Battery Simulator Circuit Health Assessment Method Based on Star Operation - A3C

Abstract

The present invention discloses a method for evaluating the health of a battery simulator circuit based on star operation - A3C. This method normalizes high-frequency current signals, uses the periodic sparse method for signal reconstruction to obtain periodic sparse data; converts it into an evaluation feature map by using the recurrence plot method; introduces an improved star operation convolutional network, combines the multi-threaded Actor-Critic cooperation mechanism of the A3C algorithm, the Actor network adaptively adjusts the feature fusion weights, the Critic network evaluates the feature fusion value, and dynamically captures high-order feature correlations; uses a support vector machine optimized by the energy valley algorithm to realize the evaluation of the health state of the battery simulator circuit. Through the feature collaborative optimization and parallel computing acceleration of star operation - A3C, the present invention realizes the evaluation of the health state of the battery simulator circuit, and has significant technical advantages and application values.

Description

Technical Field

[0001] The present invention belongs to the field of reinforcement learning fault diagnosis, and specifically relates to a battery simulator circuit health assessment method based on Star Computing (A3C). Background Art

[0002] As the global energy mix shifts toward renewable energy, energy storage batteries are becoming increasingly important in areas such as electric vehicles, smart grids, photovoltaics, and wind energy storage systems. As a core energy storage unit, battery performance directly impacts system efficiency and safety. However, real-world battery testing presents challenges such as high cost, uncontrollable aging, and significant environmental risks. Battery simulators enable safe and efficient circuit health assessments. Battery simulators are an indispensable tool in battery research and development. The health of a battery simulator's circuits is affected by a combination of factors. Electrically, voltage stress can cause insulation damage; environmentally, high temperature, high humidity, dust, and contaminant accumulation can lead to circuit damage. In manufacturing processes, improper design can lead to stress concentration, heat dissipation, and electromagnetic compatibility issues, all of which can affect the health of the battery simulator's circuits. By assessing the circuit health of a battery simulator, we can develop appropriate maintenance strategies, improve the long-term reliability of batteries, and reduce losses caused by equipment failures. Summary of the Invention

[0003] The purpose of the present invention is to provide a battery simulator circuit health assessment method based on Star Operation-A3C. By evaluating the health status of the battery simulator circuit, timely preventive measures can be taken to increase the battery life and reduce the occurrence of safety accidents.

[0004] To solve the above technical problems, the present invention provides a battery simulator circuit health assessment method based on Star Computing-A3C, comprising:

[0005] The high-frequency current signal is normalized and the signal is reconstructed using the periodic sparse method to obtain periodic sparse data.

[0006] A recursive graph method is used to convert periodic sparse data into evaluation feature graphs;

[0007] Dynamically capture high-order feature associations through the Star Operation-A3C module;

[0008] The support vector machine optimized based on the energy valley algorithm is used to realize the circuit health status assessment of the battery simulator.

[0009] Optionally, normalizing the high-frequency current signal and reconstructing the signal using a periodic sparse method to obtain periodic sparse data includes:

[0010] Normalize the high-frequency current signal, and use the periodic sparse method to process the inherent periodicity contained in the current data, reducing the information loss in the signal reconstruction process. First, downsample the normalized high-frequency current signal and segment it into cross-period subsequences, then predict the trend of the cross-period subsequences through sparse sliding prediction to obtain a prediction matrix, and finally transpose and reconstruct the prediction matrix through upsampling to obtain periodic sparse data.

[0011] Optionally, converting the periodic sparse data into an evaluation feature map by using the recurrence plot method includes:

[0012] Use the recurrence plot method to capture the non-linear relationships in the data for non-stationary sparse data that is difficult to handle by traditional methods. First, perform phase space reconstruction on the periodic sparse data, calculate the distances between each pair of points in the reconstructed phase space using the Euclidean distance to form a distance matrix, then binarize the distance matrix by setting a threshold to generate a recurrence plot matrix, and finally map the recurrence plot matrix to a grayscale image to obtain the evaluation feature map.

