Measurement error prediction method and system of current transformer, equipment and storage medium

Abstract

The invention discloses a measurement error prediction method and system for a current transformer, equipment and a storage medium, and the method comprises the steps: firstly collecting the error data of the current transformer under different test conditions to construct a feature data set, then constructing an error prediction model, training the model through the feature data set, and carrying out the prediction of the error of the current transformer. And finally, based on the current current data, the environment data and the trained model, predicting to obtain a measurement error of the current transformer. The model is trained by taking the environmental data as the features, so that the model can learn the influence of the environmental factors on the measurement error, the model prediction error generated by the influence of the environmental factors is reduced, and the prediction accuracy of the measurement error result is improved. And the hyper-parameters of the model are optimized by adopting a chaos guide adaptive algorithm, so that the optimization speed and the optimization precision of the model parameters are improved, the model network can quickly adapt to the influence of environmental factors on the measurement result of the current transformer, and the prediction accuracy is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of current transformer error calibration, and in particular to a current transformer measurement error prediction method and system, electronic equipment, and a computer-readable storage medium. Background Art

[0002] As current transformers are widely used in substations, their accuracy directly affects the monitoring, protection and control of power grids. Accurate measurement of current values ​​by current transformers is an important guarantee for the fairness of electricity price settlement between power grids and users. However, there are often process defects and uncontrollable factors such as wear and aging of equipment in the manufacturing process, which may cause measurement errors in current transformers and affect their measurement accuracy. Therefore, it is necessary to quantitatively detect and correct the errors of current transformers to ensure the safe and stable operation of the power system, improve the operating efficiency and reliability of the power grid, meet the needs of accurate electric energy metering, extend the service life of equipment, and reduce the operating cost of the power system. Existing current transformer error monitoring methods include sensor method and big data method. Among them, the sensor method refers to calibrating the current transformer under test according to the measurement results of the standard current transformer. The sensor method is easily affected by the working environment. Under different temperatures, humidity, electromagnetic interference, etc., the sensor has problems such as sensitivity error, linear error, temperature drift, etc., which makes the monitoring accuracy of the error state low. The big data method refers to the prediction of measurement errors for a period of time in the future based on the historical data of the current transformer through a neural network model. Compared with the sensor method, it has the advantages of high accuracy, real-time and strong robustness. For example, patent CN117195100A discloses a method for predicting the error state of a current transformer, which pre-processes the historical current data of the secondary side of the current transformer, constructs characteristic parameters in the shape of an inverted pyramid, and then designs a pyramid structure prediction model. The pyramid structure prediction model is trained using the characteristic parameters in the shape of an inverted pyramid, and the trained prediction model is used to realize the error state recognition of the current transformer. However, since the measurement error of the current transformer is greatly affected by the test environment, and the existing big data method does not consider the influence of environmental factors on the measurement error of the current transformer, the accuracy of the prediction result is low. Summary of the invention

[0003] The present invention provides a current transformer measurement error prediction method and system, electronic equipment, and computer-readable storage medium, which can enable an error prediction model to learn the influence of environmental factors on current transformer errors, thereby reducing the model prediction error caused by the influence of environmental factors, improving the accuracy of current transformer measurement error results, and not only improving the optimization speed and optimization accuracy of model parameters, but also enabling the prediction model network to quickly adapt to the influence of environmental factors on current transformer measurement results, further improving the prediction accuracy.

[0004] According to one aspect of the present invention, a method for predicting a measurement error of a current transformer is provided, comprising the following contents:

[0005] Collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, a first feature vector includes current data and environmental data, and a second feature vector includes error data of the transformer;

[0006] An error prediction model of a current transformer is constructed, the first eigenvector is used as input and the second eigenvector is used as output, the error prediction model is trained using a feature data set, and a chaos-guided adaptive algorithm is used to optimize the hyperparameters of the error prediction model until the model converges or the maximum number of iterations is reached;

[0007] The current current data and environmental data are collected and input into the trained error prediction model to predict the measurement error of the current current transformer.

[0008] Furthermore, the features in the first eigenvector include current, ambient temperature, magnetic field strength and ambient humidity, and the features in the second eigenvector include ratio difference and angle difference of the mutual inductor.

