Distributed photovoltaic equipment communication abnormality diagnosis method, system, equipment and medium

Through the diagnostic model based on the radial basis function neural network, the signal scale and difference characteristics are used to construct a diagnostic model that is suitable for different communication methods, which solves the problem of low communication diagnostic accuracy of distributed photovoltaic equipment in the existing technology, and achieves high efficiency communication abnormality diagnosis.

CN118890291BActive Publication Date: 2024-11-29HUANENG (FUZHOU CHANGLE) PHOTOVOLTAIC POWER GENERATION CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411367841.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-29
Publication Date
2024-11-29
Estimated Expiration
2044-09-29

AI Technical Summary

Technical Problem

The existing distributed photovoltaic equipment communication diagnosis methods are only diagnosed through signal strength, and cannot guarantee the accuracy of the diagnosis results, especially in different equipment models and communication methods.

Method used

Using a diagnostic model based on radial basis function neural network, a training sample set is constructed by obtaining the signal scale and signal difference characteristics of historical communication data, and a clustering center is used to train the communication abnormality diagnosis model that is suitable for different communication methods. .

Benefits of technology

It significantly improves the accuracy of communication abnormality diagnosis of distributed photovoltaic equipment, can adapt to the communication characteristics of different communication methods, simplify the data analysis process, and improves the generalization ability and computing efficiency of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890291B_ABST
    Figure CN118890291B_ABST
Patent Text Reader

Abstract

The present invention relates to a method, system, device and medium for diagnosing communication anomalies of distributed photovoltaic equipment, wherein the method comprises the following steps: acquiring historical communication data of each photovoltaic equipment arranged in a distributed manner, and preprocessing each historical communication data; the preprocessing specifically comprises: acquiring communication signal waves, and acquiring signals whose signal strength exceeds a set threshold value and the total number of signals in each predetermined time period according to the communication signal waves; performing feature extraction on each preprocessed historical communication data, acquiring communication feature data of each historical communication data, and constructing a training sample set; constructing a diagnostic model based on a radial basis function neural network, using different communication modes as clustering centers in the radial basis function neural network, and using the training sample set to train the diagnostic model to obtain a trained photovoltaic equipment communication anomaly diagnostic model; and using the trained photovoltaic equipment communication anomaly diagnostic model to perform online diagnosis of communication anomalies of distributed photovoltaic equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a distributed photovoltaic equipment communication abnormality diagnosis method, system, equipment and medium, and belongs to the technical field of photovoltaic equipment monitoring. Background Art

[0002] The choice of photovoltaic equipment layout is of vital importance to improving energy efficiency. Among the many layout methods, distributed layout has become the most widely used layout strategy for photovoltaic equipment due to its strong flexibility, wide adaptability, and convenient operation and maintenance.

[0003] A Chinese invention patent with publication number CN104201783A discloses a distributed power generation system and a method for diagnosing its communication performance, which enables a user to quickly diagnose the communication performance in the power generation system and discover communication problems in a timely manner.

[0004] However, the above method is only a simple diagnosis of communication performance through signal strength, which is not accurate. In addition, distributed photovoltaic equipment has different equipment installation times, equipment models, and access locations, which leads to different photovoltaic equipment using different communication methods. For example, some photovoltaic equipment uses LORA communication, some photovoltaic equipment uses 5G communication, and some photovoltaic equipment uses carrier communication. Different communication methods have different communication characteristics. Abnormal diagnosis only through signal strength cannot guarantee the accuracy of the diagnostic results. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention proposes a distributed photovoltaic device communication anomaly diagnosis method, system, device and medium.

[0006] The technical solution of the present invention is as follows:

[0007] On the one hand, the present invention provides a distributed photovoltaic device communication abnormality diagnosis method, comprising the following steps:

[0008] Acquire historical communication data of each distributed photovoltaic device and pre-process each historical communication data;

[0009] The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave;

[0010] Perform feature extraction on each pre-processed historical communication data, obtain communication feature data of each historical communication data, and construct a training sample set;

[0011] A diagnostic model based on radial basis function neural network is constructed. Different communication modes are used as cluster centers in the radial basis function neural network. The diagnostic model is trained using the training sample set to obtain a trained photovoltaic equipment communication anomaly diagnostic model.

