High-voltage circuit breaker mechanical state prediction method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of high-voltage circuit breakers, in particular to a high-voltage circuit breaker mechanical state prediction method and device, computer equipment and a storage medium, and the method comprises the steps: collecting related data samples of a circuit breaker; the method comprises the following steps: preprocessing collected circuit breaker related data samples, and dividing data sets; training the data set by using an LSTM neural network, predicting circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and calculating predicted mechanical characteristic parameters by using prediction results; and inputting the predicted mechanical characteristic parameters into the trained SVM model for diagnosis, and optimizing by using a parallel acceleration module to obtain a prediction result, so that efficient and accurate prediction of the mechanical state of the high-voltage circuit breaker can be realized, the change trend of the mechanical state of the circuit breaker can be mastered in time, possible mechanical faults can be predicted in advance, major accidents can be avoided, and the working efficiency of the circuit breaker can be improved. Economic loss is reduced, and safe and stable operation of a power system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage circuit breakers, and in particular to a method, device, computer equipment and storage medium for predicting a mechanical state of a high-voltage circuit breaker. Background Art

[0002] As an indispensable electrical switchgear in the power system, high-voltage circuit breakers (HVCBs) undertake the important task of cutting off or closing circuits. They have the dual functions of control and protection and are the key link in ensuring the safe operation of the power system. Ensuring the normal operation of HVCBs is crucial to the stability of the power system. By predicting their mechanical state, we can timely understand the changing trend of the HVCB mechanical state and predict possible mechanical failures in advance, thereby avoiding major accidents and reducing economic losses. Based on this, it is of great significance to study the mechanical state prediction method.

[0003] In the current technical field, there are mainly two types of methods, physical modeling and data-driven, for predicting the mechanical state of high-voltage circuit breakers. The physical modeling method is to use statistical probability methods to predict the mechanical state of HVCB based on the measured parameters of various indicators of the circuit breaker. However, this method has obvious limitations, that is, it is not applicable to all types of HVCBs, and its versatility is subject to certain restrictions. The data-driven method does not need to establish a physical model, and directly models and analyzes historical data. Algorithms such as Bayesian algorithms, clustering hierarchical clustering algorithms, and recursive neural networks (RNNs) have been applied to predict the mechanical state of HVCBs. However, when processing data with long time series, these algorithms are prone to model overfitting problems, which affects the accuracy and reliability of the prediction and is difficult to meet the needs of practical applications. Therefore, the existing technology is insufficient in the prediction of the mechanical state of high-voltage circuit breakers, and the existing technology needs to be improved.

[0004] The above information is presented as background information only to assist with understanding the present disclosure and no determination or admission is made as to whether any of the above may be used as prior art with respect to the present disclosure. Summary of the invention

[0005] The present invention provides a method, device, computer equipment and storage medium for predicting the mechanical state of a high-voltage circuit breaker to solve the problems existing in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for predicting a mechanical state of a high-voltage circuit breaker comprises the following steps:

[0008] Collecting circuit breaker related data samples in four mechanical states, wherein the circuit breaker related data includes a circuit breaker signal, contact displacement data, and coil current data;

[0009] Preprocess the collected circuit breaker related data samples and divide the data set;

[0010] The data set is trained using an LSTM neural network to predict the circuit breaker-related data of the HVCB of the circuit breaker in four mechanical states, and the predicted mechanical characteristic parameters are calculated using the prediction results;

[0011] The predicted mechanical characteristic parameters are input into the trained SVM model for diagnosis and optimized using the parallel acceleration module to obtain the prediction results.

[0012] Optionally, the collecting of circuit breaker related data in four mechanical states includes:

[0013] Identify test subjects;

[0014] For the circuit breaking signal, contact displacement data and coil current data of the test object, corresponding data acquisition equipment and acquisition methods are used to measure data under normal conditions, low operating voltage conditions, opening coil core jamming conditions and closing spring fatigue to obtain circuit breaker related data samples.