[0009] Furthermore, the process of optimizing the hyperparameters of the error prediction model using the chaos-guided adaptive algorithm includes the following:

[0010] Initialize the optimal position parameters and speed parameters of the chaos-guided adaptive algorithm;

[0011] Calculate the characteristic data volatility of the first characteristic vector in the characteristic data set;

[0012] The characteristic data volatility is used as the initial value of the chaotic factor, and the optimal position parameters and speed parameters are updated adaptively;

[0013] Iterate continuously until the maximum number of iterations is reached, and output the optimal solution for the hyperparameters of the error prediction model.

[0014] Furthermore, the characteristic data volatility is calculated based on the following formula:

[0015]

[0016] Among them, δ represents the volatility of feature data, h represents the number of features in the first feature vector, Represents the average value of the i-th feature data in the feature data set, Represents the standard deviation of the i-th feature data in the feature data set.

[0017] Furthermore, the optimal position parameters and speed parameters are adaptively updated based on the following formula:

[0018]

[0019] Among them, ε represents the adaptive factor, ε max and ε min represents the maximum and minimum values ​​of the weight coefficient corresponding to the first eigenvector, k represents the number of iterations, k max represents the maximum number of iterations, ζ represents the control parameter of the chaotic system, which takes values ​​in (0, 4], and r k represents the chaos factor. When k=1, the chaos factor r1 is the characteristic data volatility of the first characteristic vector in the characteristic data set. When k>1, r k Chaotic factors mapped to environmental changes, represents the position of particle a when the number of iterations is k, represents the speed of particle a when the number of iterations is k, c1 and c2 represent the trade-off p best and g best The cognitive factor, p best and g best denote the global optimal position and the individual optimal position respectively, represents the individual optimal position of particle a when the number of iterations is k, It represents the global optimal position of particle a when the number of iterations is k.

[0020] Furthermore, the error prediction model of the current transformer includes a hidden layer, an output layer and an error function, which are respectively expressed as:

[0021]

[0022] in, represents the hidden layer, represents the output layer, Loss represents the error function, represents the weight from the i-th neuron in the (n-1)th layer of the hidden layer to the j-th neuron in the nth layer of the hidden layer, represents the bias term of the jth neuron in the nth layer of the hidden layer, f() represents the hidden layer activation function, represents the weight from the jth neuron in the nth hidden layer to the kth neuron in the output layer, represents the bias term of the kth neuron in the output layer, g() represents the activation function of the output layer, M represents the number of sample data contained in the feature data set, and They represent the true values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data, and The predicted values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data respectively.

[0023] Furthermore, the following contents are also included:

[0024] The predicted measurement error is used to correct the current value measured by the current transformer.

[0025] In addition, the present invention also provides a current transformer measurement error prediction system, comprising:

[0026] An error data acquisition module is used to collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, the first feature vector includes current data and environmental data, and the second feature vector includes error data of the transformer;

[0027] An error prediction model training module is used to construct an error prediction model of a current transformer, taking the first eigenvector as input and the second eigenvector as output, using a feature data set to train the error prediction model, and using a chaos-guided adaptive algorithm to optimize hyperparameters of the error prediction model until the model converges or reaches a maximum number of iterations;

[0028] The measurement error prediction module is used to collect current current data and environmental data, input them into the trained error prediction model, and predict the measurement error of the current current transformer.

[0029] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0030] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for predicting the measurement error of a current transformer, wherein the computer program executes the steps of the method described above when running on a computer.

[0031] The present invention has the following beneficial effects:

[0032] The current transformer measurement error prediction method of the present invention first collects the error data of the current transformer under different test conditions to construct a feature data set, then constructs the error prediction model of the current transformer, uses the feature data set to train the error prediction model, and finally, predicts the measurement error of the current current transformer based on the current current data, environmental data and the trained error prediction model. By training the error prediction model with environmental data as a feature, the error prediction model can learn the influence of environmental factors on the current transformer error, thereby reducing the model prediction error caused by the influence of environmental factors and improving the accuracy of the current transformer measurement error result. In addition, a chaos-guided adaptive algorithm is used to optimize the hyperparameters of the prediction model, which not only improves the optimization speed and optimization accuracy of the model parameters, but also enables the prediction model network to quickly adapt to the influence of environmental factors on the current transformer measurement results, further improving the prediction accuracy.

[0033] In addition, the measurement error prediction system of the current transformer of the present invention also has the above advantages.

[0034] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0036] Figure 1 It is a flow chart of a method for predicting measurement errors of a current transformer according to a preferred embodiment of the present application.