[0012] The trained photovoltaic equipment communication anomaly diagnosis model is used to perform online diagnosis of communication anomalies in distributed photovoltaic equipment.

[0013] As a preferred implementation, the step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes:

[0014] For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically:

[0015] ;

[0016] in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; It represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time;

[0017] For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically:

[0018] ;

[0019] in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period;

[0020] The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period:

[0021] ;

[0022] ;

[0023] in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient.

[0024] As a preferred implementation, in the step of obtaining the communication feature data of each historical communication data and constructing it as a training sample set, a step of reducing the feature dimension of the communication feature data is also included, specifically:

[0025] Communication characteristic data based on each historical communication data , construct the covariance matrix of each historical communication data:

[0026] ;

[0027] in, For the The covariance matrix of historical communication data, ,in, , =1,2,3,…m, , They correspond to the rows and columns of the covariance matrix respectively, and m is the order of the covariance matrix;

[0028] Get the eigenvectors of the covariance matrix:

[0029] ;

[0030] ;

[0031] in, is the eigenvector of the covariance matrix, is the eigenvalue of the covariance matrix; is a matrix variable The average express expectations;

[0032] Take the eigenvector of the covariance matrix of each historical communication data As the communication feature data after dimension reduction.

[0033] As a preferred implementation, the step of training the diagnostic model using the training sample set is specifically as follows:

[0034] Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as:

[0035] ;

[0036] in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters;

[0037] Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix;

[0038] The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as:

[0039] ;

[0040] in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster;

[0041] Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

[0042] On the other hand, the present invention also provides a distributed photovoltaic equipment communication abnormality diagnosis system, comprising:

[0043] A data acquisition module is used to obtain the historical communication data of each distributed photovoltaic device and pre-process each historical communication data;

[0044] The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave;

[0045] A sample set construction module is used to extract features from each pre-processed historical communication data, obtain communication feature data of each historical communication data, and construct a training sample set;

[0046] The model training module is used to construct a diagnostic model based on a radial basis function neural network. Different communication modes are used as clustering centers in the radial basis function neural network. The diagnostic model is trained using a training sample set to obtain a trained photovoltaic equipment communication anomaly diagnostic model.

[0047] The diagnosis module is used to perform online diagnosis of communication anomalies of distributed photovoltaic equipment using the trained photovoltaic equipment communication anomaly diagnosis model.

[0048] As a preferred implementation, the step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes:

[0049] For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically:

[0050] ;

[0051] in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; It represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time;

[0052] For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically:

[0053] ;

[0054] in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period;

[0055] The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period:

[0056] ;

[0057] ;

[0058] in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient.

[0059] As a preferred implementation, in the step of obtaining the communication feature data of each historical communication data and constructing it as a training sample set, a step of reducing the feature dimension of the communication feature data is also included, specifically:

[0060] Communication characteristic data based on each historical communication data , construct the covariance matrix of each historical communication data:

[0061] ;

[0062] in, For the The covariance matrix of historical communication data, ,in, , =1,2,3,…m, , They correspond to the rows and columns of the covariance matrix respectively, and m is the order of the covariance matrix;

[0063] Get the eigenvectors of the covariance matrix:

[0064] ;

[0065] ;

[0066] in, is the eigenvector of the covariance matrix, is the eigenvalue of the covariance matrix; is a matrix variable The average express expectations;

[0067] Take the eigenvector of the covariance matrix of each historical communication data As the communication feature data after dimension reduction.

[0068] As a preferred implementation, the step of training the diagnostic model using the training sample set is specifically as follows:

[0069] Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as:

[0070] ;

[0071] in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters;

[0072] Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix;

[0073] The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as:

[0074] ;

[0075] in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster;

[0076] Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

[0077] On the other hand, the present invention further proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present invention when executing the program.