[0015] Optionally, preprocessing the collected circuit breaker related data samples and dividing the data sets includes:

[0016] Among the circuit breaker related data samples, 80% of the data is used as a training set for model training, and the remaining 20% ​​of the data is used as a test set for verification and test results.

[0017] Optionally, preprocessing the collected circuit breaker related data samples includes:

[0018] Divide and process the data;

[0019] The normalization method is used to normalize the data, that is, mapping the data to (0, 1). The formula is as follows, where and are the maximum and minimum values ​​of the variables to be normalized, respectively.

[0020] Optionally, the data set is trained using an LSTM neural network to predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and the predicted mechanical characteristic parameters are calculated using the prediction results, including:

[0021] Construct multiple LSTM neural network units corresponding to each other for multiple parameter indicators;

[0022] Select mean square error (MSE) as the loss function, train the data set in each LSTM neural network unit, and obtain the prediction results;

[0023] The mechanical characteristic parameters are obtained by calculation based on the prediction results.

[0024] Optionally, the predicted mechanical characteristic parameters are input into a trained SVM model for diagnosis, and optimized using a parallel acceleration module, including:

[0025] Determine the initial parameters of the support vector machine fault diagnosis model;

[0026] Determine the parameters of the support vector machine model and optimize them through cross-validation;

[0027] Import the data set into the trained SVM model for training to obtain the optimal parameters;

[0028] Based on the SVM diagnosis model with optimal parameters, the mechanical state of HVCB is diagnosed using the predicted HVCB mechanical characteristic parameters;

[0029] A parallel acceleration module is used to improve diagnostic efficiency. The parallel acceleration module includes using the multi-process runtime library multiprocessing to call multi-core CPU parallel acceleration calculations to improve the diagnostic algorithm.

[0030] Optionally, it also includes:

[0031] Using the same training data set and predicted mechanical property parameters, the random forest, back propagation neural network, and K-nearest neighbor neural network were trained and tested with the results of SVM diagnosis;

[0032] The diagnostic accuracy of each algorithm was compared to obtain the algorithm with the highest accuracy, and applied to the trained SVM model.

[0033] The present invention further provides a high-voltage circuit breaker mechanical state prediction device, which is used to implement the high-voltage circuit breaker mechanical state prediction method as described in any one of the above items, comprising:

[0034] A data acquisition module, used to collect circuit breaker related data samples in four mechanical states, wherein the circuit breaker related data includes a circuit breaker signal, contact displacement data and coil current data;

[0035] A data preprocessing module is used to preprocess the collected circuit breaker related data samples and divide the data sets;

[0036] A neural network prediction module, used to train the data set using an LSTM neural network, predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and calculate the predicted mechanical characteristic parameters using the prediction results;

[0037] The support vector machine diagnosis module is used to input the predicted mechanical characteristic parameters into the trained SVM model for diagnosis, and use the parallel acceleration module to optimize and obtain the prediction results.

[0038] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the method for predicting the mechanical state of a high-voltage circuit breaker as described in any one of the above items when executing the computer program.

[0039] The present invention also provides a storage medium containing computer executable instructions, wherein the computer executable instructions are executed by a computer processor to implement the method for predicting the mechanical state of a high-voltage circuit breaker as described in any one of the above items.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention provides a method, device, computer equipment and storage medium for predicting the mechanical state of a high-voltage circuit breaker, which can comprehensively collect key data of the circuit breaker under four mechanical states, accurately calculate mechanical characteristic parameters through preprocessing and specific neural network training and prediction, and then use the SVM model for diagnosis and use the parallel acceleration module for optimization, thereby achieving efficient and accurate prediction of the mechanical state of the high-voltage circuit breaker, helping to timely grasp the trend of changes in the mechanical state of the circuit breaker, predict possible mechanical failures in advance, avoid major accidents, reduce economic losses, and ensure the safe and stable operation of the power system.

