Monitoring method, device, electronic equipment, storage medium and program product of photovoltaic power generation equipment

Abstract

The present disclosure relates to a monitoring method, device, electronic device, storage medium and program product for photovoltaic power generation equipment; the monitoring method for photovoltaic power generation equipment includes: obtaining string current data corresponding to the photovoltaic power generation equipment, the string current data including multiple string current values; extracting features from the string current data to obtain target string current features; determining the monitoring results of the photovoltaic power generation equipment according to the target string current features and a pre-trained isolation forest model, wherein the monitoring results of the photovoltaic power generation equipment are used to characterize the abnormal conditions of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output the abnormal evaluation value of the photovoltaic power generation equipment according to the input target string current features. This technical solution can realize efficient and accurate analysis of abnormal conditions of photovoltaic power generation equipment and improve the operational stability and safety of photovoltaic power generation equipment.

Description

Technical Field

[0001] The present disclosure relates to the field of renewable energy technology, and in particular to a monitoring method, device, electronic device, storage medium and program product for photovoltaic power generation equipment. Background Art

[0002] With the development of renewable energy technology, photovoltaic power generation technology has also developed. Photovoltaic power generation technology realizes photovoltaic power generation through photovoltaic power generation equipment.

[0003] During the power generation process of photovoltaic power generation equipment, there may be multiple abnormal factors such as misconnection of photovoltaic equipment, equipment interruption, voltage abnormality, etc., which may lead to abnormalities in photovoltaic power generation equipment. Summary of the invention

[0004] The purpose of the present disclosure is to provide a monitoring method, device, electronic device, storage medium and program product for photovoltaic power generation equipment. The monitoring method, device, electronic device, storage medium and program product for photovoltaic power generation equipment can realize efficient and accurate analysis of abnormal conditions of photovoltaic power generation equipment and improve the operating stability and safety of photovoltaic power generation equipment.

[0005] In order to achieve the above-mentioned purpose, in a first aspect, the present disclosure provides a monitoring method for photovoltaic power generation equipment, comprising: obtaining string current data corresponding to the photovoltaic power generation equipment, the string current data comprising multiple string current values; performing feature extraction on the string current data to obtain target string current features; determining a monitoring result of the photovoltaic power generation equipment according to the target string current features and a pre-trained isolation forest model, wherein the monitoring result of the photovoltaic power generation equipment is used to characterize an abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output an abnormal evaluation value of the photovoltaic power generation equipment according to the input target string current features.

[0006] Optionally, obtaining the string current corresponding to the photovoltaic power generation equipment includes: obtaining multiple original string current values ​​of the photovoltaic power generation equipment within a preset time length; aggregating the multiple original string current values ​​to obtain the multiple string current values, and the number of the multiple string current values ​​is less than the number of the multiple original string current values.

[0007] Optionally, the aggregating the multiple original string current values ​​to obtain the multiple string current values ​​includes: dividing the multiple original string current values ​​according to a preset time window to obtain multiple string current sets, each string current set including at least one original string current value, and the length of the preset time window is less than the preset time length; determining a target original string current value corresponding to each string current set based on the original string current value included in each of the string current sets; and determining the multiple string current values ​​based on the multiple target original string current values.

[0008] Optionally, determining the multiple string current values ​​based on the multiple target original string current values ​​includes: if the multiple target original string current values ​​do not include null values, determining the multiple target original string current values ​​as the multiple string current values; if the multiple target original string current values ​​include null values, performing null value interpolation processing on the multiple target original string current values, and determining the multiple string current values ​​based on the multiple target original string current values ​​after the interpolation processing.

[0009] Optionally, the feature extraction of the string current to obtain the target string current feature includes: determining the standard deviations corresponding to the multiple string current values; determining a target string current value with a standard deviation abnormality from the multiple string current values ​​according to the standard deviations corresponding to the multiple string current values; and determining the target string current value as the target string current feature.

[0010] Optionally, the monitoring method also includes: obtaining a training data set, the training data set including multiple training samples, each training sample including a sample group string current characteristic; performing unsupervised training on the isolation forest model to be trained according to the training data set to obtain the pre-trained isolation forest model, wherein the isolation forest model to be trained is unsupervisedly trained by constructing an isolated tree, and the depth of the isolated tree is less than a preset depth.

[0011] Optionally, the photovoltaic power generation equipment is any photovoltaic power generation equipment in a distributed photovoltaic power generation system, and the monitoring result of the photovoltaic power generation equipment is determined according to the target group string current characteristics and the pre-trained isolation forest model, including: inputting the target group string current characteristics into the pre-trained isolation forest model to obtain the abnormal assessment value of the photovoltaic power generation equipment output by the pre-trained isolation forest model; obtaining the abnormal assessment values ​​of other photovoltaic power generation equipment in the distributed photovoltaic power generation system; and determining the monitoring result of the photovoltaic power generation equipment based on at least the abnormal assessment value of the photovoltaic power generation equipment and the abnormal assessment values ​​of other photovoltaic power generation equipment.

