Dry-type air-core reactor turn-to-turn short circuit fault identification method based on random matrix

Abstract

The invention relates to a dry-type air-core reactor turn-to-turn short circuit fault identification method and device based on a random matrix, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal comprises current electrical characteristic change information of the dry-type air-core reactor; constructing a random matrix based on the real-time pulse response signal; performing characteristic value analysis on the random matrix to obtain an operation state characteristic parameter of the dry-type air-core reactor; the running state characteristic parameters represent the running state of the dry-type air-core reactor; and inputting the running state characteristic parameters of the dry-type air-core reactor into a pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor. By adopting the method, the turn-to-turn short circuit fault of the dry-type air-core reactor can be accurately and efficiently detected.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method, device, computer equipment, storage medium and computer program product for identifying turn-to-turn short-circuit faults of dry-type air-core reactors based on a random matrix. Background Art

[0002] Dry-type air-core reactors are widely used in power systems. As common power equipment in power systems, dry-type air-core reactors may pose safety hazards once a failure occurs.

[0003] At present, the methods for detecting inter-turn short-circuit faults of dry-type air-core reactors include detection coil method, temperature monitoring method, smoke monitoring method, equivalent impedance monitoring method, etc. The detection coil method is easily affected by the spatial magnetic field, resulting in misjudgment of the state; the temperature detection method relies on the temperature change of the dry-type air-core reactor for detection, but the sensitivity for early detection of faults is not high; the smoke monitoring method is affected by the gas in the environment, and can only provide early warning when a serious fault occurs, and the reliability is poor; the equivalent impedance monitoring method uses the equivalent impedance transformation characteristics of the reactor to monitor inter-turn short-circuit faults, but the impedance change is small in the early stage of the fault, and the monitoring sensitivity is not high.

[0004] Therefore, there is a problem in the traditional technology that it is impossible to accurately and efficiently detect the inter-turn short-circuit fault of the dry-type air-core reactor. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for identifying inter-turn short-circuit faults of dry-type air-core reactors based on random matrices, which can accurately and efficiently detect inter-turn short-circuit faults of dry-type air-core reactors.

[0006] A method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on random matrix, the method comprising:

[0007] Acquire a real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor;

[0008] Based on the real-time impulse response signal, a random matrix is ​​constructed;

[0009] Performing eigenvalue analysis on the random matrix, obtaining the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters characterize the current operating state of the dry-type air-core reactor;

[0010] The operating state characteristic parameters of the dry-type air-core reactor are input into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor.

[0011] In an exemplary embodiment, constructing a random matrix based on a real-time impulse response signal includes:

[0012] Performing filtering processing on the real-time impulse response signal to obtain a filtered response signal;

[0013] Performing equal-interval sampling on the filtered response signal to obtain an equal-interval sampling signal;

[0014] Based on the equally spaced sampling signal, a random matrix is ​​constructed.

[0015] In an exemplary embodiment, a random matrix is ​​constructed based on an equally spaced sampling signal, including:

[0016] Based on the equally spaced sampling signal, an initial matrix is ​​constructed; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling;

[0017] The initial matrix is ​​transformed to obtain a random matrix; the row-column ratio of the random matrix is ​​greater than the row-column ratio of the initial matrix.

[0018] In an exemplary embodiment, the operating state characteristic parameters include the characteristic root average spectrum radius of the random matrix and the eigenvalue of the covariance matrix corresponding to the random matrix. The operating state characteristic parameters of the dry-type air-core reactor are input into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor, including:

[0019] Inputting the input instance composed of the characteristic root average spectrum radius and the eigenvalue of the covariance matrix into the pre-trained fault recognition model, and determining the preset number of training instances closest to the input instance in the training instance set by the pre-trained fault recognition model; the pre-trained fault recognition model is obtained by training using the training instance set;

[0020] The fault degree of the dry-type air-core reactor is determined according to the fault degree corresponding to each training instance in a preset number of training instances.

[0021] A method for training a fault identification model, the method comprising:

[0022] Obtaining the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions; the original pulse response signal includes the electrical characteristic change information of the dry-type air-core reactor;

[0023] Based on each original pulse response signal, a random matrix corresponding to the dry-type air-core reactor under different working conditions is constructed;

[0024] The eigenvalue analysis is performed on each random matrix to obtain the corresponding operating state characteristic parameters of the dry-type air-core reactor under different working conditions; the operating state characteristic parameters corresponding to any working condition represent the operating state of the dry-type air-core reactor under the working condition;

[0025] Generate a training instance set based on the operating state characteristic parameters of the dry-type air-core reactor under different working conditions;

[0026] The training instance set is used to train the fault recognition model to be trained.

[0027] In an exemplary embodiment, obtaining the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions includes:

[0028] The pulse signal generator is used to inject pulse signals of different frequencies and amplitudes into the dry-type air-core reactor;

[0029] The response signals of the dry-type air-core reactor under pulse signals of different frequencies and amplitudes are collected as the original pulse response signals corresponding to the dry-type air-core reactor under different working conditions.

[0030] A dry-type air-core reactor inter-turn short-circuit fault identification device based on a random matrix, the device comprising:

[0031] An acquisition module is used to acquire a real-time pulse response signal of a dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor;

[0032] A building module, used for building a random matrix based on a real-time impulse response signal;

[0033] The analysis module is used to perform eigenvalue analysis on the random matrix to obtain the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters represent the current operating state of the dry-type air-core reactor;

[0034] The identification module is used to input the operating state characteristic parameters of the dry-type air-core reactor into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor.