[0042] The present invention has other features and advantages, which will be apparent from or will be described in detail in the accompanying drawings and the following detailed description incorporated herein, which together serve to explain the specific principles of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a flow chart of a method for predicting the mechanical state of a high-voltage circuit breaker provided by the present invention;

[0045] Figure 2 It is a logic diagram of a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0046] Figure 3It is a BS measurement auxiliary circuit in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0047] Figure 4 It is a result diagram of the collected data of CT, CC and BS in a method for predicting the mechanical state of a high-voltage circuit breaker provided by the present invention;

[0048] Figure 5 This is a LSTM network structure diagram in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0049] Figure 6 It is a comparison diagram of BS prediction results in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0050] Figure 7 It is a comparison diagram of CT prediction results in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0051] Figure 8 It is a comparison diagram of CC prediction results in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0052] Fig. 9 It is a diagnostic result diagram in a high-voltage circuit breaker mechanical state prediction method provided by the present invention;

[0053] Fig.10 It is a structural block diagram of a high-voltage circuit breaker mechanical state prediction device provided by the present invention.

[0054] Figure numerals: 10, data acquisition module; 20, data preprocessing module; 30, neural network prediction module; 40, support vector machine diagnosis module. DETAILED DESCRIPTION

[0055] In order to explain in detail the possible application scenarios, technical principles, specific schemes that can be implemented, and the purposes and effects that can be achieved, the following is a detailed description of the specific embodiments listed in conjunction with the accompanying drawings. The embodiments described herein are only used to more clearly illustrate the technical solutions of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0056] Reference to "embodiment" herein means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The term "embodiment" appearing in various places in the specification does not necessarily refer to the same embodiment, nor does it particularly limit its independence or association with other embodiments. In principle, in the present application, as long as there is no technical contradiction or conflict, the various technical features mentioned in the embodiments can be combined in any way to form a corresponding implementable technical solution.

[0057] Unless otherwise defined, the technical terms used in this document have the same meanings as those generally understood by those skilled in the art to which this application belongs; the use of relevant terms in this document is only for describing specific embodiments and is not intended to limit this application.

[0058] In the description of this application, the term "and / or" is an expression used to describe the logical relationship between objects, indicating that three relationships may exist, for example, A and / or B, which means: A exists, B exists, and A and B exist at the same time. In addition, the character " / " in this article generally indicates that the objects before and after are in an "or" logical relationship.

[0059] In the present application, terms such as “first” and “second” are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship of quantity, priority or sequence between these entities or operations.

[0060] Without further limitations, in this application, the words "include", "comprises", "has" or other similar expressions used in the sentences are intended to cover non-exclusive inclusion. These expressions do not exclude the presence of additional elements in the process, method or product including the elements, so that the process, method or product including a series of elements may include not only those limited elements, but also other elements not explicitly listed, or also include elements inherent to such process, method or product.

[0061] Similar to the understanding in the Examination Guidelines, in this application, expressions such as "greater than", "less than", "exceed" and the like are understood to exclude the number itself; expressions such as "above", "below", "within" and the like are understood to include the number itself. In addition, in the description of the embodiments of this application, "multiple" means more than two (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups", "multiple times", etc., unless otherwise clearly and specifically limited.

[0062] In the description of the embodiments of the present application, space-related expressions used, such as "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "vertical", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicate the orientation or position relationship based on the orientation or position relationship shown in the specific embodiments or drawings, and are only for the convenience of describing the specific embodiments of the present application or facilitating the reader's understanding, and do not indicate or imply that the referred device or component must have a specific position, a specific orientation, or be constructed or operated in a specific orientation, and therefore cannot be understood as a limitation on the embodiments of the present application.

