Lightning arrester leakage current monitoring method and system, electronic equipment and storage medium

Abstract

The invention discloses a lightning arrester leakage current monitoring method and system, electronic equipment and a storage medium, and relates to the technical field of lightning arrester leakage current measurement, and the method comprises the steps: measuring the standard true value of the leakage current when a lightning arrester operates stably under the condition of eliminating environmental data interference; acquiring multiple groups of historical environment data of the lightning arrester and historical leakage current measured under the historical environment data, and calculating correlation between each group of historical environment data and the corresponding historical leakage current; taking the historical environment data of which the correlation is higher than a preset threshold value and the corresponding historical leakage current as input data, taking the standard true value of the leakage current as output data, and performing model training by adopting an NARX neural network to obtain a leakage current prediction model; the current environment data of the lightning arrester are input into the leakage current prediction model, the environment normalization reference value of the current leakage current is obtained, the influence of environment factors on leakage current monitoring is eliminated, and the accuracy of leakage current monitoring is improved.

Description

Technical Field

[0001] This invention belongs to the field of surge arrester leakage current measurement technology, and relates to a surge arrester leakage current monitoring method, system, electronic device and storage medium. Background Technology

[0002] With the rapid development of power systems, the construction of new power systems has become one of the key tasks in the industry. Among these, the digitalization of power systems is a crucial characteristic of this transformation, requiring power facilities to possess a higher level of intelligence to achieve more efficient and safer operation. Under the guidance of the State Grid Equipment Department, the construction of smart substations is progressing steadily, with the adoption of intelligent online monitoring devices for various main equipment becoming an important means of improving the intelligence level of substations.

[0003] Surge arresters are crucial protective devices in power systems, and their operational status significantly impacts the safe and stable operation of the power system. Therefore, condition monitoring of surge arresters and timely acquisition of their operational data are of great importance for accident prevention and ensuring system safety. In surge arrester condition monitoring, the measurement of the total current is a fundamental and critical step. The application of digital remote meters makes the measurement of the total current of surge arresters more convenient and accurate, providing strong support for surge arrester condition assessment.

[0004] However, in practical applications, the measurement of the total current of surge arresters is affected by various factors, among which meteorological factors are particularly significant. Studies in laboratory environments show that the total current and its resistive current component of zinc oxide surge arresters increase with increasing temperature, and increased humidity also leads to an increase in the total current. While these studies reveal the influence of meteorological factors on the total current of surge arresters to some extent, the laboratory environment differs fundamentally from actual field operating conditions. Laboratory studies of valve plate temperature characteristics do not consider the influence of the current in the arrester casing, and temperature characteristic tests of the entire surge arrester cannot fully simulate the complex conditions of real-world operation.

[0005] Furthermore, while online monitoring can acquire the total current data of surge arresters in real time and analyze the trend of total current variation with temperature and humidity, in-depth research is still lacking on the specific impact of various meteorological factors such as precipitation on the total current measurement. This results in the surge arrester's total current measurement data often being interfered with by meteorological factors in practical applications, making it difficult to accurately reflect the true leakage current of the surge arrester's varistor.

[0006] Therefore, reducing the interference of meteorological factors on the total current measurement in the online monitoring of surge arresters has become an urgent problem to be solved. Researching effective methods to eliminate or reduce these interference factors is of great significance for improving the accuracy of surge arrester condition monitoring and ensuring the safe and stable operation of the power system. Summary of the Invention

[0007] The purpose of this invention is to solve the technical problem that the measurement data of the total current of surge arresters in the prior art is often affected by meteorological factors and is difficult to accurately reflect the true leakage current of the surge arrester varistor, and to provide a method and system for monitoring the leakage current of surge arresters.

