Power system fault diagnosis method, device and related equipment based on big data

Abstract

The present application discloses a method, device and related equipment for fault diagnosis of electric power system based on big data, and belongs to the technical field of intelligent electric power system. The present application collects fault fluctuation data, constructs a fault fluctuation database, performs feature analysis on fault data, extracts and constructs a fault feature library, and then extracts the corresponding features of the recorded data to be identified. Finally, the coefficient of variation is used to characterize the load fluctuation intensity, and a construction similarity calculation method for constructing a comprehensive load fluctuation intensity is proposed. The similarity between the recorded data to be identified and the fault features in the fault feature library is calculated and sorted according to the constructed similarity calculation method, and the fault identification result of the recorded data to be identified is obtained according to the similarity sorting. The present application introduces big data technology and feature engineering into the fault diagnosis process of the electric power system, which can make full use of the potential value of historical data, has the characteristics of strong robustness and strong adaptability, and can improve the accuracy of fault identification.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent power systems, and in particular to a power system fault diagnosis method and device based on big data and related equipment. Background Art

[0002] As a key component of electric power generation, transmission and distribution, the power system is indispensable in people's daily life. The safe and stable operation of the power system is a prerequisite for electric energy to exert its effectiveness.

[0003] Currently, there are many research topics on the safety of power systems, but the main technical idea is to build expert systems and traditional fault identification methods, such as positive and negative sequence decomposition, which contain a large amount of expert-level knowledge and experience in a certain field, and can use the knowledge and problem-solving methods of human experts to deal with problems in this field. It does not rely on a large amount of data and has strong interpretability, but poor portability and learning ability. In addition, the scope of what people can consider is limited, which does not conform to the development direction of the increasingly complex power system. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a power system fault diagnosis method, device and related equipment based on big data to solve the technical problems of low recognition accuracy, poor portability and poor learning ability in the existing power system through manual fault identification.

[0005] In order to solve the above technical problems, the present application embodiment provides a power system fault diagnosis method based on big data, which adopts the following technical solutions:

[0006] A method for power system fault diagnosis based on big data, the method comprising the following steps:

[0007] S1, obtaining the recorded wave data to be identified, and determining whether the recorded wave data to be identified is fault data;

[0008] S2, if the recorded wave data to be identified is fault data, extracting the fault feature of the recorded wave data to be identified to obtain the fault feature of the data to be identified;

[0009] S3, using a preset coefficient of variation to characterize the load fluctuation amplitude of the recorded data to be identified, and calculating the load fluctuation intensity of the recorded data to be identified;

[0010] S4, constructing a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity;

[0011] S5, calculating the similarity between the fault feature of the data to be identified and the fault features of each historical data in a preset fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics;

[0012] S6, sorting the similarities, and taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0013] Furthermore, before obtaining the recorded wave data to be identified in step S1 and determining whether the recorded wave data to be identified is fault data, the method further includes:

[0014] S11, collect historical fault data and build a fault waveform database;

[0015] S12, extracting the fault features of each historical fault data in the fault waveform database to obtain the fault features of the historical data, and constructing a fault feature library based on the fault features of the historical data;

[0016] The extracting of the fault characteristics of each historical fault data in the fault waveform database specifically includes:

[0017] S121, standardizing the historical fault data, wherein the standardization calculation formula is as follows:

[0018]

[0019] in, represents the mth voltage data of the jth data x-phase voltage, represents the mth current data of the jth data x-phase current, where 1≤j≤N, x∈{a,b,c}, a, b, c represent the three phases of data x respectively, N is the number of data in the database, and n is the sequence length of voltage and current;

[0020] S122, calculating the average voltage intensity and the average current intensity of the historical fault data to obtain the average voltage fluctuation characteristics and the average current fluctuation characteristics. The calculation formulas for the average voltage intensity and the average current intensity are as follows:

[0021]

[0022]

[0023] Among them, k represents the kth data in the fault waveform database, is the average voltage intensity, is the average current intensity, is the average voltage fluctuation characteristic, is the average fluctuation characteristic of current;

[0024] S123, calculating the average voltage difference value and the average current difference value between the historical fault data to obtain voltage fluctuation difference characteristics and current fluctuation difference characteristics, wherein the calculation formulas of the average voltage difference value and the average current difference value are as follows:

[0025]

[0026] Among them, 1≤j≤N, x,y∈{a,b,c},x≠y, They represent the average voltage difference and current difference between phases x and y respectively.

[0027] Further, in step S1, determining whether the recorded wave data to be identified is fault data specifically includes:

[0028] S31, comparing the voltage of the recorded data to be identified with the limit voltage to obtain a voltage comparison result. The limit voltage calculation formula is as follows:

[0029]

[0030] Among them, u e Indicates the rated voltage, represents the limit voltage, and |·| represents the absolute value;

[0031] S32, comparing the current of the recorded data to be identified with the limiting current to obtain a current comparison result. The limiting current calculation formula is as follows:

[0032] or

[0033] Among them, i e Indicates the rated current, Indicates the limit voltage;

[0034] S33: judging whether the recorded data to be identified is fault data based on the voltage comparison result and the current comparison result.