[0063] Unless otherwise expressly specified or limited, in the description of the embodiments of the present application, the terms such as "install", "connect", "connect", "fix", "set", etc. used should be understood in a broad sense. For example, the "connection" can be a fixed connection, a detachable connection, or an integrated setting; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For technicians in the technical field to which the present application belongs, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0064] Embodiment 1

[0065] Please refer to Figure 1 and Figure 2 The embodiment of the present invention provides a method for predicting the mechanical state of a high-voltage circuit breaker, comprising the following steps:

[0066] S1. Collecting circuit breaker related data samples in four mechanical states; the circuit breaker related data includes circuit breaking signal, contact displacement data and coil current data;

[0067] S2, preprocessing the collected circuit breaker related data samples and dividing the data set;

[0068] S3, training the data set using an LSTM neural network to predict the circuit breaker-related data of the HVCB of the circuit breaker in four mechanical states, and using the prediction results to calculate the predicted mechanical characteristic parameters;

[0069] S4. Input the predicted mechanical characteristic parameters into the trained SVM model for diagnosis, and use the parallel acceleration module to optimize and obtain the prediction results.

[0070] Each step is described in detail below.

[0071] In step S1, the circuit breaker related data in four mechanical states are collected, including:

[0072] Identify test subjects;

[0073] For the circuit breaking signal, contact displacement data and coil current data of the test object, corresponding data acquisition equipment and acquisition methods are used to measure data under normal conditions, low operating voltage conditions, opening coil core jamming conditions and closing spring fatigue to obtain circuit breaker related data samples.

[0074] The experiment in this invention uses VS1-12HVCB as the test object, collects test data for example analysis, and measures and collects the breaking signal (BS), contact displacement (CT) and coil current (CC) of the HVCB, which are key operating parameters of the high-voltage circuit breaker when working, and can intuitively reflect the mechanical state of the HVCB.

[0075] The MP24025378S series data acquisition card is selected to collect the break signal (BS), with a sampling rate of 128kHz and a sampling time of 300ms. The collected data is uploaded to the host computer for real-time processing. A 560Ω resistor and a 5V DC power supply are connected to the circuit breaker to form an auxiliary circuit to obtain BS. The auxiliary circuit is as follows Figure 3 shown.

[0076] The coil current (CC) is measured using a Hall current sensor with an operating voltage of 5V. The sensor is installed on the opening / closing coil loop. A conductive plastic potentiometer is used as a displacement sensor to measure the displacement of the HVCB contact.

[0077] BS, CT and CC are collected under four mechanical states: normal (including opening operation (normal_O) and closing operation (normal_C)), low operating voltage (Fault1), opening coil core jamming (Fault2) and closing spring fatigue (Fault3). For example, 50 samples are collected for each state, for a total of 200 samples.

[0078] Then, the data is divided and processed; the collected circuit breaker related data samples are preprocessed and the data sets are divided, including:

[0079] 80% of the circuit breaker related data samples are used as training sets for model training, and the remaining 20% ​​of the data are used as test sets for verification and test results. Figure 4 As shown, (a) is CT, (b) is CC, and (c) is BS.

[0080] It can be seen from the figure that in the case of a fault, the time for contact displacement, reaching the predetermined coil current and circuit breaker signal are much longer than the normal closing and opening time.

[0081] In S2, the collected circuit breaker related data samples are preprocessed, including:

[0082] The normalization method is used to normalize the data, that is, mapping the data to (0, 1). The formula is as follows, where and are the maximum and minimum values ​​of the variables to be normalized, respectively.

[0083] Furthermore, the predicted mechanical characteristic parameters are input into the trained SVM model for diagnosis, and then optimized using parallel acceleration modules, including:

[0084] Determine the initial parameters of the support vector machine fault diagnosis model;

[0085] Determine the parameters of the support vector machine model and optimize them through cross-validation;

[0086] Import the data set into the trained SVM model for training to obtain the optimal parameters;

[0087] Based on the SVM diagnosis model with optimal parameters, the mechanical state of HVCB is diagnosed using the predicted HVCB mechanical characteristic parameters;

[0088] A parallel acceleration module is used to improve diagnostic efficiency. The parallel acceleration module includes using the multi-process runtime library multiprocessing to call multi-core CPU parallel acceleration calculations to improve the diagnostic algorithm.

[0089] Furthermore, it also includes:

[0090] Using the same training data set and predicted mechanical property parameters, the random forest, back propagation neural network, and K-nearest neighbor neural network were trained and tested with the results of SVM diagnosis;

[0091] The diagnostic accuracy of each algorithm was compared to obtain the algorithm with the highest accuracy, and applied to the trained SVM model.