[0037] Figure 2 yes Figure 1 Schematic diagram of the sub-process of step S2 in FIG.

[0038] Figure 3 It is another flow chart of the method for predicting the measurement error of the current transformer according to the preferred embodiment of the present application.

[0039] Figure 4 It is a schematic diagram of the module structure of a measurement error prediction system for a current transformer according to another embodiment of the present application. DETAILED DESCRIPTION

[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0041] Reference Figure 1The preferred embodiment of the present application provides a method for predicting the measurement error of a current transformer, comprising the following contents:

[0042] Step S1: Collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, the first feature vector includes Including current data and environment data, the second eigenvector includes error data of the mutual inductor;

[0043] Step S2: constructing an error prediction model for the current transformer, taking the first eigenvector as input and the second eigenvector as output, using the characteristic data set to train the error prediction model, and using a chaos-guided adaptive algorithm to optimize the hyperparameters of the error prediction model until the model converges or reaches a maximum number of iterations;

[0044] Step S3: Collect current current data and environmental data, input them into the trained error prediction model, and predict the measurement error of the current current transformer.

[0045] It can be understood that the current transformer measurement error prediction method of this embodiment first collects the error data of the current transformer under different test conditions to construct a feature data set, then constructs the error prediction model of the current transformer, uses the feature data set to train the error prediction model, and finally, predicts the measurement error of the current current transformer based on the current current data, environmental data and the trained error prediction model. By training the error prediction model with environmental data as a feature, the error prediction model can learn the influence of environmental factors on the current transformer error, thereby reducing the model prediction error caused by the influence of environmental factors and improving the accuracy of the current transformer measurement error result. In addition, the chaos-guided adaptive algorithm is used to optimize the hyperparameters of the prediction model, which not only improves the optimization speed and optimization accuracy of the model parameters, but also enables the prediction model network to quickly adapt to the influence of environmental factors on the current transformer measurement results, further improving the prediction accuracy.

[0046] It can be understood that in step S1, the test conditions of the current transformer are set, the test conditions include current x1, ambient temperature x2, magnetic field strength x3, ambient humidity x4, and then the error data of the current transformer under different test conditions are collected, and the error data include the ratio difference y1 and the angle difference y2 of the transformer. Among them, the error data obtained under each test condition constitutes a sample data, that is, each sample data includes two feature vectors, the first feature vector includes current x1 and environmental parameters (i.e. ambient temperature x2, magnetic field strength x3, ambient humidity x4), and the second feature vector includes the ratio difference y1 and the angle difference y2 of the transformer, which can be expressed as: x = {x1, x2, ..., x4}

[0047] y={y1,y2}, several sample data obtained under various test conditions can constitute the feature data set z.

[0048] It can be understood that in step S2, an error prediction model of the current transformer is first constructed, wherein the error prediction model specifically includes a hidden layer, an output layer and an error function, which are respectively expressed as:

[0049]

[0050]

[0051] in, represents the hidden layer, represents the output layer, Loss represents the error function, represents the weight from the i-th neuron in the (n-1)th layer of the hidden layer to the j-th neuron in the nth layer of the hidden layer, represents the bias term of the jth neuron in the nth layer of the hidden layer, f() represents the hidden layer activation function, represents the weight from the jth neuron in the nth hidden layer to the kth neuron in the output layer, represents the bias term of the kth neuron in the output layer, g() represents the activation function of the output layer, M represents the number of sample data contained in the feature data set, and They represent the true values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data, and The predicted values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data respectively.

[0052] Then, the first eigenvector is used as input and the second eigenvector is used as output, and the error prediction model is trained using the feature data set z. In the model training process, the hyperparameters of the error prediction model are optimized using a chaos-guided adaptive algorithm until the model converges or reaches the maximum number of iterations, and the optimal hyperparameters are retained.

[0053] Among them, Figure 2 As shown, the process of optimizing the hyperparameters of the error prediction model using the chaos-guided adaptive algorithm includes the following contents:

[0054] Step S21: Initializing the optimal position parameters and speed parameters of the chaos guided adaptive algorithm;

[0055] Step S22: Calculate and obtain the characteristic data volatility of the first characteristic vector in the characteristic data set;

[0056] Step S23: taking the characteristic data volatility as the initial value of the chaos factor, and adaptively updating the optimal position parameter and speed parameter;

[0057] Step S24: Continuously iterate until the maximum number of iterations is reached, and output the optimal solution for the hyperparameters of the error prediction model.