[0012] Optionally, if the photovoltaic power generation equipment monitoring result indicates that the photovoltaic power generation equipment is abnormal, a maintenance suggestion for the photovoltaic power generation equipment is determined based on the abnormal evaluation value; the maintenance suggestion for the photovoltaic power generation equipment is fed back to the maintenance end so that the maintenance personnel at the maintenance end can perform maintenance on the photovoltaic power generation equipment.

[0013] In a second aspect, the present disclosure provides a monitoring device for photovoltaic power generation equipment, including: an acquisition module, configured to acquire string current data corresponding to the photovoltaic power generation equipment, the string current data including multiple string current values; a feature extraction module, configured to perform feature extraction on the string current data to obtain target string current features; a monitoring module, configured to determine a monitoring result of the photovoltaic power generation equipment based on the target string current features and a pre-trained isolation forest model, wherein the monitoring result of the photovoltaic power generation equipment is used to characterize an abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output an abnormal evaluation value of the photovoltaic power generation equipment based on the input target string current features.

[0014] Optionally, the acquisition module is further configured to: acquire multiple original string current values ​​of the photovoltaic power generation equipment within a first preset time length; aggregate the multiple original string current values ​​to obtain the multiple string current values ​​within a second preset time length, and the second preset time length is shorter than the first preset time length.

[0015] Optionally, the acquisition module is further configured to: divide the multiple original string current values ​​according to preset time slices to obtain multiple string current sets, each string current set including at least one original string current value; determine the target original string current values ​​corresponding to the multiple string current sets respectively; and determine the multiple string current values ​​based on the multiple target original string current values.

[0016] In a third aspect, the present disclosure provides a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for monitoring a photovoltaic power generation device as described in the first aspect.

[0017] In a fourth aspect, the present disclosure provides an electronic device, comprising: a memory on which a computer program is stored; and a processor for executing the computer program in the memory to implement the monitoring method of the photovoltaic power generation equipment as described in the first aspect.

[0018] In a fifth aspect, the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the monitoring method for the photovoltaic power generation equipment as described in the first aspect.

[0019] Through the above technical solution, the string current data of the photovoltaic power generation equipment is feature extracted to obtain the target string current characteristics, and the monitoring results of the photovoltaic power generation equipment are determined by using the pre-trained isolation forest model and the target string current characteristics. Among them, the string current of the photovoltaic power generation equipment is related to the current of each component of the photovoltaic power generation equipment, which can be used to analyze the abnormal situation of the photovoltaic power generation equipment; the pre-trained isolation forest model can ensure the efficiency and accuracy of the abnormal situation analysis. Therefore, by using the pre-trained isolation forest model and string current characteristics, it is possible to achieve efficient and accurate analysis of the abnormal situation of the photovoltaic power generation equipment, thereby improving the operating stability and safety of the photovoltaic power generation equipment.

[0020] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings: Figure 1 The present invention is a flow chart of a method for monitoring a photovoltaic power generation device according to an exemplary embodiment.

[0022] Figure 2 The figure is a schematic diagram of a model training process according to an exemplary embodiment.

[0023] Figure 3 The present invention is a schematic diagram of a photovoltaic power generation equipment abnormality monitoring process according to an exemplary embodiment.

[0024] Figure 4 The present invention is a structural block diagram of a monitoring device for photovoltaic power generation equipment according to an exemplary embodiment.

[0025] Figure 5 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0026] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0027] It should be noted that all actions of acquiring signals, information or data in the present disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the device is located and with the authorization given by the owner of the corresponding device.

[0028] With the development of renewable energy technology, photovoltaic power generation technology has also developed. Photovoltaic power generation technology realizes photovoltaic power generation through photovoltaic power generation equipment.

[0029] During the power generation process of photovoltaic power generation equipment, there may be multiple abnormal factors such as misconnection of photovoltaic equipment, equipment interruption, voltage abnormality, etc., which may lead to abnormalities in photovoltaic power generation equipment.

[0030] Therefore, during the power generation process of photovoltaic power generation equipment, it is necessary to monitor the photovoltaic power generation equipment for abnormalities in order to improve the operating stability and safety of the photovoltaic power generation equipment.

[0031] In the related art, a relatively simple monitoring method is used for abnormal monitoring of photovoltaic power generation equipment. Specifically, a threshold value analysis is performed based on real-time monitoring data. If the threshold condition is not met, it is defined as an abnormality.