[0092] Among them, the long short-term memory (LSTM) network is a neural network structure that is widely used in anomaly monitoring and prediction. Figure 5 Figure 1 shows the internal structure of the LSTM network. Each network unit has three gates: forget gate, input gate, and output gate, which can remove or add information to the storage unit. The forget gate determines which state information needs to be discarded; the input gate defines which information needs to be retained from the unit state; and the output gate determines which information is output from the unit state.

[0093] The calculations for the three gates are as follows:

[0094] (1)

[0095] (2)

[0096] (3)

[0097] (4)

[0098] (5)

[0099] (6)

[0100] Among them, ft, it, ot, ht, ct, ct are the forget gate, input gate, output gate, candidate cell state, output matrix and cell state vector. Wf, Wi, Wo, Wc are weight matrices; bf, bi, bo, bc represent bias vectors. σ is the sigmoid activation function. The symbol ∘ represents matrix multiplication by elements. Specifically, x is the historical monitoring operation data collected by HVCB, and h is the predicted value of the LSTM network output.

[0101] First, the data of BS, CT and CC are normalized and mapped to (0, 1) for processing, which is convenient for data extraction and quick and easy to use. The formula is shown in (7):

[0102] (7)

[0103] xmax and xmin are the maximum and minimum values ​​of the variable to be normalized, respectively.

[0104] Construct three LSTM network units to predict a parameter indicator respectively. In this model, mean square error (MSE) is selected as the loss function. The smaller the MSE value, the better the prediction effect. The calculation formula is as follows:

[0105] (8)

[0106] Where n is the number of samples; Yi and Y are the actual and predicted values, respectively.

[0107] During the prediction process, the various parameters in the LSTM model are shown in Table 1.

[0108] Table 1 LSTM model parameters

[0109]

[0110] The MSE values ​​of the prediction results are shown in Table 2:

[0111] Table 2 Predicted MSE values

[0112]

[0113] The results of comparing the predicted value with the actual value are as follows Figure 6-8 As shown, Figure 6 For comparison of BS prediction results, Figure 7 For comparison of CT prediction results, Figure 8 Comparison of CC prediction results. Figure 6-8From the MSE data in Table 2, it can be seen that the values ​​of various indicators of the circuit breaker predicted by LSTM are very accurate.

[0114] In step S3, the data set is trained using an LSTM neural network to predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and the predicted mechanical characteristic parameters are calculated using the prediction results, including:

[0115] Construct multiple LSTM neural network units corresponding to each other for multiple parameter indicators;

[0116] Select mean square error (MSE) as the loss function, train the data set in each LSTM neural network unit, and obtain the prediction results;

[0117] The mechanical characteristic parameters are obtained based on the prediction results.

[0118] Support vector machine (SVM) is a machine learning method. The basic idea of ​​support vector machine is to solve the optimal hyperplane that meets the conditions in the high-dimensional feature space, which can accurately divide the training samples. The expression of the hyperplane is shown in (9):

[0119] (9)

[0120] Where w is the weight vector; φ is the objective function; xi is the input of the sample; b is the optimal hyperplane offset. After solving, the optimal classification decision function of the sample can be obtained.

[0121] The SVM model takes the mechanical characteristic parameters under four mechanical conditions as input and outputs four mechanical states of the HVCB, including normal state, low operating voltage, opening coil core jamming, and closing spring fatigue. First, the parameters of the support vector machine fault diagnosis model are determined. The weight vectors w and b are optimized by cross validation. The optimization range of the parameters is set to [-10 10], and the iteration step is 0.2.

[0122] The BS, CT and CC data trained by the LSTM model from the collected test data and the predicted 200 sets of mechanical characteristic parameters were imported into the trained SVM model. According to the training, the optimal parameters w=0.000977 and b=0.000977 were obtained. Based on the SVM diagnosis model with the optimal parameters, the mechanical state of the HVCB was diagnosed using the predicted HVCB mechanical characteristic parameters.