[0078] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0079] The present invention has the following beneficial effects:

[0080] 1. A method for diagnosing communication anomalies of distributed photovoltaic equipment in the present invention performs preprocessing when acquiring communication data of distributed photovoltaic equipment to obtain communication feature data that can characterize communication characteristics, and constructs a diagnostic model based on a radial basis function neural network. Cluster centers are established corresponding to communication modes of different photovoltaic equipment as the center points of each node in the hidden layer of the network. The model can adapt to the communication characteristics of different communication modes to perform communication anomaly diagnosis, which greatly improves the accuracy of communication anomaly diagnosis of distributed photovoltaic equipment.

[0081] 2. A distributed photovoltaic device communication anomaly diagnosis method of the present invention characterizes the signal scale size of a predetermined time period by the ratio of the maximum signal strength and the average value of the signal exceeding the threshold, and then characterizes the signal deviation degree within the predetermined time period according to the ratio of the number of signals exceeding the signal threshold and the total number of signals. The communication signal characteristics are constructed by the signal scale size and the signal deviation degree, which can well express the communication characteristics within the predetermined time period.

[0082] 3. Through feature dimensionality reduction operations, the number of features in the data set is reduced, while important information of the original data is retained as much as possible, thereby simplifying the data analysis process and improving the generalization ability and computational efficiency of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 The figure is a schematic diagram of the method flow of the first embodiment of the present invention. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0086] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0087] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0088] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0089] Embodiment 1:

[0090] See also Figure 1 This embodiment provides a method for diagnosing abnormal communication of distributed photovoltaic equipment, which specifically includes the following steps:

[0091] S100, acquiring historical communication data of each photovoltaic device in a distributed arrangement, and preprocessing each historical communication data;

[0092] The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave;

[0093] S200, extracting features from each pre-processed historical communication data, obtaining communication feature data of each historical communication data, and constructing a training sample set;

[0094] S300, constructing a diagnostic model based on a radial basis function neural network, using different communication modes as clustering centers in the radial basis function neural network, and using a training sample set to train the diagnostic model to obtain a trained photovoltaic equipment communication abnormality diagnostic model;

[0095] S400: Use the trained photovoltaic equipment communication anomaly diagnosis model to perform online diagnosis of communication anomalies in distributed photovoltaic equipment.

[0096] As a preferred implementation of this embodiment, in step S200, the step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes:

[0097] For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically:

[0098] ;

[0099] in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; It represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time;

[0100] For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically:

[0101] ;

[0102] in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period;

[0103] The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period:

[0104] ;

[0105] ;

[0106] in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient.

[0107] This embodiment characterizes the signal scale size of the predetermined time period by the ratio of the maximum signal strength and the average value of the signal exceeding the threshold, and then characterizes the degree of signal deviation within the predetermined time period according to the ratio of the number of signals exceeding the signal threshold and the total number of signals. The communication signal characteristics are constructed by the signal scale size and the signal deviation degree, which can well express the communication characteristics within the predetermined time period.

[0108] The amount of communication feature data obtained in step S200 is large. If it is directly used for model training, the required hardware performance is high and the training time is long. As a preferred implementation of this embodiment, in the step of obtaining the communication feature data of each historical communication data and constructing it as a training sample set, a feature dimensionality reduction operation is also performed; specifically:

[0109] Communication characteristic data based on each historical communication data , construct the covariance matrix of each historical communication data:

[0110] ;

[0111] in, For the The covariance matrix of historical communication data, ,in, , =1,2,3,…m, , They correspond to the rows and columns of the covariance matrix respectively, and m is the order of the covariance matrix;

[0112] Get the eigenvectors of the covariance matrix:

[0113] ;

[0114] ;

[0115] in, is the eigenvector of the covariance matrix, is the eigenvalue of the covariance matrix; is a matrix variable The average express expectations;

[0116] Take the eigenvector of the covariance matrix of each historical communication data As the communication feature data after dimension reduction.