[0035] Further, in step S33, judging whether the recorded wave data to be identified is fault data based on the voltage comparison result and the current comparison result specifically includes:

[0036] S331, determining, based on the voltage comparison result and the current comparison result, that the voltage of the recorded data to be identified is greater than the limit voltage, and that the current of the recorded data to be identified is greater than the limit current for a time period;

[0037] S332, when the time period is less than three cycles, determining that the recorded data to be identified is fault-free data;

[0038] S333: When the time period is greater than or equal to three cycles, determine that the recorded data to be identified is fault data.

[0039] Furthermore, the formula for calculating the load fluctuation intensity of the recorded data to be identified in step S3 is as follows:

[0040]

[0041] Among them, c p,x is the load fluctuation intensity of phase x, μ p,x , σ p,x They represent the power mean and standard deviation of phase x respectively, Represents the value of the kth point in the power sequence of load phase x.

[0042] Furthermore, the calculation formula of the similarity measurement method of the comprehensive load fluctuation characteristics in step S4 is as follows:

[0043]

[0044] Among them, d j represents the Euclidean distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, r j is the Euclidean fluctuation distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, 1≤j≤N, Represents the value of the kth point in the power sequence of the jth fault characteristic load x phase in the fault characteristic library.

[0045] Furthermore, in step S6, the similarities are sorted, and the fault type of the historical data fault feature corresponding to the maximum similarity value is used as the fault identification result of the recorded data to be identified, which specifically includes:

[0046] S81, sorting the similarities using a preset sorting algorithm to obtain a similarity sorting result;

[0047] S82, obtaining the maximum similarity value in the similarity sorting results, and determining the historical data fault feature corresponding to the maximum similarity value;

[0048] S83, taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0049] In order to solve the above technical problems, the embodiment of the present application also provides a power system fault diagnosis device based on big data, which adopts the following technical solution:

[0050] A power system fault diagnosis device based on big data, the device comprising:

[0051] A first fault identification module is used to obtain the recorded wave data to be identified and determine whether the recorded wave data to be identified is fault data;

[0052] A first feature extraction module is used to extract the fault feature of the recorded data to be identified when the recorded data to be identified is fault data, so as to obtain the fault feature of the data to be identified;

[0053] A load fluctuation calculation module, used to characterize the load fluctuation amplitude of the recorded data to be identified by using a preset coefficient of variation, and calculate the load fluctuation intensity of the recorded data to be identified;

[0054] A measurement method construction module, used for constructing a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity;

[0055] A similarity calculation module, used for calculating the similarity between the fault feature of the data to be identified and the fault features of each historical data in the fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics;

[0056] The second fault identification module is used to sort the similarities and use the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0057] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0058] A computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the power system fault diagnosis method based on big data as described in any one of the above items are implemented.

[0059] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0060] A computer-readable storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement the steps of the power system fault diagnosis method based on big data as described in any one of the above.

[0061] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0062] 1. This application collects and utilizes a large amount of historical fault data, constructs a historical data feature library, and fully utilizes the value of data assets. It overcomes the traditional method of relying solely on expert experience for online fault identification and improves the scalability of the identification algorithm. If a new fault type needs to be identified, it is only necessary to expand the feature library of the corresponding fault in the feature library. On the other hand, as data continues to accumulate, the number of feature libraries is regularly expanded to increase data diversity, thereby improving the accuracy of fault identification.

[0063] 2. The feature extraction process in this application takes into account both voltage and current characteristics to improve the accuracy of fault identification. On the one hand, the feature extraction is performed on the original standardized recording data, and the standardization process improves the robustness of the feature. On the other hand, the difference transformation is performed on each phase of the standardized recording data to extract the difference feature, which can well distinguish the phase where the fault occurs. The transferability is improved by extracting feature data from multiple angles.

[0064] 3. This application proposes a similarity measurement method that comprehensively considers load fluctuations, avoiding the limitations of traditional similarity measurement methods, which are prone to low recognition due to fluctuations and thus cause fault filtering risks. Introducing load fluctuation intensity into the similarity measurement method improves the accuracy of the recognition algorithm to a certain extent.