[0123] At the same time, in order to increase the training speed of the model, the multi-process runtime library multiprocessing is used to call multi-core CPU parallel acceleration calculation to improve the diagnosis algorithm.

[0124] The multiprocessing library is a module in the Python standard library that implements parallel computing between multiple processes. It provides the ability to create and manage processes, and can execute diagnostic tasks simultaneously on multiple CPU cores for accelerated computing.

[0125] Multiprocessing is used to generate process tasks. By creating a Process object to create a process, the task here is to perform fault diagnosis of valve cooling equipment. Use the process pool Pool function to create a large number of child processes. The number should not exceed the number of CPU cores of the running computer, otherwise it will affect the diagnosis results and performance. In the experiment, up to 4 fault diagnosis processes can be created. Each process communicates through the queue function queue to achieve data sharing and collaborative work.

[0126] Using the same training data set and predicted mechanical property parameters, random forest (RF), back propagation (BP) neural network and K nearest neighbor (KNN) neural network were trained and tested, and the results were compared with those of SVM diagnosis. The diagnostic accuracy of the four algorithms is shown in Table 3. Fig. 9 Diagnosis result diagram.

[0127] Table 3 Diagnostic results

[0128]

[0129] From Table 3 and Fig. 9 It can be seen that the diagnostic accuracy of the RF model, KNN model and BP neural network are 96.5%, 93% and 90% respectively. The diagnostic accuracy of the SVM model is 98%. The diagnostic accuracy of the support vector machine is higher than that of the other three models. Therefore, the support vector machine can better identify the mechanical status of HVCB.

[0130] Therefore, the present invention proposes that predicting the mechanical state of a high-voltage circuit breaker based on a long short-term memory (LSTM) neural network and a support vector machine (SVM) is an accurate and reliable technical solution.

[0131] Please refer to Fig.10 Based on the above embodiments, the present invention also provides a high-voltage circuit breaker mechanical state prediction device, which is used to implement the high-voltage circuit breaker mechanical state prediction method in the above embodiments, including:

[0132] The data acquisition module 10 is used to collect circuit breaker related data samples in four mechanical states, and the circuit breaker related data includes circuit breaking signal, contact displacement data and coil current data;

[0133] The data preprocessing module 20 is used to preprocess the collected circuit breaker related data samples and divide the data sets;

[0134] A neural network prediction module 30 is used to train the data set using an LSTM neural network, predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and calculate the predicted mechanical characteristic parameters using the prediction results;

[0135] The support vector machine diagnosis module 40 is used to input the predicted mechanical characteristic parameters into the trained SVM model for diagnosis, and use the parallel acceleration module for optimization to obtain the prediction result.

[0136] Based on the foregoing embodiments, the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any of the above methods for predicting the mechanical state of a high-voltage circuit breaker is implemented.

[0137] The present invention also provides a storage medium containing computer executable instructions, and the computer executable instructions are executed by a computer processor to implement the high-voltage circuit breaker mechanical state prediction method in the above embodiment.

[0138] Finally, it should be noted that although the above embodiments have been described in the specification and drawings of this application, this does not limit the scope of patent protection of this application. All technical solutions generated by replacing or modifying equivalent structures or equivalent processes based on the essential concept of this application using the contents recorded in the specification and drawings of this application, as well as directly or indirectly implementing the technical solutions of the above embodiments in other related technical fields, are included in the scope of patent protection of this application.

Claims

1. A method for predicting the mechanical state of a high-voltage circuit breaker, characterized in that: The following steps are involved: Collecting circuit breaker related data samples in four mechanical states, wherein the circuit breaker related data includes a circuit breaker signal, contact displacement data, and coil current data; Preprocess the collected circuit breaker related data samples and divide the data set; The data set is trained using an LSTM neural network to predict the circuit breaker-related data of the HVCB of the circuit breaker in four mechanical states, and the predicted mechanical characteristic parameters are calculated using the prediction results; The predicted mechanical characteristic parameters are input into the trained SVM model for diagnosis and optimized using the parallel acceleration module to obtain the prediction results.

