Remote fault diagnosis method for photovoltaic power generation equipment

Abstract

The invention provides a remote fault diagnosis method for photovoltaic power generation equipment, and relates to the technical field of data processing, and the method comprises the steps: carrying out the clustering analysis of distributed photovoltaic equipment according to an environment change feature set, an equipment model set and a performance parameter set, and determining a plurality of equipment sets; performing abnormal power generation identification based on the plurality of power generation capacity sets, and determining an abnormal equipment set; and matching and determining the associated fault feature library and the associated fault prediction model according to the equipment model, the performance parameters and the environment monitoring data of the abnormal equipment, performing fault prediction, and determining a predicted fault type. According to the invention, the technical problem that the operation efficiency, stability and safety of a conventional distributed photovoltaic power generation system are poor due to the fact that the number of photovoltaic power generation devices is large and the photovoltaic power generation devices are dispersed, and the abnormal working state of a single device cannot be timely and accurately found can be solved; the convenience and the accuracy of identifying the abnormal working state of the equipment can be improved, and the intelligence and the automation of equipment fault diagnosis are realized.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a remote fault diagnosis method for photovoltaic power generation equipment. Background Art

[0002] Distributed photovoltaic power generation system is a power generation system that installs photovoltaic power generation equipment in a dispersed manner at the user end, such as on the roof or near residential, commercial buildings or industrial facilities, to achieve effective use of solar energy. This system is different from traditional centralized photovoltaic power stations. It does not rely on large-scale solar power stations, but installs photovoltaic modules near users to directly provide electricity to users or the power grid.

[0003] At present, the existing regional distributed photovoltaic power generation system is unable to timely and accurately detect the abnormal working status of a single device in the area due to the large number and dispersion of photovoltaic power generation equipment. Therefore, it is impossible to take timely measures to diagnose and maintain equipment faults, resulting in technical problems such as poor operating efficiency, stability and safety of the photovoltaic power generation system. Summary of the invention

[0004] The purpose of this application is to provide a remote fault diagnosis method for photovoltaic power generation equipment to solve the technical problem that the existing regional distributed photovoltaic power generation system cannot timely and accurately detect the abnormal working status of a single device in the area due to the large number and dispersion of photovoltaic power generation equipment, and therefore cannot take timely measures to diagnose and maintain equipment faults, resulting in poor operating efficiency, stability and safety of the photovoltaic power generation system.

[0005] In view of the above problems, the present application provides a remote fault diagnosis method for photovoltaic power generation equipment, wherein the method comprises: collecting equipment attribute information of distributed photovoltaic equipment in a target area, obtaining an equipment attribute set, wherein the equipment attribute set comprises an equipment model set, a performance parameter set, an installed capacity set and a location coordinate set; according to predetermined power generation influencing factors and the location coordinate set, calling a regional environmental monitoring log to perform environmental change analysis, performing cluster analysis on distributed photovoltaic equipment according to the environmental change feature set, the equipment model set and the performance parameter set, and determining multiple equipment sets; under a preset monitoring node, obtaining multiple power generation sets of multiple equipment sets, performing abnormal power generation identification based on the multiple power generation sets and the installed capacity set, and determining abnormal equipment set; select a first abnormal device from the abnormal device set, activate the associated monitoring array of the first abnormal device to perform device status monitoring and environmental monitoring, and obtain status monitoring data and environmental monitoring data, wherein the first abnormal device is any abnormal device in the abnormal device set; determine an associated fault feature library and an associated fault prediction model according to the device model, performance parameters and the environmental monitoring data of the first abnormal device, input the status monitoring data into the associated fault feature library and the associated fault prediction model respectively, and determine the predicted fault type based on the fusion of the output results; generate a first fault diagnosis result according to the predicted fault type and the position coordinates of the first abnormal device, and generate an equipment fault diagnosis report for the target area based on the first fault diagnosis result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] By collecting the device attribute information of the distributed photovoltaic equipment in the target area, the device attribute set is obtained, wherein the device attribute set includes a device model set, a performance parameter set, an installed capacity set and a location coordinate set; according to the predetermined power generation influencing factors and the location coordinate set, the regional environmental monitoring log is called to perform environmental change analysis, and the distributed photovoltaic equipment is clustered according to the environmental change feature set, the device model set and the performance parameter set to determine multiple device sets; under the preset monitoring node, multiple power generation sets of the multiple device sets are obtained, and abnormal power generation is identified based on the multiple power generation sets and the installed capacity set to determine the abnormal device set; the first abnormal device is selected from the abnormal device set, Activate the associated monitoring array of the first abnormal device to perform device status monitoring and environmental monitoring to obtain status monitoring data and environmental monitoring data, wherein the first abnormal device is any abnormal device in the abnormal device set; determine the associated fault feature library and the associated fault prediction model according to the device model, performance parameters and the environmental monitoring data of the first abnormal device, input the status monitoring data into the associated fault feature library and the associated fault prediction model respectively, and determine the predicted fault type based on the fusion of the output results; generate a first fault diagnosis result according to the predicted fault type and the location coordinates of the first abnormal device, and generate an equipment fault diagnosis report for the target area based on the first fault diagnosis result. That is to say, by clustering the distributed photovoltaic equipment once according to the environmental change feature set, multiple initial equipment sets are obtained, and then the multiple initial equipment sets are clustered again according to the equipment model set and performance parameter set to determine multiple equipment sets; then, based on the multiple power generation sets of the multiple equipment sets, abnormal power generation is identified to determine the abnormal equipment set, which can timely detect the abnormal working state of a single device and improve the convenience, accuracy and efficiency of abnormal equipment identification; further activate the associated monitoring array of the abnormal equipment to obtain equipment status monitoring data and environmental monitoring data; then, according to the equipment model, performance parameters and environmental monitoring data of the abnormal equipment, the associated fault feature library and the associated fault prediction model are matched to perform fault prediction, and the predicted fault type is determined based on the fusion of the output results; finally, the equipment fault diagnosis report of the target area is generated according to the predicted fault type and the location coordinates of the abnormal equipment. It can improve the convenience and accuracy of the abnormal working state identification of the equipment, and at the same time, it can realize the intelligence and automation of remote fault diagnosis of the equipment, improve the efficiency, accuracy and reliability of fault diagnosis, and achieve the technical effect of ensuring the safe and stable operation of the photovoltaic power generation system and improving the operation efficiency.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A schematic diagram of a flow chart of a remote fault diagnosis method for photovoltaic power generation equipment in this application;