[0032] This abnormal monitoring method requires threshold setting, which will affect the monitoring accuracy. In addition, the threshold needs to be constantly updated with the actual working conditions, resulting in low accuracy and efficiency of abnormal monitoring.

[0033] Based on this, the disclosed embodiment provides a technical solution to extract features from the string current data of photovoltaic power generation equipment, obtain target string current features, and use the pre-trained isolation forest model and target string current features to determine the monitoring results of the photovoltaic power generation equipment. Among them, the string current of the photovoltaic power generation equipment is related to the current of each component of the photovoltaic power generation equipment, and can be used to analyze abnormal situations of the photovoltaic power generation equipment; the pre-trained isolation forest model can ensure the efficiency and accuracy of the abnormal situation analysis. Therefore, by using the pre-trained isolation forest model and string current features, it is possible to achieve efficient and accurate analysis of abnormal situations of photovoltaic power generation equipment, thereby improving the operational stability and safety of photovoltaic power generation equipment.

[0034] The technical solution of the embodiment of the present disclosure can be applied to abnormal monitoring of a single photovoltaic power generation device, and can also be applied to abnormal monitoring of each photovoltaic power generation device in a distributed photovoltaic power generation system.

[0035] The technical solution of the embodiment of the present disclosure can be executed by a monitoring device or system of a photovoltaic power generation device. Alternatively, it can also be executed by a photovoltaic power generation device. For example, in a distributed photovoltaic power generation system, multiple distributed photovoltaic power generation devices are included, and these multiple distributed photovoltaic power generation devices can respectively perform abnormal monitoring on themselves and / or other distributed photovoltaic power generation devices.

[0036] Figure 1 The present invention is a flow chart of a method for monitoring a photovoltaic power generation device according to an exemplary embodiment. The monitoring method comprises the following steps: Step S11, obtaining string current data corresponding to the photovoltaic power generation equipment, where the string current data includes a plurality of string current values.

[0037] Step S12, extracting features from the string current data to obtain target string current features.

[0038] Step S13, determining the monitoring result of the photovoltaic power generation equipment according to the target group string current characteristics and the pre-trained isolation forest model, wherein the monitoring result of the photovoltaic power generation equipment is used to characterize the abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output the abnormal evaluation value of the photovoltaic power generation equipment according to the input target group string current characteristics.

[0039] In some embodiments, the photovoltaic power generation equipment involves photovoltaic power generation components, which have a corresponding topological connection relationship. The string current can be the total current generated by multiple photovoltaic power generation components connected in series under the topological connection relationship, which is obtained by adding the output current of each photovoltaic power generation component.

[0040] In addition, the output current of the photovoltaic power generation component generally flows to the inverter, so the string current can also be equivalent to the input current of the inverter branch.

[0041] In step S11, the string current data includes multiple string current values, which are not real-time string current values, but multiple string current values ​​obtained by aggregating string current data within a long time window. Different string current values ​​correspond to different string current collection moments.

[0042] In some embodiments, the photovoltaic power generation device may be configured with a current sensor to collect the string current. For example, a current sensor may be configured at the output end of the photovoltaic power generation device to collect the string current.

[0043] In some embodiments, the current sensor may detect the string current according to a preset current detection cycle, for example, detecting the string current every second to obtain second-level string current data.

[0044] In some embodiments, the photovoltaic power generation equipment may be monitored for abnormalities according to a preset equipment monitoring cycle, for example, the photovoltaic power generation equipment may be monitored for abnormalities every 10 minutes.

[0045] As an optional implementation, step S11 includes: obtaining multiple original string current values ​​of the photovoltaic power generation equipment within a preset time length; aggregating the multiple original string current values ​​to obtain multiple string current values, and the number of the multiple string current values ​​is less than the number of the multiple original string current values.

[0046] In some embodiments, the preset time length may be consistent with the equipment monitoring cycle. For example, if the photovoltaic power generation equipment abnormality monitoring is performed once every 10 minutes, the multiple original string current values ​​are the string current values ​​detected by the current sensor within these 10 minutes.

[0047] In some embodiments, the corresponding string current value may not be detected at some moments, and the string current value at such moments may be a null value.

[0048] It can be understood that the multiple original string current values ​​within the preset time period may involve more data, which can be aggregated to obtain multiple string current values ​​that are less than the number of the multiple original string current values.

[0049] In some embodiments, multiple original string current values ​​are aggregated to obtain multiple string current values, including: dividing the multiple original string current values ​​according to a preset time window to obtain multiple string current sets, each string current set includes at least one original string current value, and the length of the preset time window is less than the preset duration; based on the original string current values ​​included in each string current set, determining the target original string current value corresponding to each string current set; based on the multiple target original string current values, determining multiple string current values.