[0008] To achieve the above objectives, the present invention employs the following technical solution: The first aspect of this invention provides a method for monitoring leakage current of a surge arrester, the method comprising: After eliminating interference from environmental data, the standard true value of leakage current is measured when the surge arrester is operating stably. Environmental data includes temperature, humidity, precipitation, air pressure, and wind speed. Multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data are obtained. After the historical leakage current is preprocessed using the LOF algorithm, the correlation between each set of historical environmental data and its corresponding historical leakage current is calculated. Historical environmental data with a correlation higher than a preset threshold and their corresponding historical leakage currents are used as input data, and the standard true value of the leakage current is used as output data. The NARX neural network is used to train the model to obtain the leakage current prediction model. Input the current environmental data of the surge arrester into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

[0009] Furthermore, in some embodiments of this disclosure, calculating the correlation between each set of historical environmental data and its corresponding historical leakage current includes: calculating the mean of the historical environmental data and the mean of the historical leakage current; based on the mean of the historical environmental data and the mean of the historical leakage current, calculating the covariance of the historical environmental data and the historical leakage current, the standard deviation of the historical environmental data, and the standard deviation of the historical leakage current respectively; using the product of the standard deviation of the historical environmental data and the standard deviation of the historical leakage current as the denominator, using the covariance as the numerator, and dividing the numerator by the denominator to obtain the correlation coefficient between the historical environmental data and the historical leakage current.

[0010] In some embodiments of this disclosure, the network structure of the NARX neural network includes an open-loop structure and a closed-loop structure. Training the model using the NARX neural network to obtain a leakage current prediction model includes: converting input and output data into sample pairs required by the open-loop NARX neural network based on a set delay order; using the error between the predicted value of the leakage current and the standard true value of the leakage current as a loss function; inputting the sample pairs into the open-loop NARX neural network, outputting the predicted value of the leakage current; adjusting the parameters of the open-loop NARX neural network with the goal of minimizing the loss function until a preset model convergence condition is met, thereby obtaining the leakage current prediction model.

[0011] In some embodiments of this disclosure, the method further includes: converting the network structure of the leakage current prediction model into a closed-loop structure; using the feedback mechanism of the closed-loop structure to feed back the predicted value of the leakage current to the input of the leakage current prediction model; and combining the predicted value of the leakage current to perform multi-step prediction through the leakage current prediction model to obtain the environmental normalized reference value of the leakage current at future times.

[0012] In some embodiments of this disclosure, the method further includes: obtaining the changing trend of the environmental normalized reference value of the leakage current based on the current environmental normalized reference value and the future environmental normalized reference value of the leakage current; and determining whether the operation status of the surge arrester is normal based on the changing trend.

[0013] In some embodiments of this disclosure, the LOF algorithm is used to preprocess historical leakage current data, including: for each data point of historical leakage current, calculating the local reachability density of the data point; obtaining the local outlier factor of the data point by calculating the average of the ratio of the local reachability density of all points in the neighborhood of the data point to the local reachability density of the data point; marking data points with local outlier factors exceeding a preset threshold as abnormal data points and removing abnormal data points; and filling the gaps after removing abnormal data points using piecewise cubic spline interpolation.

[0014] A second aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor performs the above-described surge arrester leakage current monitoring method.

[0015] A third aspect of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described surge arrester leakage current monitoring method.

[0016] A fourth aspect of the present invention provides a computer program product, the computer program product including computer instructions, the computer instructions instructing a computer to execute the above-described surge arrester leakage current monitoring method.