[0011] Figure 2 This is a schematic diagram of a flow chart for determining multiple equipment sets in a remote fault diagnosis method for photovoltaic power generation equipment of the present application. DETAILED DESCRIPTION

[0012] This application provides a remote fault diagnosis method for photovoltaic power generation equipment, which solves the technical problem that the existing regional distributed photovoltaic power generation system cannot timely and accurately detect the abnormal working state of a single device in the region due to the large number and dispersion of photovoltaic power generation equipment, and therefore cannot take timely measures to diagnose and maintain equipment faults, resulting in poor operating efficiency, stability and safety of the photovoltaic power generation system. It can improve the convenience and accuracy of identifying abnormal working states of equipment, and at the same time realize the intelligence and automation of remote fault diagnosis of equipment, improve the efficiency, accuracy and reliability of fault diagnosis, and achieve the technical effect of ensuring the safe and stable operation of the photovoltaic power generation system and improving operating efficiency.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] For examples, please see the attached Figure 1The present application provides a remote fault diagnosis method for photovoltaic power generation equipment, the method specifically comprising the following steps:

[0015] Step 1: Collect the device attribute information of the distributed photovoltaic equipment in the target area and obtain the device attribute set, wherein the device attribute set includes a device model set, a performance parameter set, an installed capacity set and a location coordinate set.

[0016] Specifically, by accessing the database or operation and maintenance management system of the distributed photovoltaic system in the target area, the device attribute information of all photovoltaic devices in the target area is collected, where the device attribute information includes device model, performance parameters, installed capacity and location coordinates, where the device model includes device type, working principle, etc., such as: different working principles of solar panels; performance parameters include output voltage, output current, output power, conversion efficiency, etc. By obtaining the device attribute set, data support is provided for the next step of device clustering analysis.

[0017] Step 2: Based on the predetermined power generation influencing factors and the location coordinate set, call the regional environmental monitoring log to perform environmental change analysis, perform cluster analysis on the distributed photovoltaic equipment based on the environmental change feature set, the equipment model set and the performance parameter set, and determine multiple equipment sets.

[0018] Specifically, the predetermined power generation influencing factors are obtained, wherein the predetermined power generation influencing factors include light intensity, ambient temperature and ambient humidity, which directly affect the power generation efficiency and output power of the photovoltaic equipment. For example, the higher the light intensity, the greater the current and voltage generated by the photovoltaic module, and the higher the power generation efficiency; excessive humidity can cause condensation or corrosion on the surface of the photovoltaic panel, affecting the power generation efficiency. Then, based on the predetermined power generation influencing factors and the position coordinate set, the regional environmental monitoring log is called to perform environmental change analysis, that is, to determine the environmental change characteristics of a single photovoltaic device and obtain an environmental change feature set, wherein the environmental change characteristics correspond to the device location coordinates one by one.

[0019] Then, based on the environmental change feature set, a cluster analysis is performed on the distributed photovoltaic equipment, that is, any two photovoltaic devices whose environmental feature deviations are within the preset deviation threshold are clustered into one category to obtain multiple initial device sets; further, a secondary cluster analysis is performed on the multiple initial device sets based on the device model set and the performance parameter set, that is, the initial devices with the same or similar device models and performance parameters in the initial device set are clustered into one category to obtain multiple device sets, wherein the power generation environment and device information of multiple devices in each device set can be considered the same. By determining multiple device sets through clustering division, a basis is provided for the subsequent identification of abnormal device status, and at the same time, the convenience and accuracy of abnormal device identification in the same device set can be improved, thereby improving the accuracy, reliability and efficiency of abnormal device detection.

[0020] Step three: Under a preset monitoring node, multiple power generation sets of multiple equipment sets are obtained, abnormal power generation is identified based on the multiple power generation sets and the installed capacity set, and an abnormal equipment set is determined.

[0021] Specifically, under the preset monitoring node, the power generation of photovoltaic devices in multiple device sets is collected to obtain multiple power generation sets, where the preset monitoring node is the time node for data collection, which can be set according to the actual situation, for example: a monitoring node is set every 10 minutes. Then, the unit power generation of a single device is calculated based on the multiple power generation sets and the installed capacity set, and the abnormal unit power generation in the same power generation set is identified, where the abnormal unit power generation refers to the unit power generation that deviates greatly from other unit power generation in the same power generation set. Then, the photovoltaic devices corresponding to the abnormal unit power generation are abnormally marked to obtain an abnormal device set. By constructing a device set to identify the abnormal state of the device, since this method is simple to operate, it can improve the convenience and efficiency of identifying the abnormal state of the device.

[0022] Step 4: Select a first abnormal device from the abnormal device set, activate the associated monitoring array of the first abnormal device to perform device status monitoring and environmental monitoring, and obtain status monitoring data and environmental monitoring data, wherein the first abnormal device is any abnormal device in the abnormal device set.

[0023] Specifically, a first abnormal device is then selected from the abnormal device set, wherein the first abnormal device is any abnormal device in the abnormal device set; an associated monitoring array of the first abnormal device is further activated to perform device status monitoring and environmental monitoring. By activating the associated monitoring array for monitoring after the abnormal device is determined, unnecessary monitoring resource consumption can be reduced, and the reasonable utilization rate of monitoring resources and computing power resources can be improved. The associated monitoring array includes an equipment status monitoring array and an environmental parameter monitoring array, and the equipment status monitoring indicators include at least battery panel temperature, voltage, current and output power; the environmental parameter monitoring indicator set includes at least weather type, temperature, humidity, wind speed and wind direction, so as to obtain status monitoring data and environmental monitoring data of the first abnormal device.

[0024] Step 5: Determine the associated fault feature library and the associated fault prediction model based on the equipment model, performance parameters and the environmental monitoring data of the first abnormal equipment, input the state monitoring data into the associated fault feature library and the associated fault prediction model respectively, and determine the predicted fault type based on the output result fusion.

