Fault diagnosis method and device for high-voltage direct-current power transmission system, computer equipment and storage medium

Abstract

The invention relates to a high-voltage direct-current power transmission system fault diagnosis method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring fault recording data of the high-voltage direct-current power transmission system, and preprocessing the fault recording data to obtain a fault sample of the high-voltage direct-current power transmission system; the complexity of the load change rule of the fault sample is lower than that of the load change rule of the fault recording data; performing parameter optimization processing on the classification model to obtain a to-be-trained fault diagnosis model; performing iterative training on a to-be-trained fault diagnosis model based on the historical fault sample to obtain a target fault diagnosis model; and performing fault diagnosis processing on the current data of the high-voltage direct-current power transmission system through the target fault diagnosis model to obtain a fault diagnosis result of the high-voltage direct-current power transmission system. By adopting the method, the accuracy of fault diagnosis of the target fault diagnosis model on the high-voltage direct-current power transmission system can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of fault diagnosis of high-voltage direct current transmission systems, and in particular to a method, apparatus, computer equipment, storage medium and computer program product for fault diagnosis of high-voltage direct current transmission systems. Background Art

[0002] In recent years, as the serious impact of climate change on the global environment has become increasingly prominent, carbon neutrality has gradually become a focus of common concern among countries. Carbon neutrality has promoted the development of global renewable energy, while promoting the large-scale use of clean energy such as hydropower, wind power, and solar energy, laying a solid foundation for achieving (net) zero carbon emissions. In response to the problems of large-scale clean energy grid connection, transmission, and consumption, high voltage direct current (HVDC) transmission technology is more economical, stable, and reliable than traditional AC transmission technology. Therefore, high voltage direct current transmission technology will be a necessary means to further improve the utilization rate of clean energy and assist in the construction of new power systems.

[0003] Since the HVDC transmission system is large in scale, complex in structure and expensive in cost, once a fault occurs, not only the expensive electrical equipment will be damaged, but also the stability of the power system will be seriously endangered, causing immeasurable losses to the power system and load. Therefore, in order to reduce the economic losses caused by system failures, ensure the safe and stable operation of the power system, and improve the stability and reliability of the power system, it is very meaningful to study the fault diagnosis technology of the HVDC transmission system. Summary of the invention

[0004] Based on this, it is necessary to provide a high-voltage direct current transmission system fault diagnosis method, device, computer equipment, computer-readable storage medium and computer program product that can improve the fault diagnosis accuracy of the high-voltage direct current transmission system in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for diagnosing faults in a high-voltage direct current transmission system. The method comprises:

[0006] Acquire fault recording data of a high-voltage direct current transmission system, pre-process the fault recording data, and obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data;

[0007] Perform parameter optimization processing on the classification model to obtain the fault diagnosis model to be trained;

[0008] Based on the fault samples, iteratively training the fault diagnosis model to be trained to obtain a target fault diagnosis model;

[0009] The target fault diagnosis model is used to perform fault diagnosis processing on current data of the high-voltage direct current power transmission system to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0010] In one embodiment, based on the fault sample, the fault diagnosis model to be trained is iteratively trained to obtain a target fault diagnosis model, including:

[0011] Inputting the fault sample into the fault diagnosis model to obtain a predicted fault category of the fault sample; the predicted fault category is used to determine a loss value of the fault diagnosis model;

[0012] Based on the loss value, optimizing the initial model parameters of the fault diagnosis model through a target optimization model to obtain target model parameters;

[0013] Based on the target model parameters, the fault diagnosis model is updated to obtain the target fault diagnosis model.

[0014] In one embodiment, performing parameter optimization processing on the classification model to obtain a fault diagnosis model to be trained includes:

[0015] Determine the initial optimization model;

[0016] Based on lens imaging, the population update method in the initial optimization model is improved to obtain the target optimization model;

[0017] The model parameter updating method of the initial model parameters is updated through the target optimization model to obtain the fault diagnosis model to be trained.

[0018] In one embodiment, preprocessing the fault recording data to obtain the fault sample of the high voltage direct current transmission system includes:

[0019] Performing per-unit normalization processing on the fault recording data to obtain normalized fault recording data;

[0020] Performing variational modal decomposition processing on the normalized fault recording data to obtain modal components of the normalized fault recording data;

[0021] Determining a target modal component to be reconstructed from the modal components according to the sample entropy of the modal components;

[0022] The target modal components are reconstructed and labeled to obtain fault samples of the high-voltage direct current transmission system.