[0017] The fifth aspect of the present invention provides a surge arrester leakage current monitoring system, comprising: The measurement module is used to measure the standard true value of the leakage current of the surge arrester during stable operation, after eliminating environmental data interference. Environmental data includes temperature, humidity, precipitation, air pressure and wind speed. The data acquisition and preprocessing module is used to acquire multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data. After preprocessing the historical leakage current data using the LOF algorithm, the correlation between each set of historical environmental data and its corresponding historical leakage current is calculated. The training module is used to take historical environmental data with a correlation higher than a preset threshold and their corresponding historical leakage current as input data, and the standard true value of leakage current as output data. The NARX neural network is used to train the model to obtain the leakage current prediction model. The prediction module is used to input the current environmental data of the surge arrester into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a method for monitoring leakage current in surge arresters. By accurately measuring the true value of leakage current under the operating environment of the surge arrester and collecting current environmental data as well as corresponding historical environmental data, the method can construct a highly correlated and representative training set. This training set, combined with the powerful nonlinear fitting capability of the neural network model, enables the prediction model to learn the complex relationship between environmental parameters and leakage current, thereby significantly improving the accuracy of prediction. During the training process, the method of this invention particularly emphasizes the correlation analysis between environmental data and leakage current, using only highly correlated data as the training set. This step effectively eliminates environmental factors that are not closely related to leakage current or contain noise, reducing interference during model prediction and further improving the reliability of prediction results. This method can acquire current environmental data in real time and input it into the trained prediction model to achieve instant prediction of leakage current. By using the true value of leakage current as the benchmark, the influence of environmental factors is effectively eliminated, thus enabling high-precision monitoring of leakage current values. This high-precision, real-time, and automated characteristic allows the monitoring system to quickly respond to environmental changes and adjust monitoring strategies in a timely manner, providing a strong guarantee for the safe and stable operation of the power system. By continuously monitoring the leakage current of surge arresters and analyzing environmental factors, the method of this invention can provide a scientific basis for the maintenance and management of surge arresters. For example, when it is predicted that the leakage current may exceed the safe range, the system can issue an early warning to avoid potential safety hazards. At the same time, long-term data accumulation can also be used to analyze the aging trend of surge arresters, providing decision support for equipment replacement and upgrading.

[0019] Furthermore, introducing the coefficient of determination (COD) as an evaluation metric not only considers the magnitude of the prediction error but also its proportion relative to the overall data variability, providing a more comprehensive and accurate assessment of model performance. Combining RMSE and MAE forms a complete error analysis system, evaluating the model from different dimensions to ensure optimal prediction accuracy and stability. During model training, by continuously optimizing model parameters, RMSE and MAE gradually decrease, while the COD gradually increases until the preset accuracy requirements are met. This process not only ensures the model's high-precision prediction capability but also provides a scientific basis for subsequent model optimization. When model performance no longer significantly improves, the model can be considered to have reached its optimal state, avoiding the risk of overfitting. Through the introduction and optimization of the above error evaluation metrics, the predictive reliability of the surge arrester leakage current monitoring method has been significantly improved. This means that in practical applications, this method can more accurately predict the leakage current of surge arresters, providing strong support for the safety monitoring and early warning of power systems. Simultaneously, high-precision prediction results also help reduce unnecessary maintenance and replacement costs, improving the economy and sustainability of power systems.

[0020] Furthermore, this invention provides a surge arrester leakage current monitoring system. By integrating a data acquisition module, the system can capture key parameters in the surge arrester's operating environment in real time, such as temperature, humidity, precipitation, air pressure, and wind speed, and input this data into a trained leakage current prediction model. The model, built on a neural network, can learn the complex relationship between environmental parameters and leakage current, thereby achieving high-precision prediction. By continuously monitoring the surge arrester's leakage current, the system can provide maintenance personnel with accurate data support, helping them to formulate more precise maintenance plans. This not only reduces unnecessary power outages but also extends the surge arrester's lifespan and reduces maintenance costs. Accurate leakage current monitoring helps to promptly detect and handle surge arrester anomalies, avoiding energy waste and environmental pollution caused by equipment failure. Simultaneously, by optimizing maintenance cycles and strategies, the system can further save energy, achieving the goal of energy conservation and emission reduction. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart of a surge arrester leakage current monitoring method provided by the present invention; Figure 2An example diagram of a closed-loop structure of a NARX neural network provided by the present invention; Figure 3 A block diagram of a surge arrester leakage current monitoring system provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and marked in the accompanying drawings can generally be arranged and designed in various different configurations.