[0035] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0036] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0037] A computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0038] The above-mentioned dry-type air-core reactor inter-turn short-circuit fault identification method, device, computer equipment, storage medium and computer program product based on random matrix obtain the real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal includes the current electrical characteristic change information of the dry-type air-core reactor; based on the real-time pulse response signal, a random matrix is ​​constructed; eigenvalue analysis is performed on the random matrix to obtain the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters characterize the current operating state of the dry-type air-core reactor; the operating state characteristic parameters of the dry-type air-core reactor are input into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor; in this way, the dry-type air-core reactor response signal can be collected and processed in real time to quickly identify the fault and provide a diagnosis result, thereby shortening the time for fault detection and processing and improving the operating efficiency and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] FIG1 is an application environment diagram of a random matrix-based dry-type air-core reactor turn-to-turn short-circuit fault identification method in one embodiment;

[0041] Figure 2 A schematic flow chart of a method for identifying turn-to-turn short-circuit faults of dry-type air-core reactors based on a random matrix in one embodiment;

[0042] Figure 3 A schematic flow chart of a step of constructing a random matrix in one embodiment;

[0043] Figure 4 A single-ring law spectrum distribution diagram in one embodiment;

[0044] Figure 5 A schematic diagram of a flow chart of a method for training a fault identification model in an embodiment;

[0045] Figure 6 A flow chart of detecting a turn-to-turn short circuit fault of a dry-type air-core reactor using a random matrix theory method in one embodiment;

[0046] Figure 7 The flowchart of one embodiment is for reading data such as response signals during pulse excitation.

[0047] Figure 8 , Fig. 9 A schematic diagram of response signals of a normal dry-type air-core reactor and a dry-type air-core reactor with a turn-to-turn short circuit measured in one embodiment;

[0048] Fig.10 A schematic diagram of a single-loop law generated by an average spectrum radius of a random matrix constructed based on a response signal waveform of a dry-type air-core reactor winding when the winding is healthy in one embodiment;

[0049] Fig.11 A schematic diagram of a single-loop law generated by an average spectrum radius of a random matrix constructed based on a response signal waveform when a turn-to-turn short circuit fault occurs in a dry-type air-core reactor winding in one embodiment;

[0050] Fig.12 A schematic flow chart of a method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on a random matrix in another embodiment;

[0051] Fig.13 It is a structural block diagram of a dry-type air-core reactor turn-to-turn short-circuit fault identification device based on a random matrix in one embodiment;

[0052] Fig.14 A structural block diagram of a training device for a fault identification model in an embodiment;

[0053] FIG. 15 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The dry-type air-core reactor of the present application uses air as a cooling medium, and has the advantages of simple structure, easy maintenance, safety and environmental protection. It is usually composed of a conductive coil and a supporting structure, and does not contain an iron core, thereby avoiding core loss and saturation problems. The coil is wrapped with an insulating material in the same field, which can prevent short circuits and improve voltage resistance. The dry-type air-core reactor inter-turn short-circuit fault identification method based on a random matrix provided in an embodiment of the present application can be applied to an application environment as shown in Figure 1. Among them, the terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal includes the current electrical characteristic change information of the dry-type air-core reactor; the server 104 constructs a random matrix based on the real-time pulse response signal; the server 104 performs eigenvalue analysis on the random matrix to obtain the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters characterize the current operating state of the dry-type air-core reactor; the server 104 inputs the operating state characteristic parameters of the dry-type air-core reactor into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0056] In an exemplary embodiment, Figure 2 As shown, a method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on a random matrix is ​​provided, and the method is applied to the server 104 in FIG. 1 as an example for explanation, including the following steps 202 to 206. Among them:

[0057] Step 202: Acquire a real-time pulse response signal of the dry-type air-core reactor.

[0058] Among them, the real-time pulse response signal is the electrical characteristic change information of the dry-type air-core reactor collected in real time by high-precision current sensors and voltage sensors. The electrical characteristic change information refers to the waveform of electrical parameters such as voltage, current, and power changing over time.

[0059] Optionally, the server obtains a real-time pulse response signal of the dry-type air-core reactor.

[0060] Step 204: construct a random matrix based on the real-time impulse response signal.

[0061] Herein, the random matrix may refer to a high-dimensional random matrix.

[0062] Optionally, the server constructs a random matrix based on the real-time impulse response signal.

[0063] Step 206: Perform eigenvalue analysis on the random matrix to obtain operating state characteristic parameters of the dry-type air-core reactor.

[0064] The operating state characteristic parameter represents the current operating state of the dry-type air-core reactor.

[0065] Optionally, the server performs eigenvalue analysis on the random matrix to obtain characteristic parameters that can reflect the operating state of the dry-type air-core reactor.

[0066] Step 208: Input the operating state characteristic parameters of the dry-type air-core reactor into the pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor.

[0067] The pre-trained fault identification model may refer to a pre-trained neural network model for identifying whether the dry-type air-core reactor has sent a fault or a fault state. In practical applications, the fault identification model may be a KNN model based on instance learning or a random forest model based on a decision tree.

[0068] The fault identification result may refer to whether a fault occurs or the degree of the fault.

[0069] Optionally, the server inputs the operating state characteristic parameters of the dry-type air-core reactor into a pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor.

[0070] In the above-mentioned dry-type air-core reactor inter-turn short-circuit fault identification method based on random matrix, a real-time pulse response signal of the dry-type air-core reactor is obtained; the real-time pulse response signal includes the current electrical characteristic change information of the dry-type air-core reactor; based on the real-time pulse response signal, a random matrix is ​​constructed; eigenvalue analysis is performed on the random matrix to obtain the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters characterize the current operating state of the dry-type air-core reactor; the operating state characteristic parameters of the dry-type air-core reactor are input into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor; in this way, the response signal of the dry-type air-core reactor can be collected and processed in real time, the fault can be quickly identified and the diagnosis result can be provided, the time of fault detection and processing can be shortened, and the operating efficiency and reliability of the power system can be improved.