[0050] In some embodiments, the preset time window can be set according to the aggregation requirement. For example, if the aggregation requirement is to minimize the amount of data, the preset time window can be longer. If the aggregation requirement is to ensure accuracy and reduce a certain amount of data, the preset time window can be shorter.

[0051] As an example, if the preset duration is 10 minutes, the preset time window may be 30 seconds.

[0052] Taking the preset time window of 30 seconds as an example, each string current set involves the original string current value within 30 seconds. Different string current sets have the same time window length, but the corresponding specific time is different. For example, the first string current set is the original string current value within 0 to 30 seconds, the second string current set is the original string current value within 30 to 60 seconds, the third string current set is the original string current value within 60 to 90 seconds, and so on, until all the original string current values ​​are divided.

[0053] In some embodiments, taking the original string current value within 0 to 30 seconds as an example, some moments may correspond to specific original string current values, and some moments may correspond to null values.

[0054] Furthermore, the original string current values ​​included in each string current set may be integrated to obtain a target original string current value.

[0055] For example, the target original string current value of each string current set may be a median original string current value among the original string current values ​​included in each string current set, and the target original string current value may be determined according to a method for determining the median.

[0056] In some embodiments, considering the possibility of null values, the target original string current value finally obtained may also be a null value. Therefore, determining multiple string current values ​​according to multiple target original string current values ​​may include: if the multiple target original string current values ​​do not include null values, determining the multiple target original string current values ​​as multiple string current values; if the multiple target original string current values ​​include null values, performing null value interpolation processing on the multiple target original string current values, and determining multiple string current values ​​according to the multiple target original string current values ​​after the interpolation processing.

[0057] In this implementation, if the multiple target original string current values ​​do not include null values, the multiple target original string current values ​​can be directly determined as multiple string current values. If the multiple target original string current values ​​include null values, the multiple target original string current values ​​can be interpolated with null values, and the multiple target original string current values ​​after interpolation can be determined as multiple string current values.

[0058] In some embodiments, the null value interpolation process may adopt mature interpolation means in the art, such as linear interpolation, which will not be described in detail here.

[0059] Through this implementation, time slices can be aggregated based on second-level data sets to obtain aggregated data sets, thereby achieving refined data preprocessing and improving the sensitivity and accuracy of subsequent isolation forest model anomaly judgment.

[0060] In step S12, feature extraction may be performed based on the string current data to obtain target string current features.

[0061] As an optional implementation, step S12 includes: determining the standard deviations corresponding to multiple string current values; determining a target string current value with a standard deviation abnormality from the multiple string current values ​​according to the standard deviations corresponding to the multiple string current values; and determining the target string current value as the target string current characteristic.

[0062] In this embodiment, the standard deviation of each string current value among multiple string current values ​​can be determined, and then the string current value with abnormal standard deviation is used as the target string current characteristic, that is, the target string current characteristic is the string current value with abnormal standard deviation.

[0063] The determination of the standard deviation may refer to the mature sample standard deviation determination method in the art. In addition, whether the standard deviation is abnormal may be determined by comparing the standard deviations of the current values ​​of each string. For example, if the standard deviation of a string current value is significantly different from the standard deviations of other string current values, the standard deviation of the string current value is abnormal.

[0064] In data feature engineering, string current features based on standard deviation are constructed to help the isolation forest model gain insight into the stability of string current under the same inverter.

[0065] In some embodiments, the target string current characteristics may also be implemented in other ways, for example, determining abnormal string current values ​​in other ways, etc., which are not limited here.

[0066] In step S13, the monitoring result of the photovoltaic power generation equipment is determined according to the target string current characteristics and the pre-trained isolation forest model.

[0067] In some embodiments, the pre-trained isolation forest model can output an abnormal assessment value of the photovoltaic power generation equipment according to the input target string current characteristics.

[0068] Furthermore, in step S13, the target group string current characteristics can be first input into the pre-trained isolation forest model to obtain the abnormal evaluation value output by the pre-trained isolation forest model, and then the photovoltaic power generation equipment monitoring result can be determined according to the abnormal evaluation value.

[0069] In some embodiments, the abnormal condition of the photovoltaic power generation equipment may be directly represented by an abnormal evaluation value, or may be a representation of whether the photovoltaic power generation equipment is abnormal, for example: abnormal or normal.

[0070] In some embodiments, the anomaly assessment value output by the pre-trained forest model may be an anomaly score.

[0071] As an optional implementation, the training process of the pre-trained isolation forest model includes: obtaining a training data set, the training data set includes multiple training samples, each training sample includes a sample group string current characteristic; according to the training data set, unsupervised training is performed on the isolation forest model to be trained to obtain a pre-trained isolation forest model, wherein the isolation forest model to be trained is unsupervisedly trained by constructing an isolation tree, and the depth of the isolation tree is less than a preset depth.