[0024] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0025] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0026] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0027] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0028] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0029] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 This invention discloses a method for monitoring leakage current of a surge arrester, comprising the following steps: S101, after eliminating environmental data interference, measures the standard true value of leakage current when the surge arrester is operating stably. Environmental data includes temperature, humidity, precipitation, air pressure and wind speed.

[0030] In the embodiments of this disclosure, five meteorological environmental data points—temperature, humidity, precipitation, air pressure, and wind speed—are selected as external influencing factors during the long-term operation of the surge arrester. These data points are then matched with the leakage current measurement values ​​of the surge arrester to form a training dataset for training the leakage current prediction model.

[0031] In the embodiments of this disclosure, in order to train the leakage current prediction model and obtain the leakage current predicted after eliminating environmental influencing factors, this embodiment introduces the standard true value of leakage current. That is, an experimental model is built in the laboratory that is consistent with the working voltage level and line parameters of the surge arrester selected by the measurement value, other external environmental influencing factors are eliminated, and the leakage current value when the surge arrester is working normally and stably is accurately measured by precision instruments. This value is then introduced into the training model as a benchmark value for training.

[0032] S102: Obtain multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data. After preprocessing the historical leakage current using the LOF algorithm, calculate the correlation between each set of historical environmental data and its corresponding historical leakage current.

[0033] In some embodiments of this disclosure, the LOF algorithm is used to preprocess historical leakage current data, including: for each data point of historical leakage current, calculating the local reachability density of the data point; obtaining the local outlier factor of the data point by calculating the average of the ratio of the local reachability density of all points in the neighborhood of the data point to the local reachability density of the data point; marking data points with local outlier factors exceeding a preset threshold as abnormal data points and removing abnormal data points; and filling the gaps after removing abnormal data points using piecewise cubic spline interpolation.

[0034] Specifically, obtain the preset parameter k, and use the currently preprocessed data points as observation points. The LOF algorithm calculates the observation points. The relative density deviation is used to determine whether a point is abnormal. For observation points... Its local density can reach It can be expressed as follows: In the formula, It is a point of Nearest neighbor set It is a point With point The reachable distance between points Representative point of Distance to each point in the neighborhood. Observation point. The local outlier (LOF) value is defined as the point of The local outlier factor is calculated as the average ratio of the local reachability density of all points within its neighborhood to its own local reachability density. The data points were measured The greater the density difference between a point and its surrounding points, the more significant the difference. It might be an outlier. Representative point The local reachable density. Due to the removal of outlier data, gaps appeared in the original data. Cubic spline interpolation was used to fill in the missing leakage current data.

[0035] In some embodiments of this disclosure, calculating the correlation between each set of historical environmental data and its corresponding historical leakage current includes: calculating the mean of the historical environmental data and the mean of the historical leakage current; based on the mean of the historical environmental data and the mean of the historical leakage current, calculating the covariance of the historical environmental data and the historical leakage current, the standard deviation of the historical environmental data, and the standard deviation of the historical leakage current respectively; using the product of the standard deviation of the historical environmental data and the standard deviation of the historical leakage current as the denominator, using the covariance as the numerator, and dividing the numerator by the denominator to obtain the correlation coefficient between the historical environmental data and the historical leakage current.

[0036] Specifically, during model training, to reduce the risk of overfitting and improve the model's accuracy and interpretability, this embodiment uses the Pearson correlation coefficient as an indicator to assess the linear correlation between various meteorological data and leakage current. The formula for the Pearson correlation coefficient is as follows: In the formula, X and Y represent each set of historical environmental data and its corresponding historical leakage current, respectively. For parameters and covariance, and They represent and The standard deviation of the correlation coefficient. The closer the value is to 1 or -1, the better the parameter is. and The stronger the positive or negative linear correlation between them, the better; conversely, the weaker the positive or negative linear correlation, the better. The closer the value is to 0, the more it indicates... and It does not have a linear correlation.