[0071] In an exemplary embodiment, Figure 3 As shown, step 204 includes steps 302 to 306. Among them:

[0072] Step 302: filter the real-time impulse response signal to obtain a filtered response signal.

[0073] The filtered response signal may refer to a signal obtained by denoising the real-time impulse response signal using a wavelet transform method or a low-pass filter.

[0074] Optionally, the server performs filtering processing on the real-time impulse response signal to obtain a filtered response signal.

[0075] Step 304: Perform equal-interval sampling on the filtered response signal to obtain an equal-interval sampling signal.

[0076] Among them, the equally spaced sampling signal may refer to a uniformly sampled signal obtained by performing equal-value sampling on the response curve corresponding to the filtered response signal. Equal-spaced sampling is a prerequisite for the theoretical analysis of random matrices. In order to construct a random matrix, the filtered response signal needs to be sampled at equal intervals.

[0077] Optionally, the server performs equal-interval sampling on the filtered response signal to obtain an equal-interval sampling signal.

[0078] Step 306: construct a random matrix based on the equally spaced sampling signals.

[0079] Optionally, the server constructs a random matrix based on equally spaced sampling signals.

[0080] In this embodiment, the real-time pulse response signal is filtered to obtain a filtered response signal, and the filtered response signal is sampled at equal intervals to obtain an equally spaced sampling signal. Based on the equally spaced sampling signal, a random matrix is ​​constructed, and the real-time pulse response signal can be converted into a random matrix suitable for analysis using random matrix theory, which is beneficial to accurately analyze the real-time pulse response signal of the dry-type air-core reactor and accurately identify the fault condition of the dry-type air-core reactor.

[0081] In an exemplary embodiment, a random matrix is ​​constructed based on an equally spaced sampling signal, including: constructing an initial matrix based on the equally spaced sampling signal; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling; transforming the initial matrix to obtain a random matrix; the row-to-column ratio of the random matrix is ​​greater than the row-to-column ratio of the initial matrix.

[0082] The initial matrix may refer to a matrix constructed for the first time based on the equally spaced sampling signal, and the number of rows and columns of the matrix correspond to the dimension and the number of sampling points of the sampling signal, respectively.

[0083] Optionally, the server first constructs an initial matrix whose number of rows and columns correspond to the dimension of the sampled signal and the number of sampling points, respectively. The server then transforms the initial matrix so that the row-column ratio of the transformed matrix is ​​more suitable for extracting parameters such as eigenvalues, thereby constructing a high-dimensional random matrix.

[0084] In this embodiment, an initial matrix is ​​constructed based on an equally spaced sampling signal; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling; the initial matrix is ​​transformed to obtain a random matrix; the row-to-column ratio of the random matrix is ​​greater than the row-to-column ratio of the initial matrix; in this way, the matrix constructed for the first time can be transformed into a matrix with a more appropriate row-to-column ratio, which is beneficial for accurately analyzing the real-time pulse response signal of the dry-type air-core reactor, and further beneficial for accurately identifying the fault condition of the dry-type air-core reactor.

[0085] In an exemplary embodiment, the operating state characteristic parameters include the characteristic root average spectral radius of the random matrix and the eigenvalues ​​of the covariance matrix corresponding to the random matrix. The operating state characteristic parameters of the dry-type air-core reactor are input into a pre-trained fault recognition model to obtain a fault recognition result of the dry-type air-core reactor, including: inputting the characteristic root average spectral radius and the eigenvalues ​​of the covariance matrix into an input instance into the pre-trained fault recognition model, and determining a preset number of training instances closest to the input instance in the training instance set through the pre-trained fault recognition model; the pre-trained fault recognition model is obtained by training with the training instance set; and determining the fault degree of the dry-type air-core reactor according to the fault degree corresponding to each training instance in the preset number of training instances.

[0086] Among them, the random matrix can be expressed as X, and the characteristic root mean spectral radius MSR of the random matrix is ​​expressed as the mean of the matrix eigenvalue distribution, and its calculation formula is:

[0087] ;

[0088] in, is the eigenvalue of the matrix.

[0089] In practical applications, if the measured MSR value of the reactor is significantly different from that of a normal reactor, it can be preliminarily considered that the reactor is a faulty reactor.

[0090] Among them, the covariance matrix corresponding to the random matrix can be expressed as S, , is the conjugate transposed matrix of X.

[0091] In practical applications, solving the eigenvalue of S can obtain the eigenvalue of S , this eigenvalue can also be used as a basis for judging whether the reactor has a turn-to-turn short-circuit fault. The obtained eigenvalue is plotted on the complex plane to observe its distribution. According to the single-ring law, the eigenvalue should show a circular ring distribution on the complex plane. Figure 4 is the single-ring law spectrum distribution diagram under ideal conditions, where the red solid line (the innermost solid line) is the theoretical inner diameter, that is, , the red dotted line is the outer diameter (i.e. the outermost dotted line), which is 1 after normalization. Figure 4 The blue dots in the figure are eigenvalues, which are distributed between the inner and outer diameters, i.e., on a single ring. The thick green line (i.e., the line between the outermost and innermost circles) represents the mean of all the calculated eigenvalues, which is recorded as .