[0117] This embodiment reduces the number of features in the data set through feature dimensionality reduction operations, while retaining important information of the original data as much as possible, thereby simplifying the data analysis process and improving the generalization ability and computational efficiency of the model.

[0118] As a preferred implementation of this embodiment, the step of training the diagnostic model using the training sample set is specifically as follows:

[0119] Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as:

[0120] ;

[0121] in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters;

[0122] Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix;

[0123] The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as:

[0124] ;

[0125] in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster;

[0126] Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

[0127] Based on the above implementation, the photovoltaic equipment communication anomaly diagnosis model constructed in this embodiment can adapt to the communication characteristics of different photovoltaic equipment using different communication methods to perform communication anomaly diagnosis, which greatly improves the accuracy of communication anomaly diagnosis of distributed photovoltaic equipment.

[0128] Embodiment 2:

[0129] This embodiment provides a distributed photovoltaic device communication abnormality diagnosis system, including:

[0130] A data acquisition module, used to obtain the historical communication data of each photovoltaic device in a distributed arrangement, and pre-process each historical communication data; this module is used to implement the function of step S100 in the first embodiment, and will not be described in detail here;

[0131] The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave;

[0132] A sample set construction module is used to extract features from each pre-processed historical communication data, obtain communication feature data of each historical communication data, and construct a training sample set; this module is used to implement the function of step S200 in the first embodiment, and will not be repeated here;

[0133] A model training module is used to construct a diagnostic model based on a radial basis function neural network, using different communication modes as cluster centers in the radial basis function neural network, and using a training sample set to train the diagnostic model to obtain a trained photovoltaic device communication anomaly diagnostic model; this module is used to implement the function of step S300 in the first embodiment, and will not be repeated here;

[0134] The diagnosis module is used to perform online diagnosis of communication anomalies of distributed photovoltaic equipment using the trained photovoltaic equipment communication anomaly diagnosis model; this module is used to implement the function of step S400 in the first embodiment, which will not be repeated here.

[0135] As a preferred implementation manner of this embodiment, the step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes:

[0136] For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically:

[0137] ;

[0138] in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time;

[0139] For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically:

[0140] ;

[0141] in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period;

[0142] The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period:

[0143] ;

[0144] ;

[0145] in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient.

[0146] As a preferred implementation, in the step of obtaining the communication feature data of each historical communication data and constructing it as a training sample set, a step of reducing the feature dimension of the communication feature data is also included, specifically:

[0147] Communication characteristic data based on each historical communication data , construct the covariance matrix of each historical communication data:

[0148] ;

[0149] in, For the The covariance matrix of historical communication data, ,in, , =1,2,3,…m, , They correspond to the rows and columns of the covariance matrix respectively, and m is the order of the covariance matrix;

[0150] Get the eigenvectors of the covariance matrix:

[0151] ;

[0152] ;

[0153] in, is the eigenvector of the covariance matrix, is the eigenvalue of the covariance matrix; is a matrix variable The average express expectations;

[0154] Take the eigenvector of the covariance matrix of each historical communication data As the communication feature data after dimension reduction.

[0155] As a preferred implementation, the step of training the diagnostic model using the training sample set is specifically as follows:

[0156] Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as:

[0157] ;

[0158] in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters;

[0159] Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix;

[0160] The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as:

[0161] ;

[0162] in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster;

[0163] Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

[0164] Embodiment three:

[0165] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present invention when executing the program.

[0166] Embodiment 4:

[0167] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present invention is implemented.