[0065] 4. This application first determines whether the identification data has fault characteristics, rather than directly identifying the faults in the identification data. Since most of the actual data is normal data, directly matching the characteristics of the identification data not only requires a large amount of calculation, but also easily leads to misjudgment, which improves the processing speed and accuracy of the identification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0067] Figure 1 An exemplary system architecture diagram is shown in which the present application can be applied;

[0068] Figure 2 A flowchart of an embodiment of a power system fault diagnosis method based on big data according to the present application is shown;

[0069] Figure 3 Shows Figure 2 A flowchart of an embodiment of step S2;

[0070] Figure 4 Shows Figure 2 A flowchart of an embodiment of step S3;

[0071] Figure 5 Shows Figure 4 A flowchart of an embodiment of step S33;

[0072] Figure 6 Shows Figure 2 A flowchart of an embodiment of step S8;

[0073] Figure 7 A schematic structural diagram of an embodiment of a power system fault diagnosis device based on big data according to the present application is shown;

[0074] Figure 8 A schematic structural diagram of an embodiment of a computer device according to the present application is shown. DETAILED DESCRIPTION

[0075] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0076] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0077] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0078] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, network 104 and server 105. Network 104 is used to provide a medium for communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0079] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Terminal devices 101, 102, 103 can be installed with various communication client applications, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0080] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0081] Server 105 can be a server that provides various services, such as a background server that provides support for pages displayed on terminal devices 101, 102, and 103. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0082] It should be noted that the power system fault diagnosis method based on big data provided in the embodiment of the present application is generally executed by a server, and accordingly, the power system fault diagnosis device based on big data is generally set in the server.

[0083] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0084] Continue to refer Figure 2 , shows a flow chart of an embodiment of a method for power system fault diagnosis based on big data according to the present application, the method for power system fault diagnosis based on big data comprises the following steps:

[0085] S1, obtaining the recorded wave data to be identified, and determining whether the recorded wave data to be identified is fault data;

[0086] S2, if the recorded wave data to be identified is fault data, extracting the fault feature of the recorded wave data to be identified to obtain the fault feature of the data to be identified;

[0087] S3, using a preset coefficient of variation to characterize the load fluctuation amplitude of the recorded data to be identified, and calculating the load fluctuation intensity of the recorded data to be identified;

[0088] S4, constructing a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity;

[0089] S5, calculating the similarity between the fault feature of the data to be identified and the fault features of each historical data in a preset fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics;

[0090] S6, sorting the similarities, and taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0091] Before obtaining the recorded wave data to be identified in step S1 and determining whether the recorded wave data to be identified is fault data, the method further includes:

[0092] S11, collect historical fault data and build a fault waveform database;

[0093] S12, extracting fault features of each historical fault data in the fault waveform database to obtain historical data fault features, and constructing a fault feature library based on the historical data fault features.

[0094] Specifically, the present application collects historical fault data, including single-phase line break, single-phase grounding, two-phase short circuit, two-phase grounding, three-phase short circuit, three-phase line break and two-phase line break, etc., and constructs a fault fluctuation database; then uses wavelet analysis and time series analysis to perform feature analysis on the collected historical fault data, extracts the fault characteristics of the historical fault data, and constructs a fault feature library, the fault characteristics mainly include the fluctuation intensity before and after the fault of each phase, the average voltage and current before and after the fault, etc.; then obtains the recorded data to be identified, and extracts the data characteristics of the recorded data to be identified; finally, uses a preset coefficient of variation to characterize the load fluctuation intensity of the recorded data to be identified, and constructs a similarity measurement method for the comprehensive load fluctuation characteristics based on the load fluctuation intensity, and calculates the similarity between the recorded data to be identified and each fault feature in the fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics, sorts the calculated similarities, and uses the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0095] In the above embodiments, the present application proposes a power system fault diagnosis method based on big data. By introducing big data technology and feature engineering into the fault diagnosis process of the power system, the potential value of historical fault data can be fully utilized. It has the characteristics of strong robustness and strong adaptability. At the same time, it can improve the accuracy of fault identification and has good scalability.

[0096] For further information, please refer to Figure 3 , the extracting of the fault features of each historical fault data in the fault waveform database specifically includes:

[0097] S121, standardizing the historical fault data, wherein the standardization calculation formula is as follows:

[0098]

[0099] in, represents the mth voltage data of the jth data x-phase voltage, represents the mth current data of the jth data x-phase current, where 1≤j≤N, x∈{a,b,c}, a, b, c represent the three phases of data x respectively, N is the number of data in the database, and n is the sequence length of voltage and current.

[0100] S122, calculating the average voltage intensity and the average current intensity of the historical fault data to obtain the average voltage fluctuation characteristics and the average current fluctuation characteristics. The calculation formulas for the average voltage intensity and the average current intensity are as follows:

[0101]

[0102]

[0103] Among them, k represents the kth data in the fault waveform database, is the average voltage intensity, is the average current intensity, is the average voltage fluctuation characteristic, It is the average fluctuation characteristic of current.