[0013] Optionally, dynamically capturing high-order feature associations through the star operation - A3C module includes:

[0014] Input the evaluation feature map into the star operation - A3C module, mine the high-order relationships between features through the star operation and map the input to a high-dimensional non-linear feature space, perform two linear transformations on the evaluation feature map to obtain two feature vectors, multiply the two transformed feature vectors element-wise to obtain the star operation interaction feature, use multi-thread parallel processing, each thread independently calculates the gradient, asynchronously updates the Actor-Critic network parameters, the Actor network adaptively adjusts the feature fusion weights in different subspaces, the Critic network evaluates the feature fusion value in real time through the advantage function, and aggregates the feature interaction results of all threads to obtain the global evaluation feature.

[0015] Optionally, implementing the health state assessment of the battery simulator circuit by using the support vector machine optimized by the energy valley algorithm includes:

[0016] Optimize the parameters of the support vector machine by using the energy valley algorithm. The energy valley algorithm simulates the formation mechanism of energy valleys in physical systems, efficiently searches for the global optimal solution in the complex high-dimensional parameter space, adaptively adjusts the search step size and direction, and reaches stable convergence within fewer iterations. Input the global evaluation feature into the optimized support vector machine to implement the health state assessment of the battery simulator circuit.

[0017] The present invention provides a method for evaluating the health of a battery simulator circuit based on the star operation - A3C. The method includes normalizing high - frequency current signals, reconstructing signals using the periodic sparse method to obtain periodic sparse data; converting it into an evaluation feature map using the recurrence plot method; introducing an improved star operation convolutional network, combining the multi - thread Actor - Critic collaborative mechanism of the A3C algorithm, where the Actor network adaptively adjusts the feature fusion weights, and the Critic network evaluates the value of feature fusion to dynamically capture high - order feature associations; and using a support vector machine optimized by the energy valley algorithm to achieve the evaluation of the health state of the battery simulator circuit. Through the feature collaborative optimization and parallel computing acceleration of the star operation - A3C, the present invention realizes the evaluation of the health state of the battery simulator circuit, having significant technical advantages and application values. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a schematic flow chart of the method for evaluating the health of a battery simulator circuit provided by an embodiment of the present invention;

[0020] Figure 2 It is a schematic framework diagram of the star operation - A3C module of the method for evaluating the health of a battery simulator circuit provided by an embodiment of the present invention;

[0021] Figure 3 It is a schematic flow chart of the energy valley algorithm of the method for evaluating the health of a battery simulator circuit provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to enable those skilled in the art to better understand the solution of the present invention, the following will further elaborate on the present invention in conjunction with the drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0023] As Figure 1 shown, Figure 1 It is a schematic flow chart of the method for evaluating the health of a battery simulator circuit provided by an embodiment of the present invention, and the method specifically includes 4 contents.

[0024] S1: The high-frequency current signal is normalized and the signal is reconstructed using the periodic sparse method to obtain periodic sparse data.

[0025] It should be noted that the present invention uses a periodic sparse method to process the inherent periodicity contained in the current data, thereby reducing information loss in the signal reconstruction process. This method can provide more accurate and reliable periodic sparse data when processing complex and non-stationary current signals, laying a good foundation for subsequent feature extraction. The method specifically includes three steps.

[0026] S11: Calculate the mean of the collected high-frequency current signal, subtract the mean from each data point, and obtain a normalized high-frequency current signal.

[0027] S12: The normalized high-frequency current signal is divided into a subsequences according to period a. Each subsequence is subjected to sliding aggregation, and the information of adjacent time points is integrated. The outliers are smoothed by weighted averaging to enhance the continuity of local trends within the subsequence.

[0028] S13: Sparsely predict the trend of each subsequence through a linear layer with shared parameters to obtain a prediction matrix, which is then transposed and reconstructed into a complete time series signal through upsampling. Finally, periodic sparse data is obtained through processing.