[0037] S103, taking historical environmental data with a correlation higher than a preset threshold and their corresponding historical leakage current as input data, and the standard true value of leakage current as output data, the NARX neural network is used to train the model to obtain the leakage current prediction model.

[0038] In some embodiments of this disclosure, the network structure of the NARX neural network includes an open-loop structure and a closed-loop structure. Training the model using the NARX neural network to obtain a leakage current prediction model includes: converting input and output data into sample pairs required by the open-loop NARX neural network based on a set delay order; using the error between the predicted value of the leakage current and the standard true value of the leakage current as a loss function; inputting the sample pairs into the open-loop NARX neural network, outputting the predicted value of the leakage current; adjusting the parameters of the open-loop NARX neural network with the goal of minimizing the loss function until a preset model convergence condition is met, thereby obtaining the leakage current prediction model.

[0039] Specifically, to evaluate the model's predictive performance, training is stopped when the error between the predicted leakage current and the standard true value of the leakage current stabilizes within a certain range; the root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²) are used as the evaluation metrics. 2 As evaluation indicators, the calculation formulas for the three are as follows: ; ; ; In the formula, To verify the amount of data in the set, This is the standard true value of the leakage current. This is the predicted value of leakage current. This represents the average value of the actual leakage current.

[0040] It is understood that the present invention inputs the leakage current dataset of each surge arrester and the corresponding standard true value into the NARX neural network and combines it with relevant surge arrester operating environment factor parameters. Through training with a large number of samples, a leakage current prediction model is obtained. This leakage current prediction model can show the influence relationship between meteorological environmental factors and surge arrester leakage current.

[0041] In the embodiments of this disclosure, the NARX neural network is a type of recurrent neural network (RNN). This model introduces a delay element and a feedback mechanism, enhancing the model's ability to remember historical data, thus making it suitable for nonlinear time series prediction problems. Depending on whether it has a feedback element, the NARX neural network has two structures: open-loop and closed-loop. The open-loop NARX neural network does not have a feedback element and relies solely on the historical states of the input and output to predict the current value of the output. Its mathematical expression is as follows: In the formula, yes The output value of the time step, i.e., the predicted value of the leakage current, It is a function representing the nonlinear relationship of the model. Indicates the past The output value at each time step, i.e., the historical leakage current. Indicates the past The input values ​​at each time step, i.e., historical environmental data, reflect the impact of external factors on the output.

[0042] In some embodiments of this disclosure, the method further includes: converting the network structure of the leakage current prediction model into a closed-loop structure; using the feedback mechanism of the closed-loop structure to feed back the predicted value of the leakage current to the input of the leakage current prediction model; and combining the predicted value of the leakage current to perform multi-step prediction through the leakage current prediction model to obtain the environmental normalized reference value of the leakage current at future times.

[0043] Specifically, the closed-loop structure of the NARX neural network is as follows: Figure 2 As shown, x(t) represents Time step input variables, i.e., environmental data represent The output value of the time step feedback, i.e. the predicted value of the leakage current, This indicates the input delay order, i.e., how many past time steps of external input are used as the input. This indicates the output delay order, i.e., how many past time steps of predicted output are used as input. In the hidden layer, w represents the weight matrix from the input layer to the hidden layer, carrying historical information, h1-h... n The output / state of the hidden layer carries historical information. In the output layer, b is the bias term, and W is the weight matrix from the hidden layer to the output layer. f1 f1 is the activation function of the hidden layer, and f2 is the activation function of the output layer. The closed-loop NARX neural network introduces a feedback loop (connected by the dashed line in the diagram) to the open-loop network. This loop allows the network's output to be fed back to the input, using historical predictions to predict the next time step. The mathematical expression for the closed-loop network is: ; In the formula, yes The predicted value of the time step, i.e., the environmentally normalized reference value of the leakage current, Indicates the past The predicted value at each time step, i.e. the historical predicted value of leakage current.