[0023] In one embodiment, before acquiring the fault recording data of the high voltage direct current transmission system, the method further includes:

[0024] Acquire a DC voltage at an outlet of a converter station in the high-voltage DC power transmission system and a sampling time of the DC voltage;

[0025] Performing a derivation process on the DC voltage and the sampling time to obtain a voltage change rate of the high voltage DC power transmission system;

[0026] The voltage change rate is compared with a preset voltage threshold to obtain a comparison result; the comparison result is used to indicate that when the voltage change rate is greater than the preset voltage threshold, the fault recording data of the high voltage direct current transmission system is obtained.

[0027] In one embodiment, obtaining the fault recording data of the high voltage direct current transmission system includes:

[0028] Constructing a transmission model of the high voltage direct current transmission system;

[0029] According to the fault location of the high voltage direct current transmission system, setting sampling points in the transmission model;

[0030] Obtain the fault recording data of the sampling points before and after the fault.

[0031] In a second aspect, the present application also provides a high voltage direct current transmission system fault diagnosis device. The device comprises:

[0032] A sample acquisition module is used to acquire fault recording data of a high-voltage direct current transmission system, and pre-process the fault recording data to obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data;

[0033] The model building module is used to optimize the parameters of the classification model to obtain the fault diagnosis model to be trained;

[0034] A model training module, used for iteratively training the fault diagnosis model to be trained based on the fault samples to obtain a target fault diagnosis model;

[0035] The fault diagnosis module is used to perform fault diagnosis processing on the current data of the high-voltage direct current power transmission system through the target fault diagnosis model to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0036] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0037] Acquire fault recording data of a high-voltage direct current transmission system, pre-process the fault recording data, and obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data;

[0038] Perform parameter optimization processing on the classification model to obtain the fault diagnosis model to be trained;

[0039] Based on the fault samples, iteratively training the fault diagnosis model to be trained to obtain a target fault diagnosis model;

[0040] The target fault diagnosis model is used to perform fault diagnosis processing on current data of the high-voltage direct current power transmission system to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0042] Acquire fault recording data of a high-voltage direct current transmission system, pre-process the fault recording data, and obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data;

[0043] Perform parameter optimization processing on the classification model to obtain the fault diagnosis model to be trained;

[0044] Based on the fault samples, iteratively training the fault diagnosis model to be trained to obtain a target fault diagnosis model;

[0045] The target fault diagnosis model is used to perform fault diagnosis processing on current data of the high-voltage direct current power transmission system to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0046] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0047] Acquire fault recording data of a high-voltage direct current transmission system, pre-process the fault recording data, and obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data;

[0048] Perform parameter optimization processing on the classification model to obtain the fault diagnosis model to be trained;

[0049] Based on the fault samples, iteratively training the fault diagnosis model to be trained to obtain a target fault diagnosis model;

[0050] The target fault diagnosis model is used to perform fault diagnosis processing on current data of the high-voltage direct current power transmission system to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0051] The above-mentioned method, device, computer equipment, storage medium and computer program product for fault diagnosis of high-voltage direct current transmission system obtain the fault recording data of the high-voltage direct current transmission system, pre-process the fault recording data, and obtain the fault samples of the high-voltage direct current transmission system; the complexity of the load change law of the fault sample is lower than the complexity of the load change law of the fault recording data; the classification model is subjected to parameter optimization processing to obtain the fault diagnosis model to be trained; based on the historical fault samples, the fault diagnosis model to be trained is iteratively trained to obtain the target fault diagnosis model; through the target fault diagnosis model, the current data of the high-voltage direct current transmission system is subjected to fault diagnosis processing to obtain the fault diagnosis result of the high-voltage direct current transmission system. By adopting this method, the complexity of the load change law of the data can be reduced by pre-processing the fault recording data, thereby improving the training speed and training effect of the fault diagnosis model, and the model performance of the trained target fault diagnosis model is further improved through parameter optimization processing, thereby improving the accuracy of the target fault diagnosis model in fault diagnosis of the high-voltage direct current transmission system. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A diagram showing an application environment of a method for diagnosing faults in a high-voltage direct current transmission system in an embodiment;

[0053] Figure 2 A schematic diagram of a flow chart of a method for diagnosing a fault in a high-voltage direct current transmission system according to an embodiment;

[0054] Figure 3 A schematic diagram of a flow chart of iterative training steps for a fault diagnosis model to be trained in one embodiment;

[0055] Figure 4 is a schematic diagram of a high voltage direct current transmission model in one embodiment;

[0056] Figure 5 A schematic diagram of a flow chart of a method for diagnosing faults in a high-voltage direct current transmission system in another embodiment;

[0057] Figure 6 A schematic diagram of a flow chart of a method for diagnosing a fault in a high voltage direct current transmission system in yet another embodiment;

[0058] Figure 7A schematic diagram of the structure of a high voltage direct current transmission system fault diagnosis system in one embodiment;

[0059] Figure 8 is a structural block diagram of a high voltage direct current transmission system fault diagnosis device in one embodiment;

[0060] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0061] 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.