2. The method for predicting the mechanical state of a high-voltage circuit breaker according to claim 1, characterized in that: The collection of circuit breaker related data in four mechanical states includes: Identify test subjects; For the circuit breaking signal, contact displacement data and coil current data of the test object, corresponding data acquisition equipment and acquisition methods are used to measure data under normal conditions, low operating voltage conditions, opening coil core jamming conditions and closing spring fatigue to obtain circuit breaker related data samples.

3. The method for predicting the mechanical state of a high-voltage circuit breaker according to claim 2, characterized in that: The preprocessing of the collected circuit breaker related data samples and the division of the data sets include: Among the circuit breaker related data samples, 80% of the data is used as a training set for model training, and the remaining 20% ​​of the data is used as a test set for verification and test results.

4. The method for predicting the mechanical state of a high-voltage circuit breaker according to claim 3, characterized in that: The preprocessing of the collected circuit breaker related data samples includes: Divide and process the data; The normalization method is used to normalize the data, that is, mapping the data to (0, 1). The formula is as follows, where and are the maximum and minimum values ​​of the variables to be normalized, respectively.

5. The method for predicting the mechanical state of a high-voltage circuit breaker according to claim 1, characterized in that: The data set is trained using an LSTM neural network to predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and the predicted mechanical characteristic parameters are calculated using the prediction results, including: Construct multiple LSTM neural network units corresponding to each other for multiple parameter indicators; Select mean square error (MSE) as the loss function, train the data set in each LSTM neural network unit, and obtain the prediction results; The mechanical characteristic parameters are obtained by calculation based on the prediction results.

6. The method for predicting the mechanical state of a high-voltage circuit breaker according to claim 1, characterized in that: The predicted mechanical characteristic parameters are input into the trained SVM model for diagnosis, and the parallel acceleration module is used for optimization, including: Determine the initial parameters of the support vector machine fault diagnosis model; Determine the parameters of the support vector machine model and optimize them through cross-validation; Import the data set into the trained SVM model for training to obtain the optimal parameters; Based on the SVM diagnosis model with optimal parameters, the mechanical state of HVCB is diagnosed using the predicted HVCB mechanical characteristic parameters; A parallel acceleration module is used to improve diagnostic efficiency. The parallel acceleration module includes using the multi-process runtime library multiprocessing to call multi-core CPU parallel acceleration calculations to improve the diagnostic algorithm.

7. The method for predicting the mechanical state of a high-voltage circuit breaker according to any one of claims 1 to 6, characterized in that: Also includes: Using the same training data set and predicted mechanical property parameters, the random forest, back propagation neural network, and K-nearest neighbor neural network were trained and tested with the results of SVM diagnosis; The diagnostic accuracy of each algorithm was compared to obtain the algorithm with the highest accuracy, and applied to the trained SVM model.

8. A high voltage circuit breaker mechanical state prediction device, characterized in that: A method for predicting a mechanical state of a high-voltage circuit breaker according to any one of claims 1 to 7, comprising: A data acquisition module, used to collect circuit breaker related data samples in four mechanical states, wherein the circuit breaker related data includes a circuit breaker signal, contact displacement data and coil current data; A data preprocessing module is used to preprocess the collected circuit breaker related data samples and divide the data sets; A neural network prediction module, used to train the data set using an LSTM neural network, predict the circuit breaker related data of the HVCB of the circuit breaker in four mechanical states, and calculate the predicted mechanical characteristic parameters using the prediction results; The support vector machine diagnosis module is used to input the predicted mechanical characteristic parameters into the trained SVM model for diagnosis, and use the parallel acceleration module to optimize and obtain the prediction results.

9. 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 method for predicting the mechanical state of a high-voltage circuit breaker according to any one of claims 1 to 7 is implemented.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are executed by a computer processor to implement the high-voltage circuit breaker mechanical state prediction method according to any one of claims 1 to 7.

Images