[0072] Since the isolation forest model is an unsupervised training model, only the sample group string current characteristics need to be included in the training samples.

[0073] In some embodiments, the method for obtaining the sample string current characteristics can refer to the method for obtaining the target string current characteristics, except that the original string current data corresponding to the sample string current characteristics is a priori data, rather than the string current data collected during the monitoring process. Therefore, the detailed method for obtaining the training data set is not introduced here.

[0074] In some embodiments, in addition to configuring a training data set, a test data set may also be configured. The test data set may be used to perform an accuracy test on the isolation forest model to improve the model accuracy.

[0075] In some embodiments, the isolation forest model is trained unsupervised by constructing an isolation tree, and the depth of the isolation tree is less than a preset depth.

[0076] In some embodiments, the preset depth can be set according to the processing efficiency requirement of the model. The higher the required processing efficiency, the smaller the preset depth can be. For example, the preset depth can be 40.

[0077] By making some fine-tuning to the isolation forest algorithm in the model prediction process and setting the depth threshold to a preset depth value, the model detection efficiency can be ensured, resource overflow and waste can be prevented, and the detection model deployment cost can be reduced.

[0078] The unsupervised training method of the isolation forest model is a mature technology in this field and is briefly introduced here.

[0079] During training, samples are taken from the training data set, isolated trees are constructed, each isolated tree in the forest is tested, and the path length is recorded, including: Step 1: Randomly select ψ points from the training data set as subsamples and put them into the root node of an isolated tree.

[0080] Step 2: Randomly specify a dimension and randomly generate a cutting point p within the current node data range. The cutting point is generated between the maximum and minimum values ​​of the specified dimension in the current node data.

[0081] Step 3: The selection of this cutting point generates a hyperplane that divides the current node data space into two subspaces: points less than p under the currently selected dimension are placed on the left branch of the current node, and points greater than or equal to p are placed on the right branch of the current node.

[0082] Step 4: Recursively perform steps 2 and 3 on the left and right branch nodes of the node, and continuously construct new leaf nodes until there is only one data on the leaf node (no further cutting is possible) or the tree has grown to the set depth (such as the preset depth).

[0083] Furthermore, the anomaly score of each sample point (training sample) can be calculated according to the anomaly score calculation formula.

[0084] For example, the anomaly score calculation formula can be expressed as: , where x represents any sample point, h(x) represents the depth of sample point x in the isolated tree (i.e., path length), s(x,ψ) represents the abnormal score of the sample point, ranging from 0 to 1, E(h(x)) represents the normalization of h(x), and c(ψ) represents the depth difference of the sample point in the isolated tree, which can be determined based on the depth of the previous sample point in the isolated tree and the depth of the current sample point in the isolated tree. The normalization method and the method for determining the depth difference can refer to the mature technology in this field and will not be introduced in detail here.

[0085] Furthermore, through the above-mentioned implementation method of model training, the isolation forest model can directly output an abnormality assessment value.

[0086] In some embodiments, the abnormal evaluation value can be compared with a preset threshold to determine the abnormal monitoring result. For example, if it is greater than the preset threshold, it means that the photovoltaic power generation equipment is abnormal. If it is less than or equal to the preset threshold, it means that the photovoltaic power generation equipment is normal.

[0087] In some embodiments, in the scenario of distributed photovoltaic power generation equipment, an abnormal assessment value is determined for each distributed photovoltaic power generation equipment. At this time, the abnormal monitoring result of the photovoltaic power generation equipment can also be determined in combination with the abnormal assessment values ​​of other distributed photovoltaic power generation equipment.

[0088] Therefore, as an optional implementation, step 13 includes: inputting the target group string current characteristics into the pre-trained isolation forest model to obtain the abnormal assessment value of the photovoltaic power generation equipment output by the pre-trained isolation forest model; obtaining the abnormal assessment values ​​of other photovoltaic power generation equipment in the distributed photovoltaic power generation system; and determining the monitoring results of the photovoltaic power generation equipment based on at least the abnormal assessment value of the photovoltaic power generation equipment and the abnormal assessment values ​​of other photovoltaic power generation equipment.

[0089] In some embodiments, the average value of the abnormal evaluation value of the photovoltaic power generation equipment and the abnormal evaluation values ​​of other photovoltaic power generation equipment can be determined. If the average value is less than or equal to the aforementioned preset threshold, it proves that the photovoltaic power generation equipment and other photovoltaic power generation equipment are normal.

[0090] In some embodiments, the difference between the abnormal evaluation value of the photovoltaic power generation device and the abnormal evaluation value of other photovoltaic power generation devices can be compared. If the difference is large, it means that the photovoltaic power generation device is abnormal. If the difference is small, it can be said that the photovoltaic power generation device is normal. Alternatively, other determination methods are used, which are not limited here.