[0062] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0063] The high voltage direct current transmission system fault diagnosis method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the high voltage direct current (HVDC) transmission system 101 communicates with the server 102 through the network. The data storage system can store the data that the server 102 needs to process. The data storage system can be integrated on the server 102, or it can be placed on the cloud or other network servers. Among them, the high voltage direct current transmission system 101 is a high-power long-distance direct current transmission system that utilizes the advantages of stable direct current, such as no inductive reactance, no capacitive reactance, and no synchronization problems. The server 102 can be implemented with an independent server or a server cluster composed of multiple servers. The server 102 can schedule the operation of the high voltage direct current transmission system 101 and receive data transmitted by the high voltage direct current transmission system 101.

[0064] In one embodiment, Figure 2 As shown, a method for fault diagnosis of a high voltage direct current transmission system is provided. Figure 1 The server in the example is used to illustrate the following steps:

[0065] Step S201, acquiring fault recording data of the high voltage direct current transmission system, preprocessing the fault recording data, and obtaining fault samples of the high voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data.

[0066] Among them, the fault recording data represents the data recording the changes of various electrical quantities before and after the fault when a fault occurs in the high-voltage direct current transmission system. The fault sample refers to the sample data obtained after the fault recording data is preprocessed and used to train the fault diagnosis model.

[0067] The load variation law refers to information that characterizes the law of how the power load varies over time.

[0068] Specifically, the HVDC system transmits the collected fault recording data to the server. The server preprocesses the fault recording data, marks the corresponding classification labels on the preprocessed fault recording data, and then establishes the fault samples of the HVDC transmission system, and divides the fault samples into a training set and a test set.

[0069] Step S202, performing parameter optimization processing on the classification model to obtain a fault diagnosis model to be trained.

[0070] The fault diagnosis model refers to a model used to perform fault diagnosis and classification on a high-voltage direct current transmission system based on input data.

[0071] Specifically, the server builds a fault diagnosis model based on a classification model, which can be CatBoost, a gradient boosting framework based on symmetric decision trees (oblivious trees) as a base learner with fewer parameters, support for categorical variables and high accuracy. By improving the target optimization model, the model parameter optimization method of the classification model is improved, thereby obtaining a fault diagnosis model to be trained with a better model parameter update method.

[0072] Step S203: iteratively train the fault diagnosis model to be trained based on the fault samples to obtain a target fault diagnosis model.

[0073] Specifically, the server can use the training set of fault samples to iteratively train the fault diagnosis model, and in the process of iterative training of the fault diagnosis model, update the model parameters of the fault diagnosis model according to the target optimization model. The server also uses the test set of fault samples to perform performance testing on the trained fault diagnosis model until the performance test result of the trained fault diagnosis model reaches the preset fault identification accuracy, and the target fault diagnosis model is output.

[0074] Step S204: performing fault diagnosis processing on current data of the high voltage direct current power transmission system through the target fault diagnosis model to obtain a fault diagnosis result of the high voltage direct current power transmission system.

[0075] Specifically, the server uses a target fault diagnosis model to perform fault diagnosis processing on the real-time current data of the high-voltage direct current transmission system, and outputs the fault diagnosis results of the high-voltage direct current transmission system. The fault diagnosis results are used to characterize the fault diagnosis categories of the high-voltage direct current transmission system; the fault diagnosis categories include normal, abnormal, and other categories.

[0076] In the above-mentioned fault diagnosis method for the high-voltage direct current transmission system, the fault recording data of the high-voltage direct current transmission system is obtained, and the fault recording data is preprocessed to obtain the fault samples of the high-voltage direct current transmission system; the complexity of the load change law of the fault sample is lower than the complexity of the load change law of the fault recording data; the classification model is subjected to parameter optimization processing to obtain the fault diagnosis model to be trained; based on the historical fault samples, the fault diagnosis model to be trained is iteratively trained to obtain the target fault diagnosis model; through the target fault diagnosis model, the current data of the high-voltage direct current transmission system is subjected to fault diagnosis processing to obtain the fault diagnosis result of the high-voltage direct current transmission system. By adopting this method, the complexity of the load change law of the data can be reduced by preprocessing the fault recording data, thereby improving the training speed and training effect of the fault diagnosis model, and the model performance of the trained target fault diagnosis model is further improved through parameter optimization processing, thereby improving the accuracy of the target fault diagnosis model in fault diagnosis of the high-voltage direct current transmission system.

[0077] In one embodiment, Figure 3 As shown, the above step S203, based on the fault samples, iteratively trains the fault diagnosis model to be trained to obtain the target fault diagnosis model, which specifically includes the following contents:

[0078] Step S301, inputting the fault sample into the fault diagnosis model to obtain the predicted fault category of the fault sample; the predicted fault category is used to determine the loss value of the fault diagnosis model.