[0025] Specifically, an associated fault feature library and an associated fault prediction model are further determined based on the equipment model, performance parameters and the environmental monitoring data of the first abnormal equipment, wherein the equipment model, performance parameters, the environmental monitoring data and the associated fault feature library and the associated fault prediction model of the first abnormal equipment are in a one-to-one correspondence, the associated fault feature library includes multiple fault feature matrix sets corresponding to multiple fault types, and the associated fault prediction model is constructed based on a BP neural network, and is used to predict faults according to equipment status characteristics.

[0026] Then, the state monitoring data is input into the associated fault feature library and the associated fault prediction model respectively, and the state monitoring data is similarly compared with multiple fault feature matrix sets in the associated fault feature library, and the fault type with a comprehensive similarity greater than a predetermined similarity threshold is selected as the first predicted fault type, and the first predicted fault type set is output; the state monitoring data is intelligently predicted by the associated fault prediction model, and the second predicted fault type set is output; finally, the first predicted fault type set and the second predicted fault type set are intersected, and the intersection operation result is selected as the predicted fault type, that is, the same predicted fault type in the first predicted fault type set and the second predicted fault type set is selected for output to obtain the predicted fault type. By performing the intersection operation on the first predicted fault type set and the second predicted fault type set, and selecting the intersection operation result as the predicted fault type, the advantages of different prediction models can be integrated, the probability of false alarms can be reduced, and the accuracy and reliability of the predicted fault type can be improved.

[0027] Step six: Generate a first fault diagnosis result according to the predicted fault type and the location coordinates of the first abnormal equipment, and generate an equipment fault diagnosis report of the target area based on the first fault diagnosis result.

[0028] Specifically, the predicted fault type and the location coordinates of the first abnormal device are integrated to generate a first fault diagnosis result, and then the same method as that used to obtain the first fault diagnosis result is used to perform intelligent fault analysis on other abnormal devices in the abnormal device set to obtain multiple fault diagnosis results. Finally, a device fault diagnosis report for the target area is generated based on the multiple fault diagnosis results, which can realize the intelligent and automated remote fault diagnosis of the equipment and provide a basis for subsequent equipment maintenance.

[0029] By providing a remote fault diagnosis method for photovoltaic power generation equipment, it is possible to solve the technical problem that the existing regional distributed photovoltaic power generation system cannot detect the abnormal working state of a single device in the area in a timely and accurate manner due to the large number and dispersion of photovoltaic power generation equipment, and therefore cannot take timely measures to diagnose and maintain equipment faults, resulting in poor operating efficiency, stability and safety of the photovoltaic power generation system. First, the device attribute information of the distributed photovoltaic equipment in the target area is collected to obtain a device attribute set, wherein the device attribute set includes a device model set, a performance parameter set, an installed capacity set and a location coordinate set; then, according to the predetermined power generation influencing factors and the location coordinate set, the regional environmental monitoring log is called to perform environmental change analysis, and the distributed photovoltaic equipment is clustered according to the environmental change feature set, the device model set and the performance parameter set to determine multiple device sets; then, under the preset monitoring node, multiple power generation sets of the multiple device sets are obtained, and abnormal power generation is identified based on the multiple power generation sets and the installed capacity set to determine the abnormal device set; next, the first abnormal device is selected from the abnormal device set. The method comprises the following steps: activating an associated monitoring array of a first abnormal device to perform device status monitoring and environmental monitoring to obtain status monitoring data and environmental monitoring data, wherein the first abnormal device is any abnormal device in the abnormal device set; further, determining an associated fault feature library and an associated fault prediction model according to a device model, performance parameters and the environmental monitoring data of the first abnormal device, inputting the status monitoring data into the associated fault feature library and the associated fault prediction model respectively, and determining a predicted fault type based on the fusion of the output results; finally, generating a first fault diagnosis result according to the predicted fault type and the location coordinates of the first abnormal device, and generating an equipment fault diagnosis report for the target area based on the first fault diagnosis result. By clustering distributed photovoltaic devices once according to the environmental change feature set, multiple initial device sets are obtained, and then multiple initial device sets are clustered again according to the device model set and performance parameter set to determine multiple device sets; then abnormal power generation is identified based on multiple power generation sets of multiple device sets to determine abnormal device sets, which can timely detect the abnormal working state of a single device and improve the convenience, accuracy and efficiency of abnormal device identification; further activate the associated monitoring array of abnormal devices to obtain device status monitoring data and environmental monitoring data; then determine the associated fault feature library and associated fault prediction model according to the device model, performance parameters and environmental monitoring data of the abnormal device to predict the fault, and determine the predicted fault type based on the fusion of the output results; finally, generate the device fault diagnosis report of the target area according to the predicted fault type and the location coordinates of the abnormal device. It can improve the convenience and accuracy of the abnormal working state identification of the equipment, and at the same time realize the intelligence and automation of remote fault diagnosis of the equipment, improve the efficiency, accuracy and reliability of fault diagnosis, and achieve the technical effect of ensuring the safe and stable operation of the photovoltaic power generation system and improving the operation efficiency.

[0030] Further, cluster analysis is performed on distributed photovoltaic devices according to the environmental change feature set, the device model set and the performance parameter set, as shown in the attached figure. Figure 2 As shown, step 2 of this application includes:

[0031] According to the predetermined power generation influencing factors and the position coordinate set, information is extracted from the regional environmental monitoring log to obtain an environmental change feature set, wherein the predetermined power generation influencing factors include light intensity, ambient temperature and ambient humidity; based on the predetermined environmental parameter threshold and the environmental change feature set, the distributed photovoltaic equipment is clustered once to obtain multiple initial equipment sets; according to the equipment model set and the performance parameter set, the multiple initial equipment sets are divided twice to determine multiple equipment sets.