[0058] Specifically, first set the number of particles a and the number of iterations e, where a = 1, 2, ..., d. Initialize the optimal position parameters in the chaos-guided adaptive algorithm and speed parameters n0 and h0 represent the initial values ​​of the number of hidden layers n and the number of hidden layer neurons h, respectively. and They represent the speed parameters corresponding to n0 and h0 respectively, Indicates the initial value of the position parameter, Indicates the initial value of the speed parameter.

[0059] Then, the characteristic data volatility is calculated based on the following formula:

[0060]

[0061] Among them, δ represents the volatility of feature data, h represents the number of features in the first feature vector, and x i Represents the average value of the i-th feature data in the feature data set, Represents the standard deviation of the i-th feature data in the feature data set. For example, calculate the standard deviations of current x1, ambient temperature x2, magnetic field strength x3, and ambient humidity x4 in feature data set z respectively. and the average The standard deviation With the average The average of the ratios of is obtained to obtain the characteristic data volatility of the first eigenvector in the characteristic data set z:

[0062] Next, the characteristic data volatility is used as the initial value of the chaotic factor, and the optimal position parameter and speed parameter are adaptively updated. Specifically, the optimal position parameter and speed parameter are adaptively updated based on the following formula:

[0063]

[0064] Among them, ε represents the adaptive factor, ε max and ε min represents the maximum and minimum values ​​of the weight coefficient corresponding to the first eigenvector, k represents the number of iterations, k max represents the maximum number of iterations, ζ represents the control parameter of the chaotic system, which takes values ​​in (0, 4], and r k represents the chaos factor. When k=1, the chaos factor r1 is the characteristic data volatility of the first characteristic vector in the characteristic data set. When k>1, r kChaotic factors mapped to environmental changes, represents the position of particle a when the number of iterations is k, represents the speed of particle a when the number of iterations is k, c1 and c2 represent the trade-off p best and g best The cognitive factor, p best and g best denote the global optimal position and the individual optimal position respectively, represents the individual optimal position of particle a when the number of iterations is k, It represents the global optimal position of particle a when the number of iterations is k.

[0065] Continuously perform adaptive updates to determine whether the algorithm has reached the maximum number of iterations e. If it has reached the maximum number of iterations, stop the algorithm and output the global optimal solution to obtain n. beat and h best .

[0066] Finally, the number of hidden layers and the number of hidden layer neurons of the error prediction model are set to n beat and h best ,The back propagation algorithm is adopted and the model is trained using the feature data set z to obtain the current transformer error prediction model.

[0067] It can be understood that the present invention introduces a chaotic factor r that maps the changes in the current transformer test environment factors. k , the environmental parameter data volatility δ in the feature data set z is used as the initial value of the chaotic factor, and the optimization speed and the update of the optimization position are closely linked to the change of environmental factors, so that the prediction model network can more quickly adapt to the current change law of the current transformer output and the influence of external environmental factors such as magnetic field on the current measurement of the current transformer, thereby reducing the model prediction error caused by the influence of external environmental factors, and improving the robustness of the neural network and the accuracy of the prediction results. In addition, the present invention proposes for the first time an adaptive factor ε for the current transformer error measurement, which adds a conditional perception function The derivative of the conditional perception function w(k) gradually decreases and approaches 0 as the number of iterations k increases. Therefore, in the initial stage of optimization, the adaptive factor ε generates a larger step size to speed up the optimization. This accelerates the fitting speed of the prediction model network for input (current x1, ambient temperature x2, magnetic field strength x3 and ambient humidity x4) and output (ratio difference y1 and angle difference y2); in the later stage of optimization, the derivative of the conditional perception function w(k) gradually approaches 0 as the number of iterations k increases, causing the update step size generated by the adaptive factor ε to gradually become smaller, achieving precise search in a small area, thereby greatly improving the prediction calculation efficiency and accuracy of the current transformer error.

[0068] It can be understood that in step S3, the current current data and environmental data are collected and input into the trained error prediction model to obtain the ratio error and angle error of the current transformer in the current environment.

[0069] Alternatively, if Figure 3 As shown, the current transformer measurement error prediction method also includes the following contents:

[0070] Step S4: Correcting the current value measured by the current transformer using the predicted measurement error.