[0044] In the embodiments of this disclosure, the open-loop network is first trained using sample values, leveraging its simple structure and fast computation speed to predict leakage current. Then, the trained NARX neural network is transformed into a closed-loop structure. In this case, the model no longer predicts based on historical leakage current data, but instead relies on historical predicted values ​​of the leakage current to predict future values, using a closed-loop NARX model to predict the test set data.

[0045] It should be noted that by selecting environmental factors corresponding to the training process as external influencing factors, collecting and summarizing local environmental meteorological data during the actual use of the surge arrester, and incorporating them into the leakage current prediction model for processing, the predicted leakage current can be obtained. This predicted leakage current eliminates the influence of environmental factors.

[0046] S104. Input the current environmental data of the surge arrester into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

[0047] In some embodiments of this disclosure, the method further includes: obtaining the changing trend of the environmental normalized reference value of the leakage current based on the current environmental normalized reference value and the future environmental normalized reference value of the leakage current; and determining whether the operation status of the surge arrester is normal based on the changing trend.

[0048] Specifically, the leakage current prediction model trained by this invention can process the leakage current of surge arresters collected under corresponding meteorological conditions after knowing the operating environmental parameters of the surge arrester, thus eliminating the influence of meteorological factors on the leakage current measurement value. That is, by training on the surge arrester leakage current and environmental parameter data, a reference value (i.e., the predicted leakage current value) of the leakage current under the same external environmental parameters is obtained. Taking the standard surge arrester operating environmental parameters as an example, if the JIS-8703 standard environmental conditions are adopted, after reducing meteorological interference, the changing trend of the reference value of the surge arrester leakage current is obtained, which is used to determine whether the surge arrester is operating normally during the monitoring period and whether aging has occurred.

[0049] One embodiment of the present invention, such as Figure 3 As shown, a surge arrester leakage current monitoring system is provided, comprising: Measurement module 310 is used to measure the standard true value of leakage current of surge arrester during stable operation, after eliminating environmental data interference. Environmental data includes temperature, humidity, precipitation, air pressure and wind speed. The data acquisition and preprocessing module 320 is used to acquire multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data. After preprocessing the historical leakage current data using the LOF algorithm, the correlation between each set of historical environmental data and its corresponding historical leakage current is calculated. Training module 330 is used to take historical environmental data with a correlation higher than a preset threshold and its corresponding historical leakage current as input data, and the standard true value of leakage current as output data. The model is trained using NARX neural network to obtain leakage current prediction model. The prediction module 340 is used to input the current environmental data of the surge arrester into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

[0050] It should be noted that the specific implementation methods of the embodiments disclosed herein are different from those of the embodiments in this paper. Figure 1 The principle of the embodiments shown is the same, and will not be repeated here.

[0051] One embodiment of the present invention provides an electronic device including a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used for the operation of a surge arrester leakage current monitoring method.

[0052] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). It should be noted that more specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CDROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0053] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium can also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0054] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0055] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the surge arrester leakage current monitoring method in the above embodiments.

[0056] One embodiment of the present invention provides a computer program product containing computer instructions for guiding a computer to execute a surge arrester leakage current monitoring method. By integrating functions such as environmental data acquisition, model training, and leakage current prediction, the method achieves effective monitoring of surge arrester leakage current.

[0057] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring leakage current of a surge arrester, characterized in that, The method includes: After eliminating interference from environmental data, the standard true value of the leakage current of the surge arrester during stable operation is measured. The environmental data includes temperature, humidity, precipitation, air pressure, and wind speed. Multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data are obtained. After the historical leakage current is preprocessed using the LOF algorithm, the correlation between each set of historical environmental data and its corresponding historical leakage current is calculated. The historical environmental data with a correlation higher than a preset threshold and their corresponding historical leakage current are used as input data, and the standard true value of the leakage current is used as output data. The NARX neural network is used to train the model to obtain the leakage current prediction model. The current environmental data of the surge arrester is input into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