[0092] For higher accuracy, the mean spectral radius (MSR) and covariance matrix eigenvalues ​​obtained from the above multiple data can also be used to form a one-dimensional sequence, of which most of the sequences (90%) are used as training sets, and the remaining data are used as test sets. The training set is used to test the K-nearest neighbor model, and the test set is used to evaluate the performance of the model. Through the trained K-nearest neighbor model, the model can determine the optimal K value, which can be more accurate, so that the degree of turn-to-turn short circuit in the dry-type air-core reactor can be accurately judged. In addition to the K-nearest neighbor model, the training set can also be used to test the random forest model, and the test set is used to evaluate the performance of the model. During the training process, the random forest will randomly extract data and features to generate multiple decision trees, and each decision tree will learn according to the division of features. When new data enters the model, all decision trees of the random forest will predict the fault based on the eigenvalues, and the prediction results of all trees will be voted to obtain the final fault category, so that the degree of turn-to-turn short circuit in the dry-type air-core reactor can be accurately judged.

[0093] The input instance is a pair of numbers consisting of the eigenvalues ​​of the mean spectral radius of the eigenvalues ​​and the covariance matrix.

[0094] The preset number may be a value determined by a model.

[0095] Optionally, the server inputs a group of value pairs consisting of the characteristic root average spectral radius and the eigenvalues ​​of the covariance matrix into a pre-trained fault identification model; if the fault identification model adopts a K-nearest neighbor model, the pre-trained K-nearest neighbor model is used to determine a preset number of value pairs closest to the group of value pairs from the value pairs provided in the training phase, for example, the 7 value pairs closest to the group of value pairs are determined, among these 7 value pairs, different value pairs correspond to different fault degrees, and the number of fault degrees of minor faults, moderate faults, and severe faults in these 7 value pairs is counted, and the fault degree with the largest number of value pairs is determined as the fault degree of the dry-type air-core reactor; if the fault identification model adopts a random forest model, the fault type is determined by voting of the decision trees trained by the value pairs provided by the pre-trained random forest model in the training phase, for example, the above-mentioned data is input, each data will be judged by all decision trees, and the prediction result is output, if most of the 100 trees predict "minor fault", then the input is classified as "minor fault". In the fault identification process of the random forest model, each decision tree independently makes a fault judgment based on the input instance. Then, the model determines the fault degree of the input instance based on the voting results of all decision trees. In this way, the fault degree of the dry-type air-core reactor can be determined quickly and accurately.

[0096] In the present embodiment, an input instance consisting of characteristic root average spectrum radius and eigenvalues ​​of covariance matrix is ​​input into a pre-trained fault recognition model, and a preset number of training instances closest to the input instance are determined in a training instance set by the pre-trained fault recognition model; the pre-trained fault recognition model is obtained by training using the training instance set; the fault degree of the dry-type air-core reactor is determined according to the fault degree corresponding to each training instance in the preset number of training instances, so that the fault degree of the dry-type air-core reactor can be determined quickly and accurately.

[0097] In an exemplary embodiment, Figure 5 As shown, a method for training a fault identification model is provided, and the method is applied to the server 104 in FIG. 1 as an example for explanation, including the following steps 502 to 510. Among them:

[0098] Step 502: Acquire the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions; the original pulse response signal includes the electrical characteristic change information of the dry-type air-core reactor.

[0099] Optionally, the server obtains original pulse response signals corresponding to the dry-type air-core reactor under different working conditions.

[0100] Step 504: construct a random matrix corresponding to the dry-type air-core reactor under different working conditions based on each original impulse response signal.

[0101] Optionally, the server constructs a random matrix corresponding to the dry-type air-core reactor under different working conditions based on each original pulse response signal.

[0102] Step 506, performing eigenvalue analysis on each random matrix to obtain operating state characteristic parameters corresponding to the dry-type air-core reactor under different working conditions; the operating state characteristic parameters corresponding to any working condition represent the operating state of the dry-type air-core reactor under the working condition.

[0103] Optionally, the server performs eigenvalue analysis on each random matrix to obtain characteristic parameters reflecting the operating state of the dry-type air-core reactor under different working conditions.

[0104] Step 508: Generate a training instance set based on the operating state characteristic parameters of the dry-type air-core reactor under different working conditions.

[0105] Among them, any training instance in the training instance set is a set of digital pairs consisting of the characteristic root average spectrum radius and the eigenvalue of the covariance matrix. The training instance corresponds to the input instance in the above embodiment. The training instance is the training data input to the fault recognition model to be trained in the training phase, and the input instance is the input data input to the pre-trained fault recognition model in the test phase.

[0106] Optionally, the server generates a training instance set based on operating state characteristic parameters of the dry-type air-core reactor under different working conditions.

[0107] Step 510: Use the training instance set to train the fault recognition model to be trained.

[0108] The fault identification model to be trained may be a random forest model or a K-nearest neighbor model. The trained fault identification model may identify whether a dry-type air-core reactor is faulty and may identify the fault degree if a fault occurs.

[0109] Optionally, the server uses a training instance set to train the fault recognition model to be trained to obtain a trained fault recognition model.

[0110] The training method of the above-mentioned fault identification model obtains the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions; the original pulse response signal includes the electrical characteristic change information of the dry-type air-core reactor; based on each original pulse response signal, a random matrix corresponding to the dry-type air-core reactor under different working conditions is constructed; eigenvalue analysis is performed on each random matrix to obtain the operating state characteristic parameters corresponding to the dry-type air-core reactor under different working conditions; the operating state characteristic parameters corresponding to any working condition characterize the operating state of the dry-type air-core reactor under the working condition; based on the operating state characteristic parameters of the dry-type air-core reactor under different working conditions, a training instance is generated. set; the training instance set is used to train the fault recognition model to be trained; in this way, a large number of original pulse response signals of dry-type air-core reactors under different working conditions can be collected to obtain a large amount of original pulse response data, and a random matrix is ​​constructed based on this, and a large number of key characteristic parameters of the random matrix are extracted, such as eigenvalues, covariance matrices, etc., and the key characteristic parameters of the large number of random matrices are used as input data of the fault recognition model to train the fault recognition model. The trained fault recognition model can realize accurate detection of the inter-turn short-circuit fault of the dry-type air-core reactor, and further, the fault degree of the inter-turn short-circuit fault of the dry-type air-core reactor can be identified.