[0168] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0169] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0170] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0171] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0172] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for diagnosing abnormal communication of distributed photovoltaic equipment. It is characterized in that The following steps are involved: Acquire historical communication data of each distributed photovoltaic device and pre-process each historical communication data; The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave; Perform feature extraction on each pre-processed historical communication data, obtain communication feature data of each historical communication data, and construct a training sample set; A diagnostic model based on radial basis function neural network is constructed. Different communication modes are used as cluster centers in the radial basis function neural network. The diagnostic model is trained using the training sample set to obtain a trained photovoltaic equipment communication anomaly diagnostic model. Use the trained photovoltaic equipment communication anomaly diagnosis model to perform online diagnosis of communication anomalies in distributed photovoltaic equipment; The step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes: For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically: ; in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; It represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time; For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically: ; in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period; The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period: ; ; in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient; The specific steps of using the training sample set to train the diagnostic model are as follows: Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as: ; in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters; Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix; The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as: ; in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster; Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

2. A distributed photovoltaic equipment communication abnormality diagnosis system, It is characterized in that include: A data acquisition module is used to obtain the historical communication data of each distributed photovoltaic device and pre-process each historical communication data; The preprocessing specifically includes: obtaining a communication signal wave, obtaining a signal whose signal strength exceeds a set threshold value and a total number of signals in each predetermined time period according to the communication signal wave; A sample set construction module is used to extract features from each pre-processed historical communication data, obtain communication feature data of each historical communication data, and construct a training sample set; The model training module is used to construct a diagnostic model based on a radial basis function neural network. Different communication modes are used as clustering centers in the radial basis function neural network. The diagnostic model is trained using a training sample set to obtain a trained photovoltaic equipment communication anomaly diagnostic model. A diagnosis module, which is used to perform online diagnosis of communication anomalies of distributed photovoltaic equipment using a trained photovoltaic equipment communication anomaly diagnosis model; The step of extracting features from each pre-processed historical communication data to obtain communication feature data of each historical communication data includes: For each historical communication data, the signal scale characteristics of each predetermined period are obtained, specifically: ; in, Indicates The historical communication data is in Signal scale characteristics for a predetermined period of time; represents the maximum value function; Indicates The historical communication data is in The first scheduled time period The signal strength of the signal that exceeds the set threshold; N is the The historical communication data is in The number of signals exceeding a set threshold in a predetermined period of time; For each historical communication data, the signal difference characteristics of each predetermined period are obtained, specifically: ; in, Indicates The historical communication data is in Signal difference characteristics for a predetermined period of time; Indicates The historical communication data is in The total number of signals in a predetermined period; The communication feature data of each historical communication data is constructed according to the signal scale feature and signal difference feature of each predetermined period: ; ; in, For the Communication characteristic data of historical communication data; Indicates The historical communication data is in communication characteristics for a predetermined period of time; and is the preset weight coefficient; The step of training the diagnostic model using the training sample set is specifically as follows: Constructing the membership matrix Indicates Whether the training sample belongs to clusters, first perform the membership matrix The objective function of the training is set as: ; in, represents the minimization function, Indicates the input training samples, Indicates Cluster centers; is the membership matrix; is the fuzzy coefficient; is the number of training samples, is the number of clusters; Initialize a membership matrix, calculate each cluster center according to the current membership matrix, and then update the membership matrix with the calculated distance between each cluster center and the input data until the objective function value reaches a preset threshold, and obtain a converged membership matrix; The converged membership matrix is ​​used to train the diagnostic model, and the objective function of the diagnostic model is set as: ; in, Indicates the abnormal state label value of the input sample in the actual communication mode, The input sample indicating the output of the diagnostic model belongs to The abnormal state prediction value of each cluster; Iterate the training until the objective function value of the diagnosis model reaches a preset threshold or reaches the maximum number of iterations, and obtain a trained photovoltaic equipment communication anomaly diagnosis model.

3. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the program, the method according to claim 1 is implemented.

4. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the method according to claim 1 is implemented.

Citation Information

Patent Citations

  • Distributed power generation system and communication performance diagnosing method thereof

    CN104201783A

  • Abnormal event detection method and device

    CN111598610A

  • Photovoltaic string communication abnormity identification method and system based on XGBoost

    CN114298084A