2. The surge arrester leakage current monitoring method according to claim 1, characterized in that, The calculation of the correlation between each set of historical environmental data and its corresponding historical leakage current includes: Calculate the mean of the historical environmental data and the mean of the historical leakage current; Based on the mean of the historical environmental data and the mean of the historical leakage current, the covariance of the historical environmental data and the historical leakage current, the standard deviation of the historical environmental data, and the standard deviation of the historical leakage current are calculated respectively. The product of the standard deviation of the historical environmental data and the standard deviation of the historical leakage current is used as the denominator, and the covariance is used as the numerator. The correlation coefficient between the historical environmental data and the historical leakage current is obtained by dividing the numerator by the denominator.

3. The method for monitoring leakage current of a surge arrester according to claim 1, characterized in that, The NARX neural network structure includes an open-loop structure and a closed-loop structure. The process of training the model using the NARX neural network to obtain the leakage current prediction model includes: Based on a set delay order, the input data and the output data are converted into sample pairs required by the open-loop NARX neural network; The error between the predicted value of the leakage current and the standard true value of the leakage current is used as the loss function; The sample pairs are input into the open-loop NARX neural network, and the output is the predicted value of the leakage current. With the goal of minimizing the loss function, the parameters of the open-loop NARX neural network are adjusted until the preset model convergence condition is reached, and the leakage current prediction model is obtained.

4. The surge arrester leakage current monitoring method according to claim 3, characterized in that, The method further includes: The network structure of the leakage current prediction model is converted into the closed-loop structure. The predicted value of the leakage current is fed back to the input of the leakage current prediction model using the feedback mechanism of the closed-loop structure. By combining the predicted value of the leakage current, the environmental normalized reference value of the leakage current at future times is obtained through multi-step prediction using the leakage current prediction model.

5. The method for monitoring leakage current of a surge arrester according to claim 4, characterized in that, The method further includes: Based on the current environmental normalized reference value of the leakage current and the future environmental normalized reference value of the leakage current, the changing trend of the environmental normalized reference value of the leakage current is obtained. The operating status of the surge arrester is determined based on the changing trend.

6. The method for monitoring leakage current of a surge arrester according to claim 1, characterized in that, The data preprocessing of the historical leakage current using the LOF algorithm includes: For each data point of the historical leakage current, calculate the local reachability density of the data point; The local outlier factor of the data point is obtained by calculating the average ratio of the local reachability density of all points in the neighborhood of the data point to the local reachability density of the data point. Data points whose local outlier factor exceeds a preset threshold are marked as abnormal data points, and the abnormal data points are removed. Piecewise cubic spline interpolation was used to fill the gaps after removing the outlier data points.

7. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the surge arrester leakage current monitoring method according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the surge arrester leakage current monitoring method according to any one of claims 1-6.

9. A computer program product, the computer program product comprising computer instructions, characterized in that, The computer instructions instruct the computer to execute the surge arrester leakage current monitoring method according to any one of claims 1-6.

10. A surge arrester leakage current monitoring system, characterized in that, include: The measurement module is used to measure the standard true value of the leakage current of the surge arrester during stable operation, after eliminating environmental data interference. The environmental data includes temperature, humidity, precipitation, air pressure and wind speed. The data acquisition and preprocessing module is used to acquire multiple sets of historical environmental data of the surge arrester and the historical leakage current measured under the historical environmental data. After preprocessing the historical leakage current using the LOF algorithm, the correlation between each set of historical environmental data and its corresponding historical leakage current is calculated. The training module is used to take the historical environmental data with a correlation higher than a preset threshold and its corresponding historical leakage current as input data, the standard true value of the leakage current as output data, and use the NARX neural network to train the model to obtain the leakage current prediction model. The prediction module is used to input the current environmental data of the surge arrester into the leakage current prediction model to obtain the environmental normalized reference value of the current leakage current.

Images