[0079] Specifically, the server can build a classification model based on CatBoost; input the fault sample into the CatBoost classification model, and output the predicted fault category of the fault sample; and calculate the loss value of the fault diagnosis model based on the difference between the predicted fault category and the fault category label of the fault sample.

[0080] Step S302 : Based on the loss value, the initial model parameters of the fault diagnosis model are optimized by the target optimization model to obtain the target model parameters.

[0081] The initial model parameters refer to the model parameters of the fault diagnosis model before the parameter update in this round. The target model parameters refer to the optimal model parameters obtained after the parameter optimization process.

[0082] Specifically, before training, the server can obtain the initial model parameters by initializing the model parameters of the fault diagnosis model. During the training process, the target model parameters of the previous round can be used as the initial model parameters of this round. The server iteratively updates the initial model parameters of the fault diagnosis model based on the loss value. During the iterative update of the model parameters of the fault diagnosis model, the server can minimize the loss value of the classification model on the training set as the goal of the optimization process, use the initial model parameters as constraints, and use the target optimization model to optimize the initial model parameters to obtain the target model parameters.

[0083] In practical applications, the target optimization model can be the Improved SparrowSearch Algorithm (ISSA), and the fault diagnosis model can be ISSA-CatBoost. The process of optimizing the initial model parameters through the target optimization model includes:

[0084] (1) Initializing algorithm parameters: The server first initializes the parameters of the improved sparrow search algorithm, including the group size, the maximum number of iterations, the location range of the individual, and the proportion of discoverers, predators, and alerters in the population.

[0085] (2) Initializing the population: The initial sparrow population is obtained through the improved sparrow search algorithm, where each sparrow represents a possible solution for the model parameters of the classification model.

[0086] (3) Evaluating fitness: The fitness of each sparrow is calculated through the loss value of the classification model to evaluate its performance in the current environment.

[0087] (4) Marking individuals: Based on the fitness value, the sparrow with the best fitness is selected as the discoverer; the remaining sparrows are divided into predators and alerters according to their distance from the discoverer.

[0088] (5) Update location:

[0089] The finder's position update formula is:

[0090]

[0091] In the formula, x(t) i,j represents the position of the i-th sparrow in the j-dimension at the t-th iteration, α∈(0,1] is a random number, and T max is the maximum number of iterations, Q is a normally distributed random number, L is a vector of all 1s, R is the warning value, and ST is the safety threshold.

[0092] Furthermore, the predator's position update formula is:

[0093]

[0094] In the formula, x worst is the current worst position, n is the total number of sparrows, and A is a matrix with random elements ±1.

[0095] Furthermore, the position update formula of the alerter is:

[0096]

[0097] In the formula, x best is the current optimal position, β is the step size control parameter of the normal distribution random number, K∈[-1,1] is the random number, f i is the fitness value of the current sparrow individual, f g is the current global optimal fitness value, f w is the current worst fitness value, and ε is a constant to avoid the denominator being 0.

[0098] (6) Termination condition: If it is detected that the loss value of the classification model on the training set meets the preset conditions (such as minimization), the iterative algorithm is stopped and the sparrow position corresponding to the best fitness is output as the target model parameter; otherwise, the iterative algorithm continues.

[0099] Step S303: updating the parameters of the fault diagnosis model based on the target model parameters to obtain a target fault diagnosis model.

[0100] Specifically, the initial model parameters of the fault diagnosis model are updated to the target model parameters to obtain a trained fault diagnosis model; when a preset training termination condition is reached, the training is terminated to obtain the target fault diagnosis model.

[0101] In this embodiment, during the model training process, based on the loss value of the fault diagnosis model, the initial model parameters of the fault diagnosis model are optimized through the target optimization model to obtain the target model parameters; finally, the target fault diagnosis model is trained to effectively improve the model performance of the trained target fault diagnosis model, thereby improving the accuracy of the target fault diagnosis model in fault diagnosis of the high voltage direct current transmission system.

[0102] In one embodiment, in step S202, the classification model is subjected to parameter optimization processing to obtain a fault diagnosis model to be trained, which specifically includes the following contents: determining an initial optimization model; based on lens imaging, improving the population update method in the initial optimization model to obtain a target optimization model; and updating the model parameter update method of the initial model parameters through the target optimization model to obtain the fault diagnosis model to be trained.