[0111] In an exemplary embodiment, original pulse response signals corresponding to dry-type air-core reactors under different working conditions are obtained, including: injecting pulse signals of different frequencies and amplitudes into the dry-type air-core reactor through a pulse signal generator; and collecting response signals of the dry-type air-core reactor under pulse signals of different frequencies and amplitudes as original pulse response signals corresponding to the dry-type air-core reactors under different working conditions.

[0112] Optionally, the server generates pulse signals of different frequencies and amplitudes through a pulse signal generator, and injects these pulse signals into the dry-type air-core reactor. High-precision current sensors and voltage sensors are used to collect response signals of the dry-type air-core reactor under different working conditions. These response signals contain information on changes in electrical characteristics of the dry-type air-core reactor under various fault conditions.

[0113] In this embodiment, pulse signals of different frequencies and amplitudes are injected into the dry-type air-core reactor through a pulse signal generator; the response signals of the dry-type air-core reactor under pulse signals of different frequencies and amplitudes are collected as the original pulse response signals corresponding to the dry-type air-core reactor under different working conditions. A large number of original pulse response signals corresponding to the dry-type air-core reactor under different working conditions can be quickly acquired, which is conducive to improving the accuracy of the fault identification model.

[0114] To facilitate the understanding of those skilled in the art, the following exemplary method for diagnosing turn-to-turn short-circuit faults of dry-type air-core reactors based on random matrix theory is provided, and the specific steps include:

[0115] Step 1: Generate pulse signals of different frequencies and amplitudes through a pulse signal generator, and inject these pulse signals into the dry-type air-core reactor. Under different working conditions of the dry-type air-core reactor, use high-precision current sensors and voltage sensors to collect the response signals of the dry-type air-core reactor. These response signals contain information on the changes in the electrical characteristics of the dry-type air-core reactor under various fault conditions.

[0116] Step 2: Input the collected original impulse response signal into the data processing module. First, filter the signal to remove noise and interference, and then uniformly sample the signal to ensure that the time domain signal has a consistent sampling interval on the time axis. The time domain signal after uniform sampling is more suitable for subsequent matrix feature extraction and analysis.

[0117] Step 3: Use the preprocessed uniformly sampled time domain signal to construct a high-dimensional random matrix. First, construct an initial matrix whose rows and columns correspond to the dimension of the sampled signal and the number of sampling points, respectively; then transform the initial matrix so that the matrix row-column ratio is more suitable for extracting parameters such as eigenvalues, thereby constructing a high-dimensional random matrix.

[0118] Step 4: Through random matrix theory, perform eigenvalue analysis on the constructed high-dimensional random matrix to extract characteristic parameters that reflect the operating status of the dry-type air-core reactor. The extracted characteristic parameters are used as input to classify and identify using the random forest model or K-nearest neighbor model to diagnose whether a fault has occurred and the degree of the fault.

[0119] The above method applies a transient pulse signal to the dry air-core reactor, collects the output current, and obtains a high-dimensional random matrix, then calculates the eigenvalue of the matrix, and uses the comparison of the eigenvalues ​​to detect whether the reactor has a turn-to-turn short-circuit fault. In order to facilitate the understanding of those skilled in the art, the following two embodiments are provided to further illustrate the above method.

[0120] In Example 1, see Figure 6 , a flow chart for detecting inter-turn short-circuit faults of dry-type air-core reactors using a random matrix theory method is provided.

[0121] In step 1, in this embodiment, a 35kV reactor is used as the experimental object, and its rated capacity , system voltage U=35kV, total 20 layers of coils. Square wave pulse is used as the excitation signal, that is, a stable square wave pulse voltage signal is input at one end of the dry-type air-core reactor under test, and the output response signal is detected at the other end. Under different degrees of inter-turn short-circuit faults, under the same excitation pulse, there will be differences in the output response signal waveform, which can be used to preliminarily determine whether the dry-type air-core reactor is faulty.

[0122] In step 2, under the influence of various factors, there will inevitably be noise and interference in the output data. In the subsequent data processing, in order to avoid the influence of these factors on the results, noise and interference should be removed when processing the data. At the same time, since equal-interval sampling is a prerequisite for random matrix theory analysis, the response curve should be sampled with equal values ​​to obtain uniformly sampled data.

[0123] Figure 7 This is a flow chart for reading response signals and other data during pulse excitation.

[0124] In step three, the uniformly sampled data obtained in step two is used to form a high-dimensional matrix, so that the eigenvalues ​​and covariance matrix of the matrix can be calculated. By repeating multiple samplings, multiple equally spaced sampling data vectors are obtained, and the equally spaced data vectors obtained by each sampling are used as a row or a column of the matrix to construct the initial matrix. For example, assuming that there are N experiments, and each experiment obtains T frequency domain data points, the size of the initially constructed matrix X is T×N. However, if the number of electrical samples is small, that is, the sampling rate is large or the sampling time window is long, the row-column ratio of the formed matrix will be very small, which is not conducive to calculation. The matrix needs to be transformed to achieve a more appropriate row-column ratio. For example, there is currently a matrix X′ with an inappropriate row-column ratio. Keep the number of rows of the matrix X′ unchanged, and split the T column elements of X′ into i blocks in order, then X′ is divided into i sub-matrices, and these i sub-matrices are translated to the bottom of the original X′ in turn. Since the random matrix only requires the data to be large enough, its row-column ratio is appropriate.