[0104] S123, calculating the average voltage difference value and the average current difference value between the historical fault data to obtain voltage fluctuation difference characteristics and current fluctuation difference characteristics, wherein the calculation formulas of the average voltage difference value and the average current difference value are as follows:

[0105]

[0106] Among them, 1≤j≤N, x,y∈{a,b,c},x≠y, They represent the average voltage difference and current difference between phases x and y respectively.

[0107] It should be noted that in the present application, the historical fault data and the recorded data to be identified are both power data. Currently, the power system generally adopts a three-phase connection or a two-phase connection, so the power data generally has multiple phases. The power system fault diagnosis method of the present application is aimed at a three-phase connected power system, that is, the power data of the present application has three phases a, b, and c at any time point.

[0108] In a specific embodiment of the present application, the historical fault data in the fault waveform database can be expressed as:

[0109]

[0110] Each historical fault data includes three phases a, b, and c. j Standardization and fault feature extraction are performed, and a fault feature library is constructed, denoted as (F j ,g j ), (F j ,g j ) is expressed as follows:

[0111]

[0112] Among them, 1≤j≤N, N is the number of data in the database, g j Marks the fault type.

[0113] In the above embodiment, the feature extraction process takes into account both voltage and current characteristics to improve the accuracy of fault identification. On the one hand, the features are extracted from the original standardized recorded data, and the standardization process improves the robustness of the features. On the other hand, the phases of the standardized recorded data are differentially transformed to extract differential features, which can well distinguish the phases where the fault occurs. The transferability is improved by extracting feature data from multiple angles.

[0114] For further information, please refer to Figure 4 In step S1, determining whether the recorded wave data to be identified is fault data specifically includes:

[0115] S31, comparing the voltage of the recorded data to be identified with the limit voltage to obtain a voltage comparison result. The limit voltage calculation formula is as follows:

[0116]

[0117] Among them, u e Indicates the rated voltage, represents the limit voltage, and |·| represents the absolute value;

[0118] S32, comparing the current of the recorded data to be identified with the limiting current to obtain a current comparison result. The limiting current calculation formula is as follows:

[0119]

[0120] Among them, i e Indicates the rated current, Indicates the limit voltage;

[0121] S33: judging whether the recorded data to be identified is fault data based on the voltage comparison result and the current comparison result.

[0122] For further information, please refer to Figure 5 In step S33, judging whether the recorded wave data to be identified is fault data based on the voltage comparison result and the current comparison result specifically includes:

[0123] S331, determining, based on the voltage comparison result and the current comparison result, that the voltage of the recorded data to be identified is greater than the limit voltage, and that the current of the recorded data to be identified is greater than the limit current for a time period;

[0124] S332, when the time period is less than three cycles, determining that the recorded data to be identified is fault-free data;

[0125] S333: When the time period is greater than or equal to three cycles, determine that the recorded data to be identified is fault data.

[0126] In a specific embodiment of the present application, the expression of the recorded data H to be identified is as follows:

[0127]

[0128] By comparing the voltage of the recorded data H to be identified with the limit voltage, a voltage comparison result is obtained; by comparing the current of the recorded data H to be identified with the limit current, a current comparison result is obtained; and according to the voltage comparison result and the current comparison result, it is determined whether the recorded data H to be identified is a fault number. The specific determination process is as follows:

[0129] Count the time period when the voltage of the recorded data H to be identified is greater than the limit voltage and the current of the recorded data H to be identified is greater than the limit current;

[0130] When the time period is less than three cycles, the recorded data H to be identified is determined to be fault-free data;

[0131] When the time period is greater than or equal to three cycles, the recorded data H to be identified is determined to be fault data.

[0132] It should be additionally explained that, when it is determined that the recorded data H to be identified is fault data, the relevant features of the recorded data H to be identified are extracted according to the above fault feature extraction method, and the feature sequence X of the recorded data H to be identified is obtained as follows:

[0133]

[0134] A similarity calculation is performed based on the feature sequence X and the fault features of the historical data, and the fault identification result of the recorded data H to be identified is determined based on the similarity calculation result.

[0135] Furthermore, the load fluctuation amplitude of the recorded data to be identified is characterized by using a preset coefficient of variation, and the load fluctuation intensity of the recorded data to be identified is calculated. The formula for calculating the load fluctuation intensity of the recorded data to be identified in step S3 is as follows:

[0136]

[0137] Among them, c p,x is the load fluctuation intensity of phase x, μ p,x , σ p,x They represent the power mean and standard deviation of phase x, which is the coefficient of variation of phase x. p,x , σ p,x The constructed equation is used to characterize the load fluctuation amplitude of the recorded data to be identified. Finally, the fluctuation index is calculated by combining the power series curve of the recorded data to be identified, that is, the load fluctuation intensity of the recorded data to be identified. Represents the value of the kth point in the power sequence of load phase x.