[0032] Specifically, the predetermined power generation influencing factors are obtained, and the predetermined power generation influencing factors include light intensity, ambient temperature and ambient humidity. Then, the regional environmental monitoring log is called, where the regional environmental monitoring log is a document or system that records and stores environmental monitoring data in a specific area, and is usually used to track and record environmental factors related to photovoltaic power generation systems, such as light intensity, temperature, humidity, etc. These factors have an important impact on the performance and power generation efficiency of photovoltaic equipment. According to the predetermined power generation influencing factors and the position coordinate set, information is extracted from the regional environmental monitoring log to obtain an environmental change feature set, where the environmental change features characterize the comprehensive power generation environment characteristics of the photovoltaic equipment, and correspond one-to-one with the position coordinates.

[0033] Further, the distributed photovoltaic devices are clustered based on the predetermined environmental parameter threshold and the environmental change feature set, that is, the photovoltaic devices whose feature deviation values ​​of any two environmental change features in the environmental change feature set meet the predetermined environmental parameter threshold are clustered into one category to obtain multiple initial device sets. Through the above steps, photovoltaic devices with similar environmental conditions can be clustered into one category, which is helpful to identify photovoltaic devices with similar environmental conditions. Then, the multiple initial device sets are divided twice according to the device model set and the performance parameter set, that is, the initial devices with the same or similar device models and performance parameters in the same initial device set are added to the same device set to obtain multiple device sets. By clustering the distributed photovoltaic devices according to the environmental change characteristics, device models and performance parameters, the precision and accuracy of the photovoltaic device clustering can be improved, thereby improving the accuracy and reliability of the device set construction, and providing a reliable basis for the subsequent abnormal device identification.

[0034] Further, to obtain an environmental change feature set, the present application further comprises the following steps:

[0035] A first position coordinate is selected from the position coordinate set, and information is extracted from the regional environmental monitoring log based on a predetermined time period, a predetermined power generation influencing factor and the first position coordinate to obtain a first environmental monitoring data set; data preprocessing is performed on the first environmental monitoring data set to obtain a first standard environmental monitoring data set, and a mean calculation is performed on the first standard environmental monitoring data set to obtain a first environmental change feature, and the first environmental change feature is added to the environmental change feature set.

[0036] Specifically, first, a first position coordinate is selected from the position coordinate set, and the first position coordinate is any one of the position coordinate set; then, information is extracted from the regional environmental monitoring log based on a predetermined time period, a predetermined power generation influencing factor and the first position coordinate, wherein the predetermined time period is a time period for data extraction, which can be set according to actual conditions, such as: in the last month, a first environmental monitoring data set is obtained. Then, data preprocessing is performed on the first environmental monitoring data set, including steps such as data cleaning, missing supplementation, data conversion and data integration, such as: checking whether there are repeated, abnormal or erroneous data points in the data set, and deleting or correcting these data points to ensure the integrity and accuracy of the data; if there are missing values ​​in the data set, interpolation (such as linear interpolation, spline interpolation, etc.) or deletion method can be used for processing, and the most appropriate method is selected to process the missing values ​​to maintain the continuity and integrity of the data; the numerical data in the data set is converted into a format suitable for model analysis, such as standardization or normalization. Then, the mean of multiple first standard environmental monitoring data in the first standard environmental monitoring data set is calculated to obtain the first environmental change feature, and finally the first environmental change feature is added to the environmental change feature set.

[0037] Furthermore, based on the predetermined environmental parameter threshold and the environmental change feature set, the distributed photovoltaic devices are clustered once. The present application further includes the following steps:

[0038] The predetermined environmental parameter thresholds include a light intensity threshold, a temperature threshold and a humidity threshold; randomly selecting a first environmental change feature and a second environmental change feature from the environmental change feature set, wherein the first environmental change feature and the second environmental change feature are any two environmental change features in the environmental change feature set; performing feature deviation analysis on the first environmental change feature and the second environmental change feature to obtain an environmental feature deviation value, wherein the environmental feature deviation value includes a light intensity deviation, a temperature deviation and a humidity deviation; when the light intensity deviation satisfies the light intensity threshold and the temperature deviation satisfies the temperature threshold and the humidity deviation satisfies the humidity threshold, the first photovoltaic device and the second photovoltaic device are added to the same initial device set, wherein the first photovoltaic device is a photovoltaic device corresponding to the first environmental change feature, and the second photovoltaic device is a photovoltaic device corresponding to the second environmental change feature; performing iterative analysis until the environmental change feature comparison is completed and multiple initial device sets are obtained.

[0039] Specifically, the predetermined environmental parameter threshold represents the range of change of the environmental parameter, which can be determined based on the analysis of historical environmental data of the target area, including a light intensity threshold, a temperature threshold, and a humidity threshold. Then, the first environmental change feature and the second environmental change feature are randomly selected from the environmental change feature set, wherein the first environmental change feature and the second environmental change feature are any two environmental change features in the environmental change feature set. The feature deviation of the first environmental change feature and the second environmental change feature is further calculated to obtain an environmental feature deviation value, wherein the environmental feature deviation value includes a light intensity deviation, a temperature deviation, and a humidity deviation.

[0040] The environmental characteristic deviation value is judged according to the predetermined environmental parameter threshold, and when the light intensity deviation is less than or equal to the light intensity threshold, the temperature deviation is less than or equal to the temperature threshold, and the humidity deviation is less than or equal to the humidity threshold, the first photovoltaic device and the second photovoltaic device are added to the same initial device set, wherein the first photovoltaic device is the photovoltaic device corresponding to the first environmental change characteristic, and the second photovoltaic device is the photovoltaic device corresponding to the second environmental change characteristic. The same method is used to perform iterative analysis until all environmental change characteristics in the environmental change characteristic set are compared, and multiple initial device sets are obtained.

[0041] Further, abnormal power generation identification is performed based on multiple power generation sets and the installed capacity set. Step three of the present application includes:

[0042] A first power generation set is randomly selected from multiple power generation sets, and a first installed capacity set of the first power generation set is obtained based on the installed capacity set, wherein the first power generation and the first installed capacity correspond one to one; unit power generation is calculated based on the first power generation set and the first installed capacity set to obtain a first unit power generation set, and a mean calculation is performed on the first unit power generation set to obtain a first unit power generation mean; a deviation calculation is performed on the first unit power generation set based on the first unit power generation mean, and a photovoltaic device corresponding to a unit power generation deviation greater than a predetermined power generation deviation threshold is set as an abnormal device, and added to the abnormal device set.