[0125] In step 4, the eigenvalues ​​of the random matrix are extracted, mainly the mean spectral radius (MSR) and the eigenvalues ​​of the covariance matrix. The mean spectral radius (MSR) is expressed as the mean of the matrix eigenvalue distribution, and its calculation formula is:

[0126] ;

[0127] in, is the eigenvalue of the matrix.

[0128] In practical applications, if the measured MSR value of the reactor is significantly different from that of the normal reactor, it can be preliminarily considered that the dry-type air-core reactor is a faulty reactor. If the empirical spectral distribution (ESD) is used to describe the distribution of the eigenvalues ​​of the matrix, the difference in the average spectral radius can be observed more intuitively. For the random matrix X, its corresponding covariance matrix S is:

[0129] ;

[0130] in, is the conjugate transposed matrix of X. Solving the eigenvalue of S gives the eigenvalue of S , this eigenvalue can also be used as a basis for judging whether the reactor has a turn-to-turn short-circuit fault. The obtained eigenvalue can be plotted on the complex plane to observe its distribution. According to the single-ring law, the eigenvalue should show a circular ring distribution on the complex plane. See Figure 4 As well as the above Figure 4 Corresponding explanation, for higher accuracy, the mean spectral radius (MSR) and covariance matrix eigenvalues ​​obtained from the above multiple data are used to form a one-dimensional sequence, most of which (90%) are used as training sets, and the remaining data are used as test sets. The training set is used to test the random forest model or the K-nearest neighbor model, and the test set is used to evaluate the performance of the model. Through the trained random forest model, voting classification can be performed through the integration of multiple decision trees, so as to more accurately determine the fault degree of the inter-turn short circuit of the dry-type air-core reactor; or, through the trained K-nearest neighbor model, the optimal K value can be determined, so as to more accurately determine the degree of the inter-turn short circuit of the dry-type air-core reactor.

[0131] In Example 2, step 1 uses a 35kV reactor as the experimental object, and its rated capacity , system voltage U=35kV, total 20 layers of coils. Square wave pulse is used as the excitation signal, and the pulse data used are: low level V1=0V, high level V2=500V, pulse delay TD=1us, pulse rise time TR=30ns, pulse fall time TF=30ns, pulse width PW=300ns, pulse period PER=1s. Under this pulse, the response signals of the normal and turn-to-turn short-circuited dry-type air-core reactors are measured as follows: Figure 8 , Fig. 9 shown.

[0132] In step 2, the interference of the measured data can be reduced by using a low-pass filter or wavelet transform to denoise the acquired raw data, remove the noise components, and retain the useful signal information. In embodiment 2, wavelet transform is used, and its mathematical expression is:

[0133] ;

[0134] in: It's a signal The wavelet coefficients at scale a and position b, a is the scale parameter, which determines the expansion and contraction of the wavelet function, and b is the translation parameter, which determines the position of the wavelet function. is the mother wavelet function, is the original signal.

[0135] Wavelet transform decomposes the signal into wavelet coefficients of different scales and positions. These coefficients contain the local characteristics and frequency information of the signal. The wavelet coefficients are then threshold processed, and the coefficients less than a certain threshold are set to zero (hard threshold) or softened (soft threshold) to remove noise components. The processed wavelet coefficients are inversely transformed to obtain the denoised signal. After denoising the signal, the signal is sampled at equal intervals. The data of the equal-interval sampling is organized to form a high-dimensional matrix.

[0136] In step 3, for the high-dimensional matrix formed by equally spaced sampling in step 2 above, row-column transformation is performed when the row-column ratio is inappropriate. For example, the waveform of the dry-type air-core reactor before and after the fault is obtained. Sampling points form the original data matrix X′, as follows:

[0137] ;

[0138] The first element in the matrix X′ represents the output current data of the first sampling point, that is, , whose row-column ratio is The original data matrix X′ is divided into 50 blocks by rows and columns, and then translated and expanded to form the state data matrix X′ as follows:

[0139] ;

[0140] The ratio of the number of rows to the number of columns of the matrix becomes 1 / 4. The above method is used to transform X′ into the matrix X. After the measured data is formed into a high-dimensional matrix, the average spectral radius of the characteristic root of the random matrix of the dry-type air-core reactor in the non-fault state is calculated according to the above transformation and the average spectral radius calculation formula. is: r = 0.0736. The characteristic root average spectral radius of the random matrix of the dry-type air-core reactor with inter-turn short-circuit fault is The average spectral radius of the random matrix constructed by the response signal waveform of the dry-type air-core reactor winding when it is healthy and when a turn-to-turn short-circuit fault occurs can generate the single-loop law diagram that meets the requirements. Fig.10 , Fig.11 shown.

[0141] In step 4, in order to more accurately test the fault degree of the inter-turn short circuit of the dry-type air-core reactor, the random forest model (or K-nearest neighbor model) is used. Specifically, the mean spectral radius (MSR) calculated above and the eigenvalue sequence of the covariance matrix are combined into a new sequence, and 90% of these new sequences are used for training the random forest algorithm (or K-nearest neighbor model), and the rest are used for recognition testing. For the distance metric, the Euclidean distance is selected in this example. For the eigenvalue spectrum distribution of the waveforms of any two types of signals, the Euclidean distance of their MSR is ,in:

[0142] ;

[0143] The present invention can extract more accurate and comprehensive characteristic parameters, such as mean spectral radius (MSR) and covariance matrix eigenvalues, from high-dimensional data by introducing random matrix theory. These parameters can more accurately reflect the operating state of the reactor and improve the accuracy of fault detection. At the same time, by combining with the random forest model (or K-nearest neighbor model), the extracted characteristic parameters are classified and identified, and different degrees of inter-turn short-circuit faults can be effectively distinguished, thereby improving the accuracy of fault identification. Since the random matrix theory has significant advantages in processing high-dimensional data and noise, it can extract effective characteristic parameters in complex and interference environments, improve the anti-interference ability of the system, solve the problems of low accuracy, high cost, cumbersome methods, and great environmental influence of the current inter-turn short-circuit fault detection technology of dry-type air-core reactors, and reduce the time of data processing by using matrix eigenvalues ​​for comparison. The intelligent detection system based on random matrix theory proposed by the present invention can realize the intelligent detection of inter-turn short-circuit faults. By collecting and processing the response signal of the reactor, the system can quickly identify the fault and provide the diagnosis result, shorten the time of fault detection and processing, and improve the operation efficiency and reliability of the power system.