[0138] Furthermore, the calculation formula of the similarity measurement method of the comprehensive load fluctuation characteristics in step S4 is as follows:

[0139]

[0140] Among them, d j represents the Euclidean distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, r j is the Euclidean fluctuation distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, 1≤j≤N, Represents the value of the kth point in the power sequence of the jth fault characteristic load x phase in the fault characteristic library.

[0141] Due to the strong fluctuation and randomness of power load, there are certain limitations in calculating similarity using the ordinary Euclidean distance calculation method. In order to eliminate the similarity calculation problem caused by load volatility as much as possible, this application describes the load fluctuation amplitude of the recorded data to be identified by presetting the coefficient of variation, and calculates the load fluctuation intensity of the recorded data to be identified. Based on the load fluctuation intensity, a similarity measurement method for comprehensive load fluctuation characteristics is constructed. The similarity measurement method for comprehensive load fluctuation characteristics is also called σ-Euclidean measurement.

[0142] In the above embodiments, the present application proposes a similarity measurement method for comprehensive load fluctuations, avoiding the limitations of traditional similarity measurement methods, which are prone to low recognition due to fluctuations and thus cause fault filtering risks. Introducing load fluctuation intensity into the similarity measurement method improves the accuracy of the recognition algorithm to a certain extent.

[0143] For further information, please refer to Figure 6 In step S6, the similarities are sorted, and the fault type of the historical data fault feature corresponding to the maximum similarity value is used as the fault identification result of the recorded data to be identified, which specifically includes:

[0144] S81, sorting the similarities using a preset sorting algorithm to obtain a similarity sorting result;

[0145] S82, obtaining the maximum similarity value in the similarity sorting results, and determining the historical data fault feature corresponding to the maximum similarity value;

[0146] S83, taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0147] Specifically, in a specific embodiment of the present application, the similarities are sorted by a preset bubble sorting method to obtain a similarity sorting result, the maximum similarity in the similarity sorting result is obtained, and the historical data fault feature corresponding to the maximum similarity is determined, and the fault type of the historical data fault feature corresponding to the maximum similarity is used as the fault identification result of the recorded data to be identified.

[0148] The present application discloses a method for diagnosing power system faults based on big data, and belongs to the technical field of intelligent power systems. The present application collects fault fluctuation data, constructs a fault fluctuation database, performs feature analysis on the fault data, extracts and constructs a fault feature library, and then extracts the corresponding features of the recorded data to be identified, and finally uses the coefficient of variation to characterize the load fluctuation intensity, proposes a construction similarity calculation method for constructing a comprehensive load fluctuation intensity, and calculates and sorts the similarity between the recorded data to be identified and the fault features in the fault feature library based on the constructed similarity calculation method, and obtains the fault identification result of the recorded data to be identified according to the similarity sorting.

[0149] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0150] 1. This application collects and utilizes a large amount of historical fault data, constructs a historical data feature library, and fully utilizes the value of data assets. It overcomes the traditional method of relying solely on expert experience for online fault identification and improves the scalability of the identification algorithm. If a new fault type needs to be identified, it is only necessary to expand the feature library of the corresponding fault in the feature library. On the other hand, as data continues to accumulate, the number of feature libraries is regularly expanded to increase data diversity, thereby improving the accuracy of fault identification.

[0151] 2. The feature extraction process in this application takes into account both voltage and current characteristics to improve the accuracy of fault identification. On the one hand, the feature extraction is performed on the original standardized recording data, and the standardization process improves the robustness of the feature. On the other hand, the difference transformation is performed on each phase of the standardized recording data to extract the difference feature, which can well distinguish the phase where the fault occurs. The transferability is improved by extracting feature data from multiple angles.

[0152] 3. This application proposes a similarity measurement method that comprehensively considers load fluctuations, avoiding the limitations of traditional similarity measurement methods, which are prone to low recognition due to fluctuations and thus cause fault filtering risks. Introducing load fluctuation intensity into the similarity measurement method improves the accuracy of the recognition algorithm to a certain extent.

[0153] 4. This application first determines whether the identification data has fault characteristics, rather than directly identifying the faults in the identification data. Since most of the actual data is normal data, directly matching the characteristics of the identification data not only requires a large amount of calculation, but also easily leads to misjudgment, which improves the processing speed and accuracy of the identification process.