[0103] Specifically, the server can use the Sparrow Search Algorithm (SSA) as the initial optimization model. However, the Sparrow Search Algorithm uses an initialization method to generate the initial population. In order to improve the quality of the initial population and speed up the convergence of the optimization model, the server can improve the Sparrow Search Algorithm, thereby using the improved Sparrow Search Algorithm as the target optimization model. The improvement process of the population update method of the Sparrow Search Algorithm is as follows: the server first determines the optimization range of the parameters, and randomly initializes the positions xi,j of n sparrows in the optimization range as the initial population positions. Then, the reverse learning formula of the lens imaging principle is used to generate the refractive reverse population x* i,j:

[0104]

[0105] Where x* i,j is the refraction reverse position of xi,j, and a j 、b j are the minimum and maximum values ​​of the j-th dimension of the search space, respectively, and k is the scaling factor.

[0106] Furthermore, the initial population and the reflected reverse population are merged, sorted according to the rise and fall of fitness values, and the first n sparrow individuals with the highest fitness values ​​are selected as the new initial population, and then the server obtains the improved sparrow search algorithm ISSA. The server uses the improved sparrow search algorithm ISSA to perform parameter optimization on the classification model CatBoost and obtains the fault diagnosis model ISSA-CatBoost.

[0107] In this embodiment, by improving the population update method in the initial optimization model, the target optimization model is obtained, which can generate the initial population more accurately rather than randomly, effectively improving the quality of the initial population. The target optimization model is then used to replace the original optimization method of the classification model, thereby improving the parameter optimization effect of the fault diagnosis model and greatly improving the training effect of the fault diagnosis model.

[0108] In one embodiment, the above step S201, pre-processing the fault recording data to obtain a fault sample of the high-voltage direct current transmission system, specifically includes the following contents: performing per-unit processing on the fault recording data to obtain the per-unit fault recording data; performing variational modal decomposition processing on the per-unit fault recording data to obtain modal components of the per-unit fault recording data; determining a target modal component to be reconstructed from the modal component according to the sample entropy of the modal component; and performing reconstruction processing and labeling processing on the target modal component to obtain a fault sample of the high-voltage direct current transmission system.

[0109] Among them, standardization is a relative unit system used to calculate the representation of various physical quantities and parameters in the high-voltage direct current transmission system to eliminate the influence of the order of magnitude and units of the physical quantities and parameters, which is convenient for comparison and analysis.

[0110] Specifically, the server performs per-unit normalization processing on the fault recording data to obtain the normalized fault recording data; the calculation process of the per-unit normalization processing is as follows:

[0111]

[0112] Where x i * is the per-unit value of the fault sample data; x i is the unprocessed fault sample data; x N It is the rated voltage of the high voltage DC transmission system.

[0113] Furthermore, the server performs variational mode decomposition (VMD) processing on the normalized fault recording data to obtain the modal components of the normalized fault recording data; calculates the sample entropy (SE) of each modal component respectively; determines the target modal component to be reconstructed from the modal component according to the sample entropy of the modal component, reconstructs and labels the target modal component, and obtains the fault sample of the high voltage direct current transmission system.

[0114] For example, the server can reconstruct the high-frequency components with sample entropy values ​​higher than 0.5 and the low-frequency components with sample entropy values ​​lower than 0.5 respectively to obtain high-frequency reconstructed components and low-frequency reconstructed components; then label the high-frequency reconstructed components and the low-frequency reconstructed components with corresponding classification labels, and obtain fault samples of the high-voltage direct current transmission system after labeling.

[0115] In this embodiment, the fault recording data is normalized to obtain normalized fault recording data; the normalized fault recording data is subjected to variational modal decomposition to obtain modal components of the normalized fault recording data; according to the sample entropy of the modal components, the target modal components to be reconstructed are determined from the modal components; the target modal components are reconstructed and labeled to obtain fault samples of the high-voltage direct current transmission system, which greatly reduces the complexity of the load variation law of the data, thereby improving the training speed and training effect of the fault diagnosis model.

[0116] In one embodiment, in step S201, before obtaining the fault recording data of the high-voltage direct current transmission system, the method further includes: obtaining the DC voltage at the converter station outlet in the high-voltage direct current transmission system and the sampling time of the DC voltage; performing derivative processing on the DC voltage and the sampling time to obtain the voltage change rate of the high-voltage direct current transmission system; comparing the voltage change rate with a preset voltage threshold to obtain a comparison result; the comparison result is used to indicate that when the voltage change rate is greater than the preset voltage threshold, the fault recording data of the high-voltage direct current transmission system is obtained.

[0117] The server can collect the operating data of the high-voltage direct current transmission system, and use the operating data to calculate the voltage change rate, so as to determine whether the conditions for executing subsequent steps are met based on the voltage change rate. Specifically, the DC voltage and the sampling time are derived to obtain the voltage change rate of the high-voltage direct current transmission system; the voltage change rate is compared with the preset voltage threshold to obtain a comparison result. If the voltage change rate is greater than the preset voltage threshold, the fault recording data of the high-voltage direct current transmission system is obtained; otherwise, the operating data of the high-voltage direct current transmission system is continued to be monitored, and its voltage change rate is calculated whether it is greater than the preset voltage threshold. In practical applications, the process of comparing the voltage change rate with the preset voltage threshold can be expressed as the following formula:

[0118]

[0119] In the formula, u dc is the DC voltage measured at the converter station exit; t is the sampling time, U set is the threshold for starting the judgment.