[0043] Specifically, first, a first power generation set is randomly selected from multiple power generation sets, wherein the first power generation set is any one of the multiple power generation sets; then, the first power generation set is matched in the installed capacity set based on the first power generation set to obtain a first installed capacity set of the first power generation set, wherein the first power generation and the first installed capacity correspond one to one. Then, unit power generation is calculated based on the first power generation set and the first installed capacity set, wherein the unit power generation is the ratio of power generation to installed capacity, and the first unit power generation set is obtained. Then, the average of multiple first unit power generation in the first unit power generation set is calculated to obtain the average of the first unit power generation.

[0044] According to the first unit power generation mean, deviations of multiple first unit power generation in the first unit power generation set are calculated to obtain multiple unit power generation deviations, wherein the unit power generation deviation is the difference between the first unit power generation mean and the first unit power generation, and then the photovoltaic device corresponding to the unit power generation deviation greater than the predetermined power generation deviation threshold is set as an abnormal device, wherein the predetermined power generation deviation threshold can be set according to the actual situation, and multiple abnormal devices are obtained to construct an abnormal device set.

[0045] Further, step 4 of this application includes:

[0046] Acquire an equipment status monitoring indicator set and an environmental monitoring indicator set, wherein the equipment status monitoring indicator set includes at least the temperature, voltage, current and output power of the battery panel, and the environmental monitoring indicator set includes at least the weather type, temperature, humidity, wind speed and wind direction; based on the equipment status monitoring indicator set, sequentially perform correlation analysis on the equipment status monitoring indicators and equipment failures, select equipment status monitoring indicators with correlation coefficients greater than a first predetermined coefficient threshold as high-frequency status monitoring indicators, and obtain a high-frequency status monitoring indicator set; based on the environmental monitoring indicator set, sequentially perform correlation analysis on the environmental monitoring indicators and equipment failures, select environmental monitoring indicators with correlation coefficients greater than a second predetermined coefficient threshold as high-frequency environmental monitoring indicators, and obtain a high-frequency environmental monitoring indicator set; and build a correlation monitoring array based on the high-frequency status monitoring indicator set and the high-frequency environmental monitoring indicator set.

[0047] Specifically, the equipment state monitoring indicator set and the environmental monitoring indicator set are obtained, wherein the equipment state monitoring indicator set includes at least the temperature, voltage, current and output power of the battery panel, and the environmental monitoring indicator set includes at least the weather type, temperature, humidity, wind speed and wind direction, and additional monitoring indicators can be added according to the actual situation. Then, based on the equipment state monitoring indicator set, the equipment state monitoring indicator and the equipment failure are analyzed in sequence. For example, multiple sample equipment failure events can be obtained, and then the occurrence frequency and influence of the equipment state monitoring indicator in the multiple sample equipment failure events are counted. The higher the occurrence frequency and the greater the influence, the greater the correlation between the equipment state monitoring indicator and the equipment failure, and the greater the correlation coefficient, and multiple correlation coefficients of multiple equipment state monitoring indicators are obtained. Then, the equipment state monitoring indicator with a correlation coefficient greater than the first predetermined coefficient threshold is selected and set as a high-frequency state monitoring indicator, wherein the first predetermined coefficient threshold is set based on the actual situation, and a high-frequency state monitoring indicator set is obtained. Based on the environmental monitoring indicator set, the same method is used to analyze the correlation between the environmental monitoring indicator and the equipment failure in sequence, and the environmental monitoring indicator with a correlation coefficient greater than the second predetermined coefficient threshold is selected and set as a high-frequency environmental monitoring indicator, and a high-frequency environmental monitoring indicator set is obtained.

[0048] Finally, an associated monitoring array is designed according to the high-frequency state monitoring indicator set and the high-frequency environmental monitoring indicator set, wherein the associated monitoring array should include all sensors and monitoring equipment required for the high-frequency state monitoring indicators and environmental monitoring indicators; then the sensors and monitoring equipment required for the associated monitoring array are installed in the distributed photovoltaic system, while ensuring that the sensors and monitoring equipment can accurately collect high-frequency state monitoring indicators and environmental monitoring indicators.

[0049] Further, according to the equipment model, performance parameters and the environmental monitoring data of the first abnormal equipment, the associated fault feature library and the associated fault prediction model are determined, and step five of the present application includes:

[0050] Acquire multiple high-frequency environmental parameter thresholds, divide the multiple high-frequency environmental parameter thresholds according to multiple preset step sizes, and determine multiple high-frequency environmental parameter interval sets; randomly select a parameter from the multiple high-frequency environmental parameter interval sets for combination to obtain multiple sample environmental parameter groups, and select a first sample environmental parameter group from the multiple sample environmental parameter groups; acquire a first device model and a first performance parameter, and construct a first associated fault feature library and a first associated fault prediction model based on the first device model, the first performance parameter, and the first sample environmental parameter group; based on the classification decision principle, construct a fault prediction matching channel according to the mapping relationship between the first device model, the first performance parameter, the first sample environmental parameter group, the first associated fault feature library, and the first associated fault prediction model; input the device model, performance parameters, and the environmental monitoring data of the first abnormal device into the fault prediction matching channel for matching, and determine the associated fault feature library and the associated fault prediction model.

[0051] Specifically, multiple high-frequency environmental parameter thresholds are obtained, and then the multiple high-frequency environmental parameter thresholds are divided according to multiple preset step sizes, where the preset step size can be set according to the environmental parameter type and fluctuation amplitude, for example: the temperature step size is set to 0.1 degrees Celsius, and multiple high-frequency environmental parameter interval sets are determined. Then, a parameter interval is randomly selected from the multiple high-frequency environmental parameter interval sets for combination to obtain multiple sample environmental parameter groups, and then a first sample environmental parameter group is selected from the multiple sample environmental parameter groups, where the first sample environmental parameter group is any one of the multiple sample environmental parameter groups.