[0091] In some embodiments, after determining the monitoring results of the photovoltaic power generation equipment, the method may further include: if the monitoring results of the photovoltaic power generation equipment indicate that the photovoltaic power generation equipment is abnormal, determining maintenance suggestions for the photovoltaic power generation equipment based on the abnormal evaluation value; and feeding back the maintenance suggestions for the photovoltaic power generation equipment to the maintenance end so that the maintenance personnel at the maintenance end can perform maintenance on the photovoltaic power generation equipment.

[0092] In this embodiment, in the event that an abnormality exists in the photovoltaic power generation equipment, a maintenance suggestion may be formulated based on the abnormality evaluation value.

[0093] In some embodiments, different maintenance suggestions may be formulated in advance according to different abnormality assessment values, and a suitable maintenance suggestion may be determined from the pre-formulated maintenance suggestions based on the current abnormality assessment value.

[0094] For example, the higher the abnormality assessment value is, the more serious the abnormality is, the more maintenance items are included in the maintenance suggestions, the longer the maintenance time is, the higher the maintenance personnel experience requirement is, etc.

[0095] In some embodiments, the maintenance end may be the end where the maintenance personnel are located, and the feedback method is not limited to SMS feedback, application feedback, and telephone feedback, etc., and is not limited here.

[0096] In the related art, the wrong abnormal result prompts greatly reduce the efficiency of maintenance work, while the isolation forest algorithm used in the present disclosure can separate outliers from the data set by cutting and dividing the depth to judge the abnormal points, and filter out a few abnormalities through the majority of cases. It helps to improve the accuracy of the abnormality detection model and greatly improve the work efficiency of the station maintenance team.

[0097] In addition, the maintenance scenarios and maintenance suggestions covered by related technologies are relatively simple, while the present invention performs in-depth segmentation through multiple dimensions of abnormal outliers, and can trace maintenance suggestions under different situations based on abnormal scores, giving maintenance personnel more abundant suggestions, making maintenance and inspection work more accurate and reducing the occurrence of misoperation.

[0098] Figure 2 is a schematic diagram of a model training process according to an exemplary embodiment. Figure 2 As shown, in the training process of the isolation forest model, the sample group string current data are first aggregated to obtain a sample group string current data set, and null value interpolation is performed based on the sample group string current data set.

[0099] Next, features are constructed based on the standard deviation to obtain the sample string current feature set.

[0100] Then, the sample string current feature set is divided into a training data set and a test data set according to the ratio.

[0101] Then, based on the training data set and the test data set, the model is trained and tested.

[0102] Figure 3 is a schematic diagram of a photovoltaic power generation equipment abnormality monitoring process according to an exemplary embodiment. Figure 3 As shown, a large amount of second-level measurement point data is first obtained, and based on these second-level measurement point data, aggregation is performed to obtain a string current data set. Based on the string current data set, null value interpolation is performed.

[0103] Next, feature extraction is performed based on the standard deviation to construct the string current characteristics.

[0104] Furthermore, the pre-trained isolation forest model is used to predict the anomaly score based on the string current characteristics.

[0105] Furthermore, the abnormal photovoltaic power generation equipment is determined based on the predicted abnormality score.

[0106] Finally, maintenance recommendations can be output based on the abnormal scores corresponding to the abnormal photovoltaic power generation equipment.

[0107] It can be seen that the present disclosure transforms and iterates the current photovoltaic power generation equipment monitoring method, uses smaller granularity time slices and the isolated forest outlier algorithm for analysis, and imposes certain restrictions on the height of the isolated forest random tree to ensure that the method is efficient. In turn, the accuracy of the abnormal analysis of the string current is improved to a certain extent, thereby ensuring the safety of the photovoltaic power generation equipment and improving the power generation efficiency of the photovoltaic power generation equipment.

[0108] Figure 4 is a structural block diagram of a monitoring device 400 for photovoltaic power generation equipment according to an exemplary embodiment. Figure 4 As shown, the monitoring device comprises: The acquisition module 401 is configured to acquire string current data corresponding to the photovoltaic power generation equipment, wherein the string current data includes a plurality of string current values.

[0109] The feature extraction module 402 is configured to extract features from the string current data to obtain target string current features.

[0110] The monitoring module 403 is configured to determine the monitoring result of the photovoltaic power generation equipment according to the target group string current characteristics and the pre-trained isolation forest model, wherein the monitoring result of the photovoltaic power generation equipment is used to characterize the abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output the abnormal evaluation value of the photovoltaic power generation equipment according to the input target group string current characteristics.

[0111] Optionally, the acquisition module 401 is further configured to: obtain multiple original string current values ​​of the photovoltaic power generation equipment within a first preset time length; aggregate the multiple original string current values ​​to obtain the multiple string current values ​​within a second preset time length, and the second preset time length is less than the first preset time length.