[0120] In this embodiment, by calculating the voltage change rate of the high-voltage direct current transmission system and then detecting whether the voltage change rate is greater than a preset voltage threshold value, it is determined whether to obtain the fault recording data of the high-voltage direct current transmission system. When the voltage of the high-voltage direct current transmission system changes greatly, data can be collected to determine whether a fault exists, rather than detecting the high-voltage direct current transmission system in real time. This reduces the energy consumption of fault diagnosis of the high-voltage direct current transmission system, improves the fault detection accuracy of the high-voltage direct current transmission system, and also improves the detection efficiency of the high-voltage direct current transmission system.

[0121] In one embodiment, the above step S201, obtaining fault recording data of the high voltage direct current transmission system, specifically includes the following contents: constructing a transmission model of the high voltage direct current transmission system; setting sampling points in the transmission model according to the fault location of the high voltage direct current transmission system; and obtaining fault recording data of the sampling points before the fault and the sampling points after the fault.

[0122] Specifically, PSCAD (Power Systems Computer Aided Design, electromagnetic transient simulation software) is used to establish Figure 4 In the HVDC transmission model shown, metallic ground faults are set at fault locations F1, F2 and F3, where F1 is a line fault. The line faults are set from near to far along the entire length of the line, with a total of 400 fault locations, and the step length is 5 km; F2 and F3 are out-of-area faults, and the out-of-area faults include the rectifier side outlet fault and the rectifier side AC system fault, the inverter side outlet fault and the inverter side AC system fault, and the sampling frequency is 10kHz. Select 3 sampling points before the DC voltage fault and 7 sampling points after the fault, a total of 10 sampling points, and obtain the fault recording data of the sampling points before the fault and the fault recording data of the sampling points after the fault.

[0123] In this embodiment, by setting sampling points in the transmission model according to the fault location of the high-voltage direct current transmission system, and then obtaining fault recording data of the sampling points before the fault and the sampling points after the fault, the collection of fault recording data at different fault locations in the high-voltage direct current transmission system is realized, providing rich and reliable data for subsequent model training.

[0124] In one embodiment, Figure 5 As shown in FIG. 1 , another fault diagnosis method for a high voltage direct current transmission system is provided. The method is applied to Figure 1 The server in the example is used to illustrate the following steps:

[0125] Step S501, obtaining fault recording data of a high voltage direct current transmission system.

[0126] Step S502, performing per-unit normalization processing on the fault recording data to obtain the normalized fault recording data; performing variational modal decomposition processing on the normalized fault recording data to obtain modal components of the normalized fault recording data.

[0127] Step S503: determining a target modal component to be reconstructed from the modal components according to the sample entropy of the modal components.

[0128] Step S504: reconstruct and label the target modal components to obtain fault samples of the high voltage direct current transmission system.

[0129] Step S505, determining an initial optimization model; based on lens imaging, improving the population update method in the initial optimization model to obtain a target optimization model.

[0130] Step S506, updating the model parameter updating method of the initial model parameters through the target optimization model to obtain the fault diagnosis model to be trained.

[0131] Step S507, inputting the fault sample into the fault diagnosis model to obtain the predicted fault category of the fault sample; the predicted fault category is used to determine the loss value of the fault diagnosis model.

[0132] Step S508 , based on the loss value, the initial model parameters of the fault diagnosis model are optimized by the target optimization model to obtain the target model parameters.

[0133] Step S509: based on the target model parameters, update the parameters of the fault diagnosis model to obtain a target fault diagnosis model.

[0134] Step S510: performing fault diagnosis processing on current data of the high voltage direct current power transmission system through a target fault diagnosis model to obtain a fault diagnosis result of the high voltage direct current power transmission system.

[0135] The above-mentioned high-voltage direct current transmission system fault diagnosis method can achieve the following beneficial effects: by preprocessing the fault recording data, the complexity of the load variation law of the data can be reduced, thereby improving the training speed and training effect of the fault diagnosis model, and further improving the model performance of the trained target fault diagnosis model through parameter optimization processing, thereby improving the accuracy of the target fault diagnosis model in fault diagnosis of the high-voltage direct current transmission system.