[0029] Based on the above discussion, in an optional embodiment of the present invention, the signal reconstruction using the periodic sparse method to obtain periodic sparse data specifically includes:

[0030] In an optional embodiment of S1, the periodic sparse method consists of two main parts: a one-dimensional convolutional layer for sliding aggregation and a shared parameter linear layer for sparse sliding prediction. The number of parameters in the one-dimensional convolutional layer is determined by the kernel size, and it is trained for 35 cycles using the Adam optimizer.

[0031] S2: A recursive graph method is used to convert periodic sparse data into evaluation feature graphs.

[0032] It should be noted that the present invention adopts the recursive graph method to convert into an evaluation feature graph because this method can intuitively display the dynamic characteristics and complex relationships of the signal, can better understand the working status and health trends of the battery simulator circuit, and also provides a rich data basis for subsequent feature extraction.

[0033] The recursive graph method is used to convert periodic sparse data into evaluation feature graphs, which includes three steps.

[0034] S21: Reconstruct the phase space of periodic sparse data, select the embedding dimension and time delay, construct a series of reconstruction vectors, and calculate each pair of points Y in the reconstructed phase spacei and Y j The Euclidean distance between .

[0035] S22: Based on the distance between points, threshold binarization is performed to generate a recursive graph matrix. The formula is as follows:

[0036]

[0037] Where θ is the Heaviside step function, if Y i and Y j The distance is less than the set threshold ε, and the corresponding position R on the recursive graph ij is marked as 1; otherwise it is marked as 0.

[0038] S23: Map the recursive graph matrix to a grayscale image to obtain the evaluation feature map.

[0039] Based on the above discussion, in an optional embodiment of the present invention, the above-mentioned recursive graph method is used to convert the periodic sparse data into an evaluation feature graph, specifically including:

[0040] In an optional embodiment of S2, the phase space reconstruction using the recursive graph method needs to satisfy the condition m≥2D+1, where m is the embedding dimension and D is the real dimension of the data; the autocorrelation function of the periodic sparse data is calculated, and the first zero crossing point is selected as the time delay.

[0041] S3: Dynamically capture high-order feature associations through the Star Operation-A3C module.

[0042] It should be noted that the present invention introduces the Star Operation-A3C module because the high-dimensional mapping and feature interaction of Star Operation can explore the complex relationships between features; through the multi-threaded parallel processing of the A3C algorithm, each thread independently calculates the gradient and asynchronously updates the network parameters, avoiding the waiting time in traditional synchronous training and greatly improving the convergence speed. The Star Operation-A3C module combines the feature interaction capabilities of Star Operation and the reinforcement learning optimization capabilities of the A3C algorithm to process complex feature relationships and optimize feature fusion strategies.

[0043] like Figure 2 As shown in the figure, high-order feature associations are dynamically captured through the Star Operation-A3C module, which specifically includes four steps.

[0044] S31: Input the evaluation feature map into the Star Operation-A3C module, and perform two linear transformations on the evaluation feature map to obtain two feature vectors. The formulas are as follows:

[0045]

[0046]

[0047] Among them, G is the evaluation feature map, W e and W f is a learnable weight parameter, b e and b f is the bias term, E and F are the transformed eigenvectors.

[0048] S32: Multiply the two transformed feature vectors element-by-element to capture the nonlinear correlation between channels and output the star operation interaction feature. The formula is as follows:

[0049]

[0050] Among them, Z is the star operation interaction feature, and ⊙ represents element-by-element multiplication.

[0051] S33: Using multi-threaded parallel processing, the Actor network adaptively adjusts the feature fusion weights of different subspaces. The formula is as follows:

[0052]

[0053] Among them, α k is the feature fusion weight coefficient of the Actor network, exp is the exponential function, T is the function parameter of the Actor network, and k is the number of subspaces.

[0054] The Critic network evaluates the feature fusion value in real time through the advantage function. The objective function of the advantage function A(G,α) is as follows:

[0055]

[0056]

[0057] Among them, Q(G,α) represents the cumulative reward for completing actions in the Critic network, V critic (Z) is the expected value reward of the Critic network, γ is the discount factor, r t is the immediate reward obtained at time t, Z t+n is the star operation interaction feature at time t+n, which serves as the input of the Critic network; n is the number of time steps, which is used for temporal difference estimation.