[0154] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above-mentioned methods. The aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0155] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0156] Further references Figure 7 , as a response to the above Figure 2 The present application provides an embodiment of a power system fault diagnosis device based on big data, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0157] like Figure 7 As shown, the power system fault diagnosis device based on big data described in this embodiment includes:

[0158] The first fault identification module 701 is used to obtain the recorded wave data to be identified and determine whether the recorded wave data to be identified is fault data;

[0159] The second feature extraction module 702 is used to extract the fault feature of the recorded data to be identified when the recorded data to be identified is fault data, so as to obtain the fault feature of the data to be identified;

[0160] The load fluctuation calculation module 703 is used to characterize the load fluctuation amplitude of the recorded data to be identified by using a preset coefficient of variation, and calculate the load fluctuation intensity of the recorded data to be identified;

[0161] A measurement method construction module 704 is used to construct a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity;

[0162] A similarity calculation module 705 is used to calculate the similarity between the fault feature of the data to be identified and each historical data fault feature in a preset fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics;

[0163] The second fault identification module 706 is used to sort the similarities and use the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0164] Furthermore, the power system fault diagnosis device based on big data also includes:

[0165] The historical data collection module 711 is used to collect historical fault data and build a fault waveform database;

[0166] A second feature extraction module 712 is used to extract the fault features of each historical fault data in the fault waveform database, obtain the fault features of the historical data, and build a fault feature library based on the fault features of the historical data;

[0167] The first feature extraction module 712 specifically includes:

[0168] The standardization unit is used to standardize the historical fault data, wherein the standardization calculation formula is as follows:

[0169]

[0170] in, represents the mth voltage data of the jth data x-phase voltage, represents the mth current data of the jth data x-phase current, where 1≤j≤N, x∈{a,b,c}, a, b, c represent the three phases of data x respectively, N is the number of data in the database, and n is the sequence length of voltage and current.

[0171] The average strength calculation unit is used to calculate the average voltage strength and average current strength of the historical fault data to obtain the average voltage fluctuation characteristics and the average current fluctuation characteristics. The calculation formulas of the average voltage strength and the average current strength are as follows:

[0172]

[0173]

[0174] Among them, k represents the kth data in the fault waveform database, is the average voltage intensity, is the average current intensity, is the average voltage fluctuation characteristic, It is the average fluctuation characteristic of current.

[0175] The average difference calculation unit is used to calculate the average voltage difference value and the average current difference value between the historical fault data to obtain the voltage fluctuation difference feature and the current fluctuation difference feature, wherein the calculation formulas of the voltage average difference value and the current average difference value are as follows:

[0176]

[0177] Among them, 1≤j≤N, x,y∈{a,b,c},x≠y, They represent the average voltage difference and current difference between phases x and y respectively.

[0178] Furthermore, the first fault identification module 701 specifically includes:

[0179] The voltage comparison unit is used to compare the voltage of the recorded data to be identified with the limit voltage to obtain a voltage comparison result. The limit voltage calculation formula is as follows:

[0180] or

[0181] Among them, u e Indicates the rated voltage, represents the limit voltage, and |·| represents the absolute value;

[0182] The current comparison unit is used to compare the current of the waveform data to be identified with the limit current to obtain a current comparison result. The limit current calculation formula is as follows:

[0183] or

[0184] Among them, i e Indicates the rated current, Indicates the limit voltage;

[0185] The first fault identification unit is used to determine whether the recorded data to be identified is fault data based on the voltage comparison result and the current comparison result.

[0186] Furthermore, the first fault identification unit specifically includes:

[0187] A time period duration subunit, used to determine, based on the voltage comparison result and the current comparison result, a time period duration during which the voltage of the recorded data to be identified is greater than the limit voltage and the current of the recorded data to be identified is greater than the limit current;

[0188] A first identification result subunit is used to determine that the recorded data to be identified is fault-free data when the time period is less than three cycles;

[0189] The second identification result subunit is used to determine that the recorded data to be identified is fault data when the time period is greater than or equal to three cycles.

[0190] Furthermore, the formula for calculating the load fluctuation intensity of the recorded data to be identified is as follows:

[0191]

[0192] Among them, c p,x is the load fluctuation intensity of phase x, μ p,x , σ p,x They represent the power mean and standard deviation of phase x respectively, Represents the value of the kth point in the power sequence of load phase x.

[0193] Furthermore, the calculation formula of the similarity measurement method of the comprehensive load fluctuation characteristics is as follows:

[0194]

[0195] Among them, d j represents the Euclidean distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, r j is the Euclidean fluctuation distance between the load of the recorded data to be identified and the jth fault feature load in the fault feature library, 1≤j≤N, Represents the value of the kth point in the power sequence of the jth fault characteristic load x phase in the fault characteristic library.