[0144] In another embodiment, Fig.12 As shown, a method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on a random matrix is ​​provided. The method is applied to the server in FIG1 as an example for explanation, and includes the following steps:

[0145] Step 1202, obtaining a real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor.

[0146] Step 1204: filter the real-time impulse response signal to obtain a filtered response signal.

[0147] Step 1206, sampling the filtered response signal at equal intervals to obtain an equal interval sampling signal.

[0148] Step 1208, constructing an initial matrix based on the equally spaced sampling signal; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling.

[0149] Step 1210, transforming the initial matrix to obtain a random matrix; the row-column ratio of the random matrix is ​​greater than the row-column ratio of the initial matrix.

[0150] Step 1212, performing eigenvalue analysis on the random matrix to obtain operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters characterize the operating state of the dry-type air-core reactor.

[0151] Step 1214: input the operating state characteristic parameters of the dry-type air-core reactor into the pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor.

[0152] It should be noted that the specific limitations of the above steps can refer to the specific limitations of the above method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on random matrices.

[0153] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0154] Based on the same inventive concept, the embodiment of the present application also provides a dry-type air-core reactor turn-to-turn short-circuit fault identification device based on a random matrix for implementing the above-mentioned dry-type air-core reactor turn-to-turn short-circuit fault identification method based on a random matrix, and a fault identification model training device for implementing the above-mentioned fault identification model training method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiments of one or more dry-type air-core reactor turn-to-turn short-circuit fault identification devices based on a random matrix provided below can refer to the limitations of the dry-type air-core reactor turn-to-turn short-circuit fault identification method based on a random matrix above, and the specific limitations in the embodiments of one or more fault identification model training devices provided below can refer to the limitations of the fault identification model training method above, which will not be repeated here.

[0155] In an exemplary embodiment, Fig.13 As shown, a dry-type air-core reactor turn-to-turn short-circuit fault identification device based on a random matrix is ​​provided, comprising: an acquisition module 1302, a construction module 1304, an analysis module 13013 and an identification module 1308, wherein:

[0156] An acquisition module 1302 is used to acquire a real-time pulse response signal of a dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor;

[0157] A construction module 1304 is used to construct a random matrix based on the real-time impulse response signal;

[0158] The analysis module 13013 is used to perform eigenvalue analysis on the random matrix to obtain the operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters represent the operating state of the dry-type air-core reactor;

[0159] The identification module 1308 is used to input the operating state characteristic parameters of the dry-type air-core reactor into the pre-trained fault identification model to obtain the fault identification result of the dry-type air-core reactor.

[0160] In an exemplary embodiment, the construction module 1304 is specifically used to filter the real-time impulse response signal to obtain a filtered response signal; perform equal-interval sampling on the filtered response signal to obtain an equal-interval sampling signal; and construct the random matrix based on the equal-interval sampling signal.

[0161] In an exemplary embodiment, the construction module 1304 is specifically used to construct an initial matrix based on the equally spaced sampling signal; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling; the initial matrix is ​​transformed to obtain the random matrix; the row-to-column ratio of the random matrix is ​​greater than the row-to-column ratio of the initial matrix.

[0162] In an exemplary embodiment, the operating state characteristic parameters include the characteristic root average spectral radius of the random matrix and the eigenvalues ​​of the covariance matrix corresponding to the random matrix. The identification module 1308 is specifically used to input the characteristic root average spectral radius and the eigenvalues ​​of the covariance matrix into an input instance and input it into the pre-trained fault identification model, and determine a preset number of training instances closest to the input instance in the training instance set through the pre-trained fault identification model; the pre-trained fault identification model is obtained by training using the training instance set; and the fault degree of the dry-type air-core reactor is determined according to the fault degree corresponding to each of the training instances in the preset number of training instances.

[0163] In an exemplary embodiment, Fig.14 As shown, a training device for a fault identification model is provided, comprising: an original data acquisition module 1402, a random matrix construction module 1404, a random matrix analysis module 1406, a training instance set generation module 1408 and a fault identification model training module 1410. Among them:

[0164] The original data acquisition module 1402 is used to obtain the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions; the original pulse response signal includes the electrical characteristic change information of the dry-type air-core reactor under various fault states;

[0165] A random matrix construction module 1404 is used to construct a random matrix corresponding to the dry-type air-core reactor under different working conditions based on each of the original pulse response signals;

[0166] The random matrix analysis module 1406 is used to perform eigenvalue analysis on each of the random matrices to obtain the operating state characteristic parameters corresponding to the dry-type air-core reactor under different working conditions; the operating state characteristic parameters corresponding to any of the working conditions represent the operating state of the dry-type air-core reactor under the working conditions;

[0167] A training instance set generating module 1408, configured to generate a training instance set based on the operating state characteristic parameters of the dry-type air-core reactor under different operating conditions;

[0168] The fault identification model training module 1410 is used to train the fault identification model to be trained using the training instance set.