[0071] Specifically, the current value measured by the current transformer is corrected based on the following formula:

[0072]

[0073] Among them, I test Indicates the current value measured by the current transformer, Indicates the current value after the ratio error correction, surface Angle The current value after the difference correction, K n Indicates the rated current ratio.

[0074] In addition, in order to verify the advantages and authenticity of the measurement error prediction method of the present invention, the inventors of the present application also conducted an experimental comparison between the scheme of the present application and the existing algorithm. Specifically, part of the data of the collected feature data set is shown in Table 1.

[0075] Table 1. Partial data of feature dataset

[0076] time Ratio difference / % Phase difference / ' Current / A Ambient temperature / ℃ Magnetic field strength / Gs Ambient humidity / %RH 9：00：00 -0.188 -15.054 10.3 18.3 0.11 65.6 9：10：00 -0.192 -16.731 11.5 19.5 0.09 65.4 9：20：00 -0.185 -16.235 10.8 20.1 0.10 64.8 9：30：00 -0.196 -15.648 9.4 20.3 0.07 64.7 9：40：00 -0.193 -15.593 11.4 21.0 0.08 62.6 9：50：00 -0.201 -15.128 10.6 21.5 0.12 62.8

[0077] The experimental results of the chaos guided adaptive algorithm of the present invention and the existing algorithm are compared as shown in Table 2.

[0078] Table 2. Example results of different algorithms

[0079]

[0080] It can be seen from Table 2 that the chaos-guided adaptive algorithm of the present invention improves the accuracy compared with the existing algorithm, shortens the calculation time and improves the calculation efficiency compared with the existing optimization algorithm.

[0081] In addition, the results of error prediction using the chaos-guided adaptive algorithm of the present invention are shown in Table III.

[0082] Table 3. Example of current transformer error prediction results

[0083]

[0084]

[0085] In addition, if Figure 4 As shown, another embodiment of the present invention further provides a measurement error prediction system for a current transformer, preferably using the measurement error prediction method as described above, the system comprising:

[0086] An error data acquisition module is used to collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, the first feature vector includes current data and environmental data, and the second feature vector includes error data of the transformer;

[0087] An error prediction model training module is used to construct an error prediction model of a current transformer, taking the first eigenvector as input and the second eigenvector as output, using a feature data set to train the error prediction model, and using a chaos-guided adaptive algorithm to optimize hyperparameters of the error prediction model until the model converges or reaches a maximum number of iterations;

[0088] The measurement error prediction module is used to collect current current data and environmental data, input them into the trained error prediction model, and predict the measurement error of the current current transformer.

[0089] It can be understood that the current transformer measurement error prediction system of this embodiment first collects the error data of the current transformer under different test conditions to construct a feature data set, then constructs the error prediction model of the current transformer, uses the feature data set to train the error prediction model, and finally, predicts the measurement error of the current current transformer based on the current current data, environmental data and the trained error prediction model. By training the error prediction model with environmental data as a feature, the error prediction model can learn the influence of environmental factors on the current transformer error, thereby reducing the model prediction error caused by the influence of environmental factors and improving the accuracy of the current transformer measurement error result. In addition, the chaos-guided adaptive algorithm is used to optimize the hyperparameters of the prediction model, which not only improves the optimization speed and optimization accuracy of the model parameters, but also enables the prediction model network to quickly adapt to the influence of environmental factors on the current transformer measurement results, further improving the prediction accuracy.

[0090] In addition, the measurement error prediction system further comprises:

[0091] The measurement error correction module is used to correct the current value measured by the current transformer using the predicted measurement error.

[0092] In addition, another embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0093] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for predicting the measurement error of a current transformer, wherein the computer program executes the steps of the method described above when running on a computer.

[0094] The general form of computer readable storage media includes: floppy disk, flexible disk, hard disk, magnetic tape, any other magnetic medium, CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash erasable programmable read-only memory (FLASH-EPROM), any other memory chip or cartridge, or any other medium that can be read by a computer. The instructions can further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium that can be used to store, encode or carry instructions for execution by a machine, and includes digital or analog communication signals or other intangible media that facilitate the communication of the above instructions. Transmission media include coaxial cables, copper wires and optical fibers, which include the wires of a bus used to transmit a computer data signal.