[0196] Furthermore, the second fault identification module 706 specifically includes:

[0197] A similarity sorting unit, used to sort the similarities using a preset sorting algorithm to obtain a similarity sorting result;

[0198] A feature matching unit, used to obtain the maximum similarity value in the similarity sorting result, and determine the historical data fault feature corresponding to the maximum similarity value;

[0199] The second fault identification unit is used to use the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

[0200] In the above embodiments, the present application discloses a power system fault diagnosis device based on big data, which belongs to the technical field of intelligent power systems. The present application collects fault fluctuation data, constructs a fault fluctuation database, performs feature analysis on the fault data, extracts and constructs a fault feature library, and then extracts the corresponding features of the recorded data to be identified. Finally, the coefficient of variation is used to characterize the load fluctuation intensity, and a construction similarity calculation method for constructing a comprehensive load fluctuation intensity is proposed. The similarity between the recorded data to be identified and the fault features in the fault feature library is calculated and sorted according to the constructed similarity calculation method, and the fault identification result of the recorded data to be identified is obtained according to the similarity sorting. The present application introduces big data technology and feature engineering into the fault diagnosis process of the power system, which can make full use of the potential value of historical fault data, has the characteristics of strong robustness and strong adaptability, and can improve the accuracy of fault identification.

[0201] To solve the above technical problems, the present application also provides a computer device. Figure 8 , Figure 8 This is a basic structural block diagram of the computer device in this embodiment.

[0202] The computer device 8 includes a memory 81, a processor 82, and a network interface 83 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 8 with components 81-83, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0203] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0204] The memory 81 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 81 can be an internal storage unit of the computer device 8, such as a hard disk or memory of the computer device 8. In other embodiments, the memory 81 can also be an external storage device of the computer device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 8. Of course, the memory 81 can also include both the internal storage unit of the computer device 8 and its external storage device. In this embodiment, the memory 81 is generally used to store the operating system and various application software installed on the computer device 8, such as computer-readable instructions of a power system fault diagnosis method based on big data, etc. In addition, the memory 81 can also be used to temporarily store various types of data that have been output or are to be output.

[0205] The processor 82 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 82 is generally used to control the overall operation of the computer device 8. In this embodiment, the processor 82 is used to run the computer-readable instructions stored in the memory 81 or process data, such as computer-readable instructions for running the power system fault diagnosis method based on big data.

[0206] The network interface 83 may include a wireless network interface or a wired network interface. The network interface 83 is generally used to establish a communication connection between the computer device 8 and other electronic devices.

[0207] The present application discloses a computer device, which belongs to the technical field of intelligent power system. The present application collects fault fluctuation data, constructs a fault fluctuation database, performs feature analysis on the fault data, extracts and constructs a fault feature library, and then extracts the corresponding features of the recorded data to be identified. Finally, the coefficient of variation is used to characterize the load fluctuation intensity, and a construction similarity calculation method for constructing a comprehensive load fluctuation intensity is proposed. The similarity between the recorded data to be identified and the fault features in the fault feature library is calculated and sorted according to the constructed similarity calculation method, and the fault identification result of the recorded data to be identified is obtained according to the similarity sorting. The present application introduces big data technology and feature engineering into the fault diagnosis process of the power system, which can make full use of the potential value of historical fault data, has strong robustness and adaptability, and can improve the accuracy of fault identification.

[0208] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the big data-based power system fault diagnosis method as described above.

[0209] The present application discloses a storage medium, which belongs to the technical field of intelligent power system. The present application collects fault fluctuation data, constructs a fault fluctuation database, performs feature analysis on the fault data, extracts and constructs a fault feature library, and then extracts the corresponding features of the recorded data to be identified. Finally, the coefficient of variation is used to characterize the load fluctuation intensity, and a construction similarity calculation method for constructing a comprehensive load fluctuation intensity is proposed. The similarity between the recorded data to be identified and the fault features in the fault feature library is calculated and sorted according to the constructed similarity calculation method, and the fault identification result of the recorded data to be identified is obtained according to the similarity sorting. The present application introduces big data technology and feature engineering into the fault diagnosis process of the power system, which can make full use of the potential value of historical fault data, has the characteristics of strong robustness and strong adaptability, and can improve the accuracy of fault identification.

[0210] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0211] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A power system fault diagnosis method based on big data, characterized in that: The method comprises the following steps: S1, obtaining the recorded wave data to be identified, and determining whether the recorded wave data to be identified is fault data; S2, if the recorded wave data to be identified is fault data, extracting the fault feature of the recorded wave data to be identified to obtain the fault feature of the data to be identified; S3, using a preset coefficient of variation to characterize the load fluctuation amplitude of the recorded data to be identified, and calculating the load fluctuation intensity of the recorded data to be identified; S4, constructing a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity; S5, calculating the similarity between the fault feature of the data to be identified and the fault features of each historical data in a preset fault feature library based on the similarity measurement method of the comprehensive load fluctuation characteristics; S6, sorting the similarities, and taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified; The formula for calculating the load fluctuation intensity of the recorded data to be identified in step S3 is as follows: ； in, for Phase load fluctuation intensity, Respectively The mean and standard deviation of the power of each phase, Indicates load The first phase in the power sequence The value of a point; The calculation formula of the similarity measurement method of the comprehensive load fluctuation characteristics in step S4 is as follows: ， ; in, Indicates the recorded data to be identified Phase load and fault feature library The Euclidean distance between the fault characteristic loads, The recorded data to be identified Phase load and fault feature library The Euclidean fluctuation distance between the fault characteristic loads, , Indicates the fault signature library Fault characteristic load The first phase in the power sequence The value of a point.