[0169] In an exemplary embodiment, the original data acquisition module 1402 is specifically used to inject pulse signals of different frequencies and different amplitudes into the dry-type air-core inductor through a pulse signal generator; and collect the response signals of the dry-type air-core inductor under the pulse signals of different frequencies and different amplitudes as the original pulse response signals corresponding to the dry-type air-core inductor under different working conditions.

[0170] Each module in the above-mentioned random matrix-based dry-type air-core reactor inter-turn short-circuit fault identification device and fault identification model training device can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0171] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG15. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the dry-type air-core reactor inter-turn short-circuit fault identification data based on a random matrix and the training data of the fault identification model. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a dry-type air-core reactor inter-turn short-circuit fault identification method based on a random matrix is ​​implemented.

[0172] Those skilled in the art will understand that the structure shown in FIG. 15 is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0173] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix, and the steps of the above-mentioned method for training a fault identification model. The steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix here may be the steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix in each of the above-mentioned embodiments, and the steps of the method for training a fault identification model here may be the steps of the method for training a fault identification model in each of the above-mentioned embodiments.

[0174] In one embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the processor executes the steps of the above-mentioned method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix, and executes the steps of the above-mentioned method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix. The steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix here may be the steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix in each of the above-mentioned embodiments, and the steps of the method for training a fault identification model here may be the steps of the method for training a fault identification model in each of the above-mentioned embodiments.

[0175] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix, and the steps of the above-mentioned method for training a fault identification model. The steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix here may be the steps of the method for identifying a dry-type air-core reactor turn-to-turn short-circuit fault based on a random matrix in each of the above-mentioned embodiments, and the steps of the method for training a fault identification model here may be the steps of the method for training a fault identification model in each of the above-mentioned embodiments.

[0176] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0177] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0178] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying inter-turn short-circuit faults of dry-type air-core reactors based on random matrix, characterized in that: The method comprises: Acquire a real-time pulse response signal of the dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor; Based on the real-time impulse response signal, construct a random matrix; Performing eigenvalue analysis on the random matrix to obtain operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters represent the current operating state of the dry-type air-core reactor; The operating state characteristic parameters of the dry-type air-core reactor are input into a pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor.

2. The method according to claim 1, characterized in that The step of constructing a random matrix based on the real-time impulse response signal comprises: Performing filtering on the real-time impulse response signal to obtain a filtered response signal; Performing equal-interval sampling on the filtered response signal to obtain an equal-interval sampling signal; The random matrix is ​​constructed based on the equally spaced sampling signals.

3. The method according to claim 2, characterized in that The step of constructing the random matrix based on the equally spaced sampling signal comprises: Based on the equally spaced sampling signal, an initial matrix is ​​constructed; the number of rows of the initial matrix corresponds to the dimension of the equally spaced sampling signal; the number of columns of the initial matrix corresponds to the number of sampling points of the equally spaced sampling; The initial matrix is ​​transformed to obtain the random matrix; the row-column ratio of the random matrix is ​​greater than the row-column ratio of the initial matrix.

4. The method according to claim 1, characterized in that: The operating state characteristic parameters include the characteristic root average spectral radius of the random matrix and the eigenvalue of the covariance matrix corresponding to the random matrix. The operating state characteristic parameters of the dry-type air-core reactor are input into a pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor, including: The characteristic root average spectrum radius and the eigenvalues ​​of the covariance matrix are used to form an input instance and inputted into the pre-trained fault recognition model, and a preset number of training instances closest to the input instance are determined in a training instance set by the pre-trained fault recognition model; the pre-trained fault recognition model is obtained by training with the training instance set; The fault degree of the dry-type air-core reactor is determined according to the fault degree corresponding to each of the preset number of training instances.

5. A method for training a fault identification model, characterized in that: The method comprises: Acquire the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions; the original pulse response signal includes the electrical characteristic change information of the dry-type air-core reactor; Based on each of the original pulse response signals, constructing a random matrix corresponding to the dry-type air-core reactor under different working conditions; Performing eigenvalue analysis on each of the random matrices to obtain operating state characteristic parameters corresponding to the dry-type air-core reactor under different working conditions; the operating state characteristic parameters corresponding to any of the working conditions characterize the operating state of the dry-type air-core reactor under the working conditions; Generate a training instance set based on the operating state characteristic parameters of the dry-type air-core reactor under different working conditions; The training instance set is used to train the fault recognition model to be trained.

6. The method according to claim 5, characterized in that The obtaining of the original pulse response signal corresponding to the dry-type air-core reactor under different working conditions includes: Injecting pulse signals of different frequencies and amplitudes into the dry-type air-core reactor through a pulse signal generator; The response signals of the dry-type air-core reactor under the pulse signals of different frequencies and different amplitudes are collected as the original pulse response signals corresponding to the dry-type air-core reactor under different working conditions.

7. A dry-type air-core reactor turn-to-turn short-circuit fault identification device based on random matrix, characterized in that: The device comprises: An acquisition module, used for acquiring a real-time pulse response signal of a dry-type air-core reactor; the real-time pulse response signal includes current electrical characteristic change information of the dry-type air-core reactor; A construction module, used for constructing a random matrix based on the real-time impulse response signal; An analysis module, used for performing eigenvalue analysis on the random matrix to obtain operating state characteristic parameters of the dry-type air-core reactor; the operating state characteristic parameters represent the current operating state of the dry-type air-core reactor; The identification module is used to input the operating state characteristic parameters of the dry-type air-core reactor into a pre-trained fault identification model to obtain a fault identification result of the dry-type air-core reactor.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images