[0095] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0096] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0099] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting measurement error of a current transformer, characterized in that: Includes the following: Collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, a first feature vector includes current data and environmental data, and a second feature vector includes error data of the transformer; An error prediction model of a current transformer is constructed, the first eigenvector is used as input and the second eigenvector is used as output, the error prediction model is trained using a feature data set, and a chaos-guided adaptive algorithm is used to optimize the hyperparameters of the error prediction model until the model converges or the maximum number of iterations is reached; The current current data and environmental data are collected and input into the trained error prediction model to predict the measurement error of the current current transformer.

2. The method for predicting the measurement error of a current transformer according to claim 1, characterized in that: The features in the first eigenvector include current, ambient temperature, magnetic field strength and ambient humidity, and the features in the second eigenvector include ratio difference and angle difference of the mutual inductor.

3. The method for predicting the measurement error of a current transformer according to claim 1, characterized in that: The process of optimizing the hyperparameters of the error prediction model using the chaos-guided adaptive algorithm includes the following contents: Initialize the optimal position parameters and speed parameters of the chaos-guided adaptive algorithm; Calculate the characteristic data volatility of the first characteristic vector in the characteristic data set; The characteristic data volatility is used as the initial value of the chaotic factor, and the optimal position parameters and speed parameters are updated adaptively; Iterate continuously until the maximum number of iterations is reached, and output the optimal solution for the hyperparameters of the error prediction model.

4. The method for predicting the measurement error of a current transformer according to claim 3, characterized in that: The characteristic data volatility is calculated based on the following formula: Among them, δ represents the volatility of feature data, h represents the number of features in the first feature vector, Represents the average value of the i-th feature data in the feature data set, Represents the standard deviation of the i-th feature data in the feature data set.

5. The method for predicting the measurement error of a current transformer according to claim 3, characterized in that: Adaptively update the optimal position and speed parameters based on the following formula: Among them, ε represents the adaptive factor, ε max and ε min represents the maximum and minimum values ​​of the weight coefficient corresponding to the first eigenvector, k represents the number of iterations, k max represents the maximum number of iterations, ζ represents the control parameter of the chaotic system, which takes values ​​in (0, 4], and r k represents the chaos factor. When k=1, the chaos factor r1 is the characteristic data volatility of the first characteristic vector in the characteristic data set. When k>1, r k Chaotic factors mapped to environmental changes, represents the position of particle a when the number of iterations is k, represents the speed of particle a when the number of iterations is k, c1 and c2 represent the trade-off p best and g best The cognitive factor, p best and g best denote the global optimal position and the individual optimal position respectively, represents the individual optimal position of particle a when the number of iterations is k, It represents the global optimal position of particle a when the number of iterations is k.

6. The method for predicting the measurement error of a current transformer according to claim 3, characterized in that: The error prediction model of the current transformer includes a hidden layer, an output layer, and an error function, which are expressed as: in, represents the hidden layer, represents the output layer, Loss represents the error function, represents the weight from the i-th neuron in the (n-1)th layer of the hidden layer to the j-th neuron in the nth layer of the hidden layer, represents the bias term of the jth neuron in the nth layer of the hidden layer, f() represents the hidden layer activation function, represents the weight from the jth neuron in the nth hidden layer to the kth neuron in the output layer, represents the bias term of the kth neuron in the output layer, g() represents the activation function of the output layer, M represents the number of sample data contained in the feature data set, and They represent the true values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data, and The predicted values ​​of the ratio difference and angle difference of the mutual inductor in the mth sample data respectively.

7. The method for predicting the measurement error of a current transformer according to claim 1, characterized in that: Also included are the following: The predicted measurement error is used to correct the current value measured by the current transformer.

8. A current transformer measurement error prediction system, characterized in that: include: An error data acquisition module is used to collect error data of the current transformer under different test conditions to construct a feature data set, wherein each sample data in the feature data set includes two feature vectors, the first feature vector includes current data and environmental data, and the second feature vector includes error data of the transformer; An error prediction model training module is used to construct an error prediction model of a current transformer, taking the first eigenvector as input and the second eigenvector as output, using a feature data set to train the error prediction model, and using a chaos-guided adaptive algorithm to optimize hyperparameters of the error prediction model until the model converges or reaches a maximum number of iterations; The measurement error prediction module is used to collect current current data and environmental data, input them into the trained error prediction model, and predict the measurement error of the current current transformer.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for predicting a measurement error of a current transformer, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images