[0052] Then, the first device model and the first performance parameter are obtained, wherein the first device model is any one of the device models of multiple photovoltaic devices in the area, and the first performance parameter is any one of the device performance parameters of multiple photovoltaic devices in the area. Then, based on the first device model, the first performance parameter and the first sample environmental parameter group, a first associated fault feature library and a first associated fault prediction model are constructed. Further, a mapping relationship between the first device model, the first performance parameter, the first sample environmental parameter group and the first associated fault feature library and the first associated fault prediction model is established, and based on the classification decision principle, a fault prediction matching channel is constructed according to the mapping relationship, that is, the first device model, the first performance parameter and the first sample environmental parameter group are used as child nodes, and the first associated fault feature library and the first associated fault prediction model with the mapping relationship are used as leaf nodes of the child nodes to construct a fault prediction matching channel. Finally, the device model, performance parameters and the environmental monitoring data of the first abnormal device are input into the fault prediction matching channel for matching, and the associated fault feature library and the associated fault prediction model are determined.

[0053] Further, based on the first device model, the first performance parameter and the first sample environment parameter group, a first associated fault feature library is constructed. The present application also includes the following steps:

[0054] Taking the first device model, the first performance parameter and the first sample environmental parameter group as constraints, data retrieval is performed online to obtain a first sample data set, wherein the first sample data set includes a first sample fault type set and multiple first sample fault feature sets, and the first sample fault feature set and the first sample fault type are in a one-to-one relationship; similarity dimensionality reduction is performed on the multiple first sample fault feature sets based on a predetermined similarity dimensionality reduction algorithm to obtain multiple first standard sample fault feature sets, wherein the minimum similarity of the comparison of the first standard sample fault features is less than a first similarity threshold and the mean of the comparison similarity is less than a second similarity threshold, and the first similarity threshold is less than the second similarity threshold; multiple first standard sample fault feature matrix sets are generated according to the multiple first standard sample fault feature sets, and a first associated fault feature library is constructed based on the first sample fault type set and the multiple first standard sample fault feature matrix sets.

[0055] Specifically, the first device model, the first performance parameter and the first sample environmental parameter group are used as retrieval constraints, and data retrieval is performed based on big data technology to obtain a first sample data set, wherein the first sample data set includes a first sample fault type set and multiple first sample fault feature sets, and the first sample fault feature set and the first sample fault type are in a one-to-one relationship.

[0056] Then, based on a predetermined similarity dimensionality reduction algorithm, a sample fault feature set is randomly selected from multiple first sample fault feature sets, and a sample fault feature is randomly selected from the sample fault feature set, and then a similarity comparison is performed on the sample fault feature and other sample fault features in the sample fault feature set, for example: a similarity comparison can be performed through an existing cosine similarity comparison algorithm to obtain multiple comparison similarities, and the mean of the multiple comparison similarities is calculated to obtain the comparison similarity mean; then a first similarity threshold and a second similarity threshold are set, wherein the first similarity threshold is less than the second similarity threshold; when the minimum similarity among the multiple comparison similarities is less than the first similarity threshold and the comparison similarity mean is less than the second similarity threshold, it indicates that the overall deviation between the sample fault feature and other sample fault features in the same sample fault feature set is large, and the sample fault feature is set as a standard sample fault feature to obtain multiple first standard sample fault feature sets.

[0057] Finally, multiple first standard sample fault feature matrix sets are generated according to multiple first standard sample fault feature sets, and based on the mapping relationship between the first sample fault type and the first standard sample fault feature matrix set, a first associated fault feature library is constructed according to the first sample fault type set and multiple first standard sample fault feature matrix sets.

[0058] Further, based on the first device model, the first performance parameter and the first sample environmental parameter group, a first associated fault prediction model is constructed. The present application also includes the following steps:

[0059] Taking the first equipment model, the first performance parameter and the first sample environmental parameter group as constraints, data retrieval is performed online to obtain a second sample data set, wherein the second sample data set includes multiple second sample fault type sets and multiple second sample fault feature sets, and the second sample fault features and the second sample fault types are in a many-to-many relationship; the second sample data set is divided into Q equal parts, and Q parts are selected with replacement to construct a first sample set, and Q sample sets are selected iteratively to generate Q sample sets; taking the second sample fault feature set as input and the second sample fault type set as supervision, supervised training and cross-validation are performed on the BP neural network through the Q sample sets to obtain Q converged fault prediction units, and a first associated fault prediction model is built based on the Q converged fault prediction units, wherein the output of the first associated fault prediction model is the mode of the output results of the Q converged fault prediction units.

[0060] Specifically, with the first device model, the first performance parameter and the first sample environmental parameter group as constraints, data retrieval is performed based on the big data technology network to obtain the second sample data set, wherein the second sample data set includes multiple second sample fault type sets and multiple second sample fault feature sets, and the second sample fault feature and the second sample fault type are in a many-to-many relationship. Then the second sample data set is divided into Q parts, and Q parts of the data set are selected with replacement Q times to construct the first sample set; the above selection process with replacement is repeated, and iterative selection is performed Q times. After each iteration, a new sample set is generated, and Q sample sets are generated.

[0061] Then, an initial fault prediction unit is constructed based on the BP neural network. The initial fault prediction unit is a neural network model that can be iteratively optimized in machine learning. The input data of the initial fault prediction unit is multiple fault features, and the output data is multiple fault types. Then, the second sample fault feature set is used as input, and the second sample fault type set is used as supervision. The BP neural network is supervised and cross-validated through Q sample sets, for example: supervised training is performed using the gradient descent method and the loss function, and after meeting the preset number of training times, the training sets are exchanged for cross-validation until the output accuracy is verified to meet the expected indicators, and Q converged fault prediction units are obtained. Finally, a first associated fault prediction model is constructed based on the Q converged fault prediction units, wherein the output of the first associated fault prediction model is the mode of the output results of the Q converged fault prediction units. By constructing the first associated fault prediction model based on Q converged fault prediction units, the prediction error can be reduced and the accuracy and reliability of fault prediction can be improved.