[0112] Optionally, the acquisition module 401 is further configured to: divide the multiple original string current values ​​according to preset time slices to obtain multiple string current sets, each string current set including at least one original string current value; determine the target original string current values ​​corresponding to the multiple string current sets respectively; and determine the multiple string current values ​​based on the multiple target original string current values.

[0113] Optionally, the acquisition module 401 is further configured to: if the multiple target original string current values ​​do not include null values, determine the multiple target original string current values ​​as the multiple string current values; if the multiple target original string current values ​​include null values, perform null value interpolation processing on the multiple target original string current values, and determine the multiple string current values ​​based on the multiple target original string current values ​​after the interpolation processing.

[0114] Optionally, the feature extraction module 402 is further configured to: determine the standard deviations corresponding to the multiple string current values ​​respectively; determine a target string current value with a standard deviation abnormality from the multiple string current values ​​according to the standard deviations corresponding to the multiple string current values ​​respectively; and determine the target string current value as the target string current feature.

[0115] Optionally, the device also includes a training module, which is configured to: obtain a training data set, the training data set includes multiple training samples, each training sample includes a sample group string current characteristic; according to the training data set, perform unsupervised training on the isolation forest model to be trained to obtain the pre-trained isolation forest model, wherein the isolation forest model to be trained is unsupervisedly trained by constructing an isolation tree, and the depth of the isolation tree is less than a preset depth.

[0116] Optionally, the photovoltaic power generation equipment is any photovoltaic power generation equipment in a distributed photovoltaic power generation system, and the monitoring module 403 is further configured to: input the target group string current characteristics into the pre-trained isolation forest model to obtain the abnormal assessment value of the photovoltaic power generation equipment output by the pre-trained isolation forest model; obtain the abnormal assessment values ​​of other photovoltaic power generation equipment in the distributed photovoltaic power generation system; and determine the monitoring result of the photovoltaic power generation equipment based on at least the abnormal assessment value of the photovoltaic power generation equipment and the abnormal assessment values ​​of other photovoltaic power generation equipment.

[0117] Optionally, the monitoring device also includes a feedback module, which is configured to: if the photovoltaic power generation equipment is characterized by an abnormality according to the monitoring result of the photovoltaic power generation equipment, determine a maintenance suggestion for the photovoltaic power generation equipment according to the abnormality evaluation value; and feed back the maintenance suggestion for the photovoltaic power generation equipment to a maintenance end, so that the maintenance personnel at the maintenance end can perform maintenance on the photovoltaic power generation equipment.

[0118] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0119] Figure 5 FIG. 5 is a block diagram of an electronic device 500 according to an exemplary embodiment. Figure 5 As shown, the electronic device 500 may include: a processor 501 , a memory 502 . The electronic device 500 may also include one or more of a multimedia component 503 , an input / output (I / O) interface 504 , and a communication component 505 .

[0120] The processor 501 is used to control the overall operation of the electronic device 500 to complete all or part of the steps in the above-mentioned photovoltaic power generation equipment monitoring method. The memory 502 is used to store various types of data to support the operation of the electronic device 500, and these data may include instructions for any application or method used to operate on the electronic device 500, and application-related data, such as contact data, messages sent and received, pictures, audio, video, etc. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 503 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 502 or sent through the communication component 505. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 504 provides an interface between the processor 501 and other interface modules, and the other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 505 is used for wired or wireless communication between the electronic device 500 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 505 may include: Wi-Fi module, Bluetooth module, NFC module, etc.

[0121] In an exemplary embodiment, the electronic device 500 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned monitoring method for photovoltaic power generation equipment.

[0122] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned monitoring method of photovoltaic power generation equipment are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 502 including program instructions, and the above-mentioned program instructions can be executed by the processor 501 of the electronic device 500 to complete the above-mentioned monitoring method of photovoltaic power generation equipment.

[0123] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0124] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0125] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for monitoring photovoltaic power generation equipment, characterized in that: include: Acquire string current data corresponding to the photovoltaic power generation equipment, wherein the string current data includes a plurality of string current values; Extracting features from the string current data to obtain target string current features; The photovoltaic power generation equipment monitoring result is determined according to the target group string current characteristics and the pre-trained isolation forest model, wherein the photovoltaic power generation equipment monitoring result is used to characterize the abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output the abnormal evaluation value of the photovoltaic power generation equipment according to the input target group string current characteristics.

2. The monitoring method according to claim 1, characterized in that: The obtaining of the string current corresponding to the photovoltaic power generation equipment includes: Acquiring multiple original string current values ​​of the photovoltaic power generation equipment within a preset time period; The multiple original string current values ​​are aggregated to obtain the multiple string current values, and the number of the multiple string current values ​​is less than the number of the multiple original string current values.