[0136] In order to more clearly illustrate the high-voltage direct current transmission system fault diagnosis method provided by the embodiment of the present disclosure, the high-voltage direct current transmission system fault diagnosis method is specifically described below with a specific embodiment. Figure 6 As shown in FIG. 1 , another fault diagnosis method for a high voltage direct current transmission system is provided, which can be applied to Figure 1 The server in the system includes the following:

[0137] (1) Collect fault recording data of HVDC transmission system (i.e. high voltage direct current transmission system);

[0138] (2) Normalize the fault recording data and use VMD to obtain the modal components of the fault recording data;

[0139] (3) Calculate the sample entropy of the modal components of the fault recording data, and reconstruct the high-frequency and low-frequency components to obtain the fault sample data;

[0140] (4) Attach corresponding classification labels to the preprocessed fault sample data, establish a fault sample data set, and divide the fault sample data set into a training set and a test set;

[0141] (5) The training set is input into CatBoost for training, and the parameter optimization process of CatBoost is optimized by ISSA algorithm to build a HVDC transmission system fault diagnosis model based on ISSA-CatBoost;

[0142] (6) The ISSA-CatBoost based HVDC transmission system fault diagnosis model is used to perform fault diagnosis of the HVDC transmission system.

[0143] In this embodiment, by preprocessing the fault recording data, the complexity of the load variation law of the data can be reduced, thereby improving the training speed and training effect of the fault diagnosis model. By further improving the model performance of the trained target fault diagnosis model through parameter optimization processing, the accuracy of the target fault diagnosis model in fault diagnosis of the high-voltage direct current transmission system is improved.

[0144] 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.

[0145] In one embodiment, Figure 7 As shown, a high voltage direct current transmission system fault diagnosis system is also provided, the system comprising:

[0146] The data acquisition unit is used to collect fault recording data in the HVDC transmission system.

[0147] The data processing unit uses VMD to obtain the modal components of the fault recording data, calculates the sample entropy of the modal components, reconstructs them according to the size of the sample entropy, and obtains the fault sample data; the corresponding classification labels are attached to the preprocessed fault sample data, a fault sample data set is established, and the fault sample data set is divided into a training set and a test set.

[0148] A model training unit, which trains the CatBoost fault diagnosis model using a training set;

[0149] The parameter optimization unit uses ISSA to intelligently optimize the parameters of the CatBoost fault diagnosis model during the model training process, and builds an ISSA-CatBoost-based HVDC transmission system fault diagnosis model.

[0150] The fault diagnosis unit adopts the HVDC transmission system fault diagnosis model based on ISSA-CatBoost for fault diagnosis of the HVDC transmission system.

[0151] Based on the same inventive concept, the embodiment of the present application also provides a high-voltage direct current transmission system fault diagnosis device for implementing the high-voltage direct current transmission system fault diagnosis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more high-voltage direct current transmission system fault diagnosis device embodiments provided below can refer to the limitations of the high-voltage direct current transmission system fault diagnosis method above, and will not be repeated here.

[0152] In one embodiment, Figure 8 As shown, a high voltage direct current transmission system fault diagnosis device 800 is provided, comprising: a sample acquisition module 801, a model construction module 802, a model training module 803 and a fault diagnosis module 804, wherein:

[0153] The sample acquisition module 801 is used to acquire the fault recording data of the high-voltage direct current transmission system, pre-process the fault recording data, and obtain the fault samples of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data.

[0154] The model building module 802 is used to perform parameter optimization processing on the classification model to obtain a fault diagnosis model to be trained.

[0155] The model training module 803 is used to iteratively train the fault diagnosis model to be trained based on the fault samples to obtain a target fault diagnosis model.

[0156] The fault diagnosis module 804 is used to perform fault diagnosis processing on the current data of the high-voltage direct current power transmission system through a target fault diagnosis model to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

[0157] In one embodiment, the model training module 803 is also used to input the fault sample into the fault diagnosis model to obtain the predicted fault category of the fault sample; the predicted fault category is used to determine the loss value of the fault diagnosis model; based on the loss value, the initial model parameters of the fault diagnosis model are optimized through the target optimization model to obtain the target model parameters; based on the target model parameters, the parameters of the fault diagnosis model are updated to obtain the target fault diagnosis model.

[0158] In one embodiment, the model building module 802 is also used to determine an initial optimization model; based on lens imaging, the population update method in the initial optimization model is improved to obtain a target optimization model; through the target optimization model, the model parameter update method of the initial model parameters is updated to obtain a fault diagnosis model to be trained.

[0159] In one embodiment, the sample acquisition module 801 is also used to perform normalization processing on the fault recording data to obtain normalized fault recording data; perform variational modal decomposition processing on the normalized fault recording data to obtain modal components of the normalized fault recording data; determine target modal components to be reconstructed from the modal components according to the sample entropy of the modal components; perform reconstruction processing and labeling processing on the target modal components to obtain fault samples of the high voltage direct current transmission system.