[0062] Further, the state monitoring data is respectively input into the associated fault feature library and the associated fault prediction model, and the predicted fault type is determined based on the output result fusion. Step five of this application includes:

[0063] The state monitoring data is compared with multiple sample fault feature matrix sets in an associated fault feature library through similarity traversal, multiple similarity sets are output, and the mean of the multiple similarity sets is calculated, and the fault type with a similarity mean greater than a predetermined abnormal similarity threshold is selected as the predicted fault type, and a first predicted fault type set is output; the state monitoring data is input into an associated fault prediction model, and a second predicted fault type set is output; an intersection operation is performed on the first predicted fault type set and the second predicted fault type set to obtain the predicted fault type.

[0064] Specifically, the state monitoring data is traversed and compared with multiple sample fault feature matrix sets in the associated fault feature library for similarity, wherein commonly used similarity comparison algorithms include cosine similarity algorithm, Euclidean distance, etc., and an adapted similarity comparison algorithm can be selected according to actual conditions to output multiple similarity sets; then, the multiple similarity sets are averaged, and the fault types whose similarity mean is greater than the predetermined abnormal similarity threshold are selected as predicted fault types, and the first predicted fault type set is output. On the other hand, the state monitoring data is input into the associated fault prediction model, and the second predicted fault type set is output. Finally, the first predicted fault type set and the second predicted fault type set are intersected, that is, the same predicted fault types in the first predicted fault type set and the second predicted fault type set are selected for output to obtain predicted fault types.

[0065] In summary, the remote fault diagnosis method for photovoltaic power generation equipment provided by the present application has the following technical effects:

[0066] 1. By clustering distributed photovoltaic devices once according to the environmental change feature set, multiple initial device sets are obtained, and then multiple initial device sets are clustered again according to the device model set and performance parameter set to determine multiple device sets; then abnormal power generation is identified based on multiple power generation sets of multiple device sets to determine the abnormal device set, which can timely detect the abnormal working state of a single device and improve the convenience, accuracy and efficiency of abnormal device identification; further activate the associated monitoring array of the abnormal device to obtain device status monitoring data and environmental monitoring data; then determine the associated fault feature library and associated fault prediction model according to the device model, performance parameters and environmental monitoring data of the abnormal device to predict the fault, and determine the predicted fault type based on the fusion of the output results; finally, generate the device fault diagnosis report of the target area according to the predicted fault type and the location coordinates of the abnormal device. It can improve the convenience and accuracy of the identification of abnormal working state of the equipment, and at the same time realize the intelligence and automation of remote fault diagnosis of the equipment, improve the efficiency, accuracy and reliability of fault diagnosis, and achieve the technical effect of ensuring the safe and stable operation of the photovoltaic power generation system and improving the operation efficiency.

[0067] 2. By clustering and dividing, multiple equipment sets are determined, which provides a basis for the subsequent identification of abnormal equipment status. At the same time, it can improve the convenience and accuracy of abnormal equipment identification in the same equipment set, thereby improving the accuracy, reliability and efficiency of abnormal equipment detection.

[0068] 3. By activating the associated monitoring array for monitoring after determining the abnormal device, unnecessary monitoring resource consumption can be reduced and the rational utilization of monitoring resources and computing resources can be improved.

[0069] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A remote fault diagnosis method for photovoltaic power generation equipment, characterized in that: The method comprises: Collecting device attribute information of distributed photovoltaic devices in the target area to obtain a device attribute set, wherein the device attribute set includes a device model set, a performance parameter set, an installed capacity set, and a location coordinate set; According to the predetermined power generation influencing factors and the location coordinate set, the regional environmental monitoring log is called to perform environmental change analysis, and the distributed photovoltaic equipment is clustered according to the environmental change feature set, the equipment model set and the performance parameter set to determine multiple equipment sets; Under a preset monitoring node, multiple power generation sets of multiple equipment sets are obtained, abnormal power generation is identified based on the multiple power generation sets and the installed capacity set, and an abnormal equipment set is determined; Selecting a first abnormal device from the abnormal device set, activating an associated monitoring array of the first abnormal device to perform device state monitoring and environment monitoring, and obtaining state monitoring data and environment monitoring data, wherein the first abnormal device is any abnormal device in the abnormal device set; Determine an associated fault feature library and an associated fault prediction model according to the equipment model, performance parameters and the environmental monitoring data of the first abnormal equipment, input the state monitoring data into the associated fault feature library and the associated fault prediction model respectively, and determine the predicted fault type based on the fusion of the output results; A first fault diagnosis result is generated according to the predicted fault type and the location coordinates of the first abnormal equipment, and an equipment fault diagnosis report of the target area is generated based on the first fault diagnosis result.

2. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 1, characterized in that: Cluster analysis is performed on distributed photovoltaic devices according to the environmental change feature set, the device model set, and the performance parameter set, including: Extracting information from the regional environment monitoring log according to the predetermined power generation influencing factors and the position coordinate set to obtain an environmental change feature set, wherein the predetermined power generation influencing factors include light intensity, ambient temperature and ambient humidity; Clustering the distributed photovoltaic devices based on a predetermined environmental parameter threshold and the environmental change feature set to obtain a plurality of initial device sets; The multiple initial device sets are divided twice according to the device model set and the performance parameter set to determine multiple device sets.

3. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 2, characterized in that: Get the environmental change feature set, including: Selecting a first position coordinate from the position coordinate set, extracting information from the regional environment monitoring log based on a predetermined time period, a predetermined power generation influencing factor, and the first position coordinate to obtain a first environment monitoring data set; The first environmental monitoring data set is subjected to data preprocessing to obtain a first standard environmental monitoring data set, and the first standard environmental monitoring data set is subjected to mean calculation to obtain a first environmental change feature, and the first environmental change feature is added to the environmental change feature set.

4. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 2, characterized in that: The distributed photovoltaic devices are clustered based on a predetermined environmental parameter threshold and the environmental change feature set, including: The predetermined environmental parameter thresholds include a light intensity threshold, a temperature threshold, and a humidity threshold; Randomly selecting a first environment change feature and a second environment change feature from the environment change feature set, wherein the first environment change feature and the second environment change feature are any two environment change features in the environment change feature set; Performing feature deviation analysis on the first environmental change feature and the second environmental change feature to obtain an environmental feature deviation value, wherein the environmental feature deviation value includes a light intensity deviation, a temperature deviation, and a humidity deviation; When the light intensity deviation satisfies the light intensity threshold, the temperature deviation satisfies the temperature threshold, and the humidity deviation satisfies the humidity threshold, the first photovoltaic device and the second photovoltaic device are added to the same initial device set, wherein the first photovoltaic device is a photovoltaic device corresponding to the first environmental change characteristic, and the second photovoltaic device is a photovoltaic device corresponding to the second environmental change characteristic; Perform iterative analysis until the environmental change characteristics are compared and multiple initial device sets are obtained.

5. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 1, characterized in that: Abnormal power generation identification is performed based on multiple power generation sets and the installed capacity set, including: Randomly selecting a first power generation set from multiple power generation sets, and acquiring a first installed capacity set of the first power generation set based on the installed capacity set, wherein the first power generation and the first installed capacity correspond one to one; Calculating unit power generation based on the first power generation set and the first installed capacity set to obtain a first unit power generation set, and performing mean calculation on the first unit power generation set to obtain a first unit power generation mean; The first unit power generation set is subjected to deviation calculation based on the first unit power generation mean value, and the photovoltaic devices corresponding to the unit power generation deviations greater than the predetermined power generation deviation threshold are set as abnormal devices and added to the abnormal device set.

6. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 1, characterized in that: Activating the associated monitoring array of the first abnormal device to perform device status monitoring and environment monitoring, which also includes: Obtaining a device status monitoring indicator set and an environmental monitoring indicator set, wherein the device status monitoring indicator set includes at least the temperature, voltage, current and output power of the battery panel, and the environmental monitoring indicator set includes at least the weather type, temperature, humidity, wind speed and wind direction; Based on the equipment status monitoring indicator set, correlation analysis is performed on the equipment status monitoring indicators and equipment failures in turn, and equipment status monitoring indicators with correlation coefficients greater than a first predetermined coefficient threshold are selected as high-frequency status monitoring indicators to obtain a high-frequency status monitoring indicator set; Based on the environmental monitoring indicator set, correlation analysis is performed on the environmental monitoring indicators and equipment failures in turn, and environmental monitoring indicators with correlation coefficients greater than a second predetermined coefficient threshold are selected as high-frequency environmental monitoring indicators to obtain a high-frequency environmental monitoring indicator set; An associated monitoring array is established according to the high-frequency state monitoring indicator set and the high-frequency environment monitoring indicator set.

7. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 6, characterized in that: According to the equipment model, performance parameters and the environmental monitoring data of the first abnormal equipment, a related fault feature library and a related fault prediction model are determined, including: Acquire multiple high-frequency environmental parameter thresholds, divide the multiple high-frequency environmental parameter thresholds according to multiple preset step sizes, and determine multiple high-frequency environmental parameter interval sets; Randomly selecting a parameter from a plurality of high-frequency environmental parameter intervals to combine, to obtain a plurality of sample environmental parameter groups, and selecting a first sample environmental parameter group from the plurality of sample environmental parameter groups; Acquire a first device model and a first performance parameter, and construct a first associated fault feature library and a first associated fault prediction model based on the first device model, the first performance parameter, and a first sample environmental parameter group; Based on the classification decision principle, a fault prediction matching channel is constructed according to the mapping relationship between the first device model, the first performance parameter, the first sample environmental parameter group, the first associated fault feature library, and the first associated fault prediction model; The device model, performance parameters and environmental monitoring data of the first abnormal device are input into the fault prediction matching channel for matching, and an associated fault feature library and an associated fault prediction model are determined.

8. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 7, characterized in that: Building a first associated fault feature library based on the first device model, the first performance parameter and the first sample environment parameter group includes: Taking the first device model, the first performance parameter and the first sample environmental parameter group as constraints, performing data retrieval online to obtain a first sample data set, wherein the first sample data set includes a first sample fault type set and a plurality of first sample fault feature sets, and the first sample fault feature set and the first sample fault type are in a one-to-one relationship; Performing similarity dimensionality reduction on the plurality of first sample fault feature sets based on a predetermined similarity dimensionality reduction algorithm to obtain a plurality of first standard sample fault feature sets, wherein the minimum similarity of the first standard sample fault features is less than a first similarity threshold and the average of the similarity is less than a second similarity threshold, and the first similarity threshold is less than the second similarity threshold; A plurality of first standard sample fault feature matrix sets are generated according to the plurality of first standard sample fault feature sets, and a first associated fault feature library is constructed based on the first sample fault type set and the plurality of first standard sample fault feature matrix sets.

9. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 7, characterized in that: Building a first associated fault prediction model based on the first device model, the first performance parameter and the first sample environmental parameter group includes: Taking the first device model, the first performance parameter and the first sample environmental parameter group as constraints, performing data retrieval online to obtain a second sample data set, wherein the second sample data set includes a plurality of second sample fault type sets and a plurality of second sample fault feature sets, and the second sample fault features and the second sample fault types are in a many-to-many relationship; The second sample data set is divided into Q equal parts, and Q parts are selected with replacement to construct the first sample set, and the selection is iterated Q times to generate Q sample sets; Taking the second sample fault feature set as input and the second sample fault type set as supervision, the BP neural network is supervised trained and cross-validated through Q sample sets to obtain Q convergence fault prediction units, and a first associated fault prediction model is built based on the Q convergence fault prediction units, wherein the output of the first associated fault prediction model is the mode of the output results of the Q convergence fault prediction units.

10. A remote fault diagnosis method for photovoltaic power generation equipment according to claim 8, characterized in that: The condition monitoring data is respectively input into the associated fault feature library and the associated fault prediction model, and the predicted fault type is determined based on the output result fusion, including: Performing a similarity traversal comparison between the state monitoring data and a plurality of sample fault feature matrix sets in an associated fault feature library, outputting a plurality of similarity sets, and calculating the mean of the plurality of similarity sets, selecting a fault type whose similarity mean is greater than a predetermined abnormal similarity threshold value as a predicted fault type, and outputting a first predicted fault type set; Inputting the condition monitoring data into an associated fault prediction model and outputting a second predicted fault type set; An intersection operation is performed on the first predicted fault type set and the second predicted fault type set to obtain the predicted fault type.

Images