3. The monitoring method according to claim 2, characterized in that: The aggregating the multiple original string current values ​​to obtain the multiple string current values ​​includes: Dividing the multiple original string current values ​​according to a preset time window to obtain multiple string current sets, each string current set includes at least one original string current value, and the length of the preset time window is less than the preset time length; Determining, according to the original string current values ​​included in each of the string current sets, a target original string current value corresponding to each of the string current sets; The plurality of string current values ​​are determined according to the plurality of target original string current values.

4. The monitoring method according to claim 3, characterized in that: The determining the multiple string current values ​​according to the multiple target original string current values ​​includes: If the multiple target original string current values ​​do not include a null value, determining the multiple target original string current values ​​as the multiple string current values; If the multiple target original string current values ​​include null values, null value interpolation processing is performed on the multiple target original string current values, and the multiple string current values ​​are determined according to the multiple target original string current values ​​after the interpolation processing.

5. The monitoring method according to claim 1, characterized in that: The extracting the characteristics of the string current to obtain the target string current characteristics includes: Determine the standard deviations corresponding to the multiple string current values ​​respectively; Determining a target string current value having an abnormal standard deviation from the plurality of string current values ​​according to standard deviations respectively corresponding to the plurality of string current values; The target string current value is determined as the target string current characteristic.

6. The monitoring method according to claim 1, characterized in that: The monitoring method further comprises: Acquire a training data set, wherein the training data set includes a plurality of training samples, each training sample includes a sample string current feature; According to the training data set, the isolation forest model to be trained is subjected to unsupervised training to obtain the pre-trained isolation forest model, wherein the isolation forest model to be trained is subjected to unsupervised training by constructing an isolation tree, and the depth of the isolation tree is less than a preset depth.

7. The monitoring method according to any one of claims 1 to 6, characterized in that: The photovoltaic power generation equipment is any photovoltaic power generation equipment in a distributed photovoltaic power generation system, and the photovoltaic power generation equipment monitoring result is determined according to the target string current characteristics and the pre-trained isolation forest model, including: Inputting the target string current characteristic into the pre-trained isolation forest model to obtain an abnormal evaluation value of the photovoltaic power generation equipment output by the pre-trained isolation forest model; Obtaining abnormal evaluation values ​​of other photovoltaic power generation equipment in the distributed photovoltaic power generation system; The photovoltaic power generation equipment monitoring result is determined based on at least the abnormality evaluation value of the photovoltaic power generation equipment and the abnormality evaluation values ​​of other photovoltaic power generation equipment.

8. The monitoring method according to any one of claims 1 to 6, characterized in that: The monitoring method further comprises: If the photovoltaic power generation equipment monitoring result indicates that the photovoltaic power generation equipment is abnormal, determining a maintenance suggestion for the photovoltaic power generation equipment according to the abnormality evaluation value; The photovoltaic power generation equipment maintenance suggestion is fed back to a maintenance end, so that maintenance personnel at the maintenance end can perform maintenance on the photovoltaic power generation equipment.

9. A monitoring device for photovoltaic power generation equipment, characterized in that: include: An acquisition module is configured to acquire string current data corresponding to the photovoltaic power generation equipment, wherein the string current data includes a plurality of string current values; A feature extraction module is configured to extract features from the string current data to obtain target string current features; The monitoring module is configured to determine the monitoring result of the photovoltaic power generation equipment according to the target group string current characteristics and the pre-trained isolation forest model, wherein the monitoring result of the photovoltaic power generation equipment is used to characterize the abnormal situation of the photovoltaic power generation equipment, and the pre-trained isolation forest model is used to output the abnormal evaluation value of the photovoltaic power generation equipment according to the input target group string current characteristics.

10. The monitoring device according to claim 9, characterized in that: The acquisition module is further configured to: Acquire multiple original string current values ​​of the photovoltaic power generation equipment within a first preset time period; The multiple original string current values ​​are aggregated to obtain the multiple string current values ​​within a second preset time length, where the second preset time length is shorter than the first preset time length.

11. The monitoring device according to claim 10, characterized in that: The acquisition module is further configured to: Dividing the multiple original string current values ​​according to preset time slices to obtain multiple string current sets, each string current set including at least one original string current value; Determine target original string current values ​​corresponding to the multiple string current sets respectively; The multiple string current values ​​are determined according to the multiple target original string current values.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the monitoring method of the photovoltaic power generation equipment described in any one of claims 1 to 8 is implemented.

13. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the photovoltaic power generation equipment monitoring method according to any one of claims 1 to 8.

14. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the monitoring method of the photovoltaic power generation equipment described in any one of claims 1 to 8 is implemented.

Images