[0160] In one embodiment, the high-voltage direct current transmission system fault diagnosis device 800 also includes a voltage judgment module, which is used to obtain the DC voltage at the converter station outlet in the high-voltage direct current transmission system and the sampling time of the DC voltage; perform derivative processing on the DC voltage and the sampling time to obtain the voltage change rate of the high-voltage direct current transmission system; compare the voltage change rate with a preset voltage threshold to obtain a comparison result; the comparison result is used to indicate that when the voltage change rate is greater than the preset voltage threshold, the fault recording data of the high-voltage direct current transmission system is obtained.

[0161] In one embodiment, the sample acquisition module 801 is also used to construct a transmission model of the high-voltage direct current transmission system; set sampling points in the transmission model according to the fault location of the high-voltage direct current transmission system; and obtain fault recording data of the sampling points before the fault and the sampling points after the fault.

[0162] Each module in the above-mentioned high-voltage direct current transmission system fault diagnosis device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0163] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig. 9As shown. 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 data such as fault recording data, fault samples and target fault diagnosis models. 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 high-voltage direct current transmission system fault diagnosis method is implemented.

[0164] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0165] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0166] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0167] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0168] 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.

[0169] 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.

[0170] 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 diagnosing faults in a high voltage direct current transmission system, characterized in that: The method comprises: Acquire fault recording data of a high-voltage direct current transmission system, pre-process the fault recording data, and obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data; Perform parameter optimization processing on the classification model to obtain the fault diagnosis model to be trained; Based on the fault samples, iteratively training the fault diagnosis model to be trained to obtain a target fault diagnosis model; The target fault diagnosis model is used to perform fault diagnosis processing on current data of the high-voltage direct current power transmission system to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

2. The method according to claim 1, characterized in that The iterative training of the fault diagnosis model to be trained based on the fault sample to obtain a target fault diagnosis model includes: Inputting the fault sample into the fault diagnosis model to obtain a predicted fault category of the fault sample; the predicted fault category is used to determine a loss value of the fault diagnosis model; Based on the loss value, optimizing the initial model parameters of the fault diagnosis model through a target optimization model to obtain target model parameters; Based on the target model parameters, the fault diagnosis model is updated to obtain the target fault diagnosis model.

3. The method according to claim 2, characterized in that The method of performing parameter optimization processing on the classification model to obtain a fault diagnosis model to be trained includes: Determine the initial optimization model; Based on lens imaging, the population update method in the initial optimization model is improved to obtain the target optimization model; The model parameter updating method of the initial model parameters is updated through the target optimization model to obtain the fault diagnosis model to be trained.

4. The method according to claim 1, characterized in that: The preprocessing of the fault recording data to obtain the fault sample of the high voltage direct current transmission system includes: Performing per-unit normalization processing on the fault recording data to obtain normalized fault recording data; Performing variational modal decomposition processing on the normalized fault recording data to obtain modal components of the normalized fault recording data; Determining a target modal component to be reconstructed from the modal components according to the sample entropy of the modal components; The target modal components are reconstructed and labeled to obtain fault samples of the high-voltage direct current transmission system.

5. The method according to claim 1, characterized in that Before obtaining the fault recording data of the high-voltage direct current transmission system, it also includes: Acquire a DC voltage at an outlet of a converter station in the high-voltage DC power transmission system and a sampling time of the DC voltage; Performing a derivation process on the DC voltage and the sampling time to obtain a voltage change rate of the high voltage DC power transmission system; The voltage change rate is compared with a preset voltage threshold to obtain a comparison result; the comparison result is used to indicate that when the voltage change rate is greater than the preset voltage threshold, the fault recording data of the high voltage direct current transmission system is obtained.

6. The method according to claim 5, characterized in that The obtaining of the fault recording data of the high voltage direct current transmission system comprises: Constructing a transmission model of the high voltage direct current transmission system; According to the fault location of the high voltage direct current transmission system, setting sampling points in the transmission model; Obtain the fault recording data of the sampling points before and after the fault.

7. A fault diagnosis device for a high voltage direct current transmission system, characterized in that: The device comprises: A sample acquisition module is used to acquire fault recording data of a high-voltage direct current transmission system, and pre-process the fault recording data to obtain a fault sample of the high-voltage direct current transmission system; the complexity of the load variation law of the fault sample is lower than the complexity of the load variation law of the fault recording data; The model building module is used to optimize the parameters of the classification model to obtain the fault diagnosis model to be trained; A model training module, used for iteratively training the fault diagnosis model to be trained based on the fault samples to obtain a target fault diagnosis model; The fault diagnosis module is used to perform fault diagnosis processing on the current data of the high-voltage direct current power transmission system through the target fault diagnosis model to obtain a fault diagnosis result of the high-voltage direct current power transmission system.

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