charging pile abnormal operation fault diagnosis method according to claim 4 is characterized in that: After calculating the information gain rate of the abnormal attribute, the obtained information gain rate RainRatio(S, M) is used as the criterion for node division: If the information gain rate of the current node is the largest, this node is selected as the split attribute.

6. The photovoltaic inverter and charging pile abnormal operation fault diagnosis method according to claim 5, characterized in that: Use the binary method to divide all abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree, which specifically includes: Arrange all abnormal operation status data sets S in the node according to the size of the values ​​to obtain the value sequence of the abnormal operation status data set S; There are n-1 binary methods in the value sequence of the abnormal operation status data set S. Each method will get a threshold, a total of n-1 thresholds, and then calculate the information gain rate under the binary result; Record n-1 information gain rates in ascending order to form an abnormal state decision tree, extract the method with the largest information gain rate, and select this method as the attribute for dividing the abnormal operation state data set S.

7. A photovoltaic inverter and charging pile abnormal operation fault diagnosis system, characterized in that: The fault diagnosis system is connected to the existing integrated energy operation management platform and the power Internet of Things terminal equipment in the power communication system to realize operation fault diagnosis, and make fault diagnosis in different states for abnormal operation states; The fault diagnosis system can automatically execute the abnormal operation fault diagnosis method of the photovoltaic inverter and the charging pile as described in any one of claims 1 to 6.

8. The photovoltaic inverter and charging pile abnormal operation fault diagnosis system according to claim 7, characterized in that: The fault diagnosis system includes: Data acquisition unit; the data acquisition unit is used to collect operating data of the photovoltaic inverter and the charging pile under normal operation, and filter out normal business data sets and abnormal operation data sets from the operating data according to actual business needs; Abnormal type judgment unit; the abnormal type judgment unit is used to analyze the causes of abnormal operation data, divide them into three abnormal types: self-environment causes, system abnormal causes and comprehensive energy operation management safety causes, and write them into the abnormal attributes corresponding to the abnormal operation data; A first calculation unit; the first calculation unit is used to discretize the abnormal operation data and calculate the information gain rate of the abnormal attribute; A decision tree generation unit; the decision tree generation unit is used to select the abnormal attribute with the largest information gain rate as the basis for dividing the root node; use the dichotomy method to divide all the abnormal attributes in the abnormal operation data into corresponding root nodes to form an abnormal state decision tree; Parsing unit: The parsing unit is used to parse the abnormal state decision tree to obtain the terminal device, abnormal type, and abnormal details corresponding to the current abnormal operation data.

9. An electronic device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the abnormal operation fault diagnosis method of the photovoltaic inverter and the charging pile according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is executed on the electronic device, the electronic device executes the abnormal operation fault diagnosis method for the photovoltaic inverter and the charging pile according to any one of claims 1 to 6.

Images