2. The power system fault diagnosis method based on big data according to claim 1 is characterized in that: Before obtaining the recorded wave data to be identified in step S1 and determining whether the recorded wave data to be identified is fault data, the method further includes: S11, collect historical fault data and build a fault waveform database; S12, extracting the fault features of each historical fault data in the fault waveform database to obtain the fault features of the historical data, and constructing a fault feature library based on the fault features of the historical data; The extracting of the fault characteristics of each historical fault data in the fault waveform database specifically includes: S121, standardizing the historical fault data, wherein the standardization calculation formula is as follows: ， ; in, Indicates Data The first voltage data of the phase voltage, Indicates Data The first current data of the phase current, where , , Respectively represent data The three phases of is the number of data in the database, is the sequence length of voltage and current; S122, calculating the average voltage intensity and the average current intensity of the historical fault data to obtain the average voltage fluctuation characteristics and the average current fluctuation characteristics. The calculation formulas for the average voltage intensity and the average current intensity are as follows: ; in, Indicates the fault waveform database Items of data, is the average voltage intensity, is the average current intensity, is the average voltage fluctuation characteristic, is the average fluctuation characteristic of current; S123, calculating the average voltage difference value and the average current difference value between the historical fault data to obtain voltage fluctuation difference characteristics and current fluctuation difference characteristics, wherein the calculation formulas of the average voltage difference value and the average current difference value are as follows: , ; Among them, various , , , Respectively and The average difference in voltage and current between phases.

3. The power system fault diagnosis method based on big data according to claim 1 is characterized in that: In step S1, determining whether the recorded wave data to be identified is fault data specifically includes: S31, comparing the voltage of the recorded data to be identified with the limit voltage to obtain a voltage comparison result. The limit voltage calculation formula is as follows: or ; in, Indicates the rated voltage, Indicates the limit voltage, Indicates taking the absolute value; S32, comparing the current of the recorded data to be identified with the limiting current to obtain a current comparison result. The limiting current calculation formula is as follows: or ; in, Indicates the rated current, Indicates the limiting current; S33: judging whether the recorded data to be identified is fault data based on the voltage comparison result and the current comparison result.

4. The power system fault diagnosis method based on big data according to claim 3 is characterized in that: In step S33, judging whether the recorded wave data to be identified is fault data based on the voltage comparison result and the current comparison result specifically includes: S331, according to the voltage comparison result and the current comparison result, counting the time period length during which the voltage of the recorded data to be identified is greater than the limit voltage and the current of the recorded data to be identified is greater than the limit current, and judging whether the recorded data to be identified is fault data according to the time period length; S332, when the time period is less than three cycles, determining that the recorded data to be identified is fault-free data; S333: When the time period is greater than or equal to three cycles, determine that the recorded data to be identified is fault data.

5. The power system fault diagnosis method based on big data according to claim 1 is characterized in that: In step S6, the similarities are sorted, and the fault type of the historical data fault feature corresponding to the maximum similarity value is used as the fault identification result of the recorded data to be identified, which specifically includes: S81, sorting the similarities using a preset sorting algorithm to obtain a similarity sorting result; S82, obtaining the maximum similarity value in the similarity sorting results, and determining the historical data fault feature corresponding to the maximum similarity value; S83, taking the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

6. A power system fault diagnosis device based on big data, characterized in that: The device is used to perform the steps of the power system fault diagnosis method based on big data according to any one of claims 1 to 5, and the device includes: A first fault identification module is used to obtain the recorded wave data to be identified and determine whether the recorded wave data to be identified is fault data; A first feature extraction module is used to extract the fault feature of the recorded data to be identified when the recorded data to be identified is fault data, so as to obtain the fault feature of the data to be identified; A load fluctuation calculation module, used to characterize the load fluctuation amplitude of the recorded data to be identified by using a preset coefficient of variation, and calculate the load fluctuation intensity of the recorded data to be identified; A measurement method construction module, used for constructing a similarity measurement method of comprehensive load fluctuation characteristics based on the load fluctuation intensity; A similarity calculation module, used for calculating the similarity between the fault characteristics of the data to be identified and the fault characteristics of each historical data in a preset fault characteristic library based on the similarity measurement method of the comprehensive load fluctuation characteristics; The second fault identification module is used to sort the similarities and use the fault type of the historical data fault feature corresponding to the maximum similarity value as the fault identification result of the recorded data to be identified.

7. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the power system fault diagnosis method based on big data as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps of the power system fault diagnosis method based on big data as described in any one of claims 1 to 5 are implemented.

Images