[0058] S34: The agents of each thread independently calculate gradients, asynchronously update the Actor-Critic network parameters, and aggregate the feature interaction results of all threads to obtain the global evaluation features.

[0059] Based on the above discussion, in an optional embodiment of the present invention, the above-mentioned dynamic capture of high-order feature associations by the star operation-A3C module specifically includes:

[0060] In an alternative embodiment of S3, the star operation - A3C module experiment is set with a training step of 512, an iteration number of 500, and a learning rate of 0.005.

[0061] S4: Use the support vector machine optimized by the energy valley algorithm to implement the health state assessment of the battery simulator circuit.

[0062] It should be noted that the health state of the battery simulator circuit is divided into four levels. Healthy means the device has no faults and the parameters are normal; sub - healthy means the device parameters deviate and the device needs to be monitored in real - time; minor fault means the device function degrades and immediate maintenance is required; severe fault means the device function fails and shutdown for maintenance and emergency treatment are needed. The energy valley algorithm adopted in the present invention can efficiently search for the global optimal solution in a complex high - dimensional parameter space, avoiding the problem that traditional optimization algorithms are prone to falling into local optima. Compared with traditional optimization algorithms, the energy valley algorithm has a faster convergence speed when optimizing the support vector machine, which is crucial for the health state assessment of the battery simulator circuit with high real - time requirements.

[0063] Use the support vector machine optimized by the energy valley algorithm to implement the health state assessment of the battery simulator circuit, which specifically includes 3 steps.

[0064] S41: Use the energy valley algorithm to optimize the hyperparameters of the support vector machine, including the regularization parameter c and the kernel function parameter d.

[0065] S42: As Figure 3 shown, in the implementation process of the energy valley algorithm, first, initialization is performed. A group of candidate solutions is randomly generated in the parameter space, and each solution represents a set of hyperparameter combinations (c, d); an energy field in the physical system is constructed, and the "energy value" of each solution is evaluated through a fitness function. The low - energy region corresponds to a better solution; a global search strategy is carried out to simulate the movement of particles in the energy field, and the candidate solutions are updated by combining gradient descent and random perturbation to avoid falling into local optima; when the maximum number of iterations is reached, it stops, and the optimal hyperparameter combination of the support vector machine is output.

[0066] S43: Input the global evaluation features into the optimized support vector machine to implement the health state assessment of the battery simulator circuit.

[0067] Based on the above discussion, in an alternative embodiment of the present invention, for the above - mentioned support vector machine optimized by the energy valley algorithm to implement the health state assessment of the battery simulator circuit, it specifically includes:

[0068] In an alternative embodiment of S4, a support vector machine is used to implement the assessment of the health state of the battery simulator circuit, and four-level dynamic health state thresholds are set. Healthy means that all parameters are within the normal range, and the probability range is greater than 95%; sub-healthy means that 1-2 parameters deviate slightly, and the probability range is 85%-95%; minor fault means that key parameters are continuously abnormal, and the probability range is 60%-85%; severe fault means that multiple parameters exceed the limit severely, and the probability range is less than 60%.

[0069] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0070] Specific examples are used in this article to elaborate on the principles and implementation methods of the present invention. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A battery simulator circuit health assessment method based on star operation - A3C, characterized in that include: The high-frequency current signal is normalized and the signal is reconstructed using the periodic sparse method to obtain periodic sparse data. A recursive graph method is used to convert periodic sparse data into evaluation feature graphs; The evaluation feature map is input into the Star Operation-A3C module. Star Operations are used to mine high-order relationships between features and map the input into a high-dimensional nonlinear feature space. Two linear transformations are performed on the evaluation feature map to obtain two feature vectors. The two transformed feature vectors are multiplied element-by-element to obtain the Star Operation interaction feature. Multi-threaded parallel processing is used, with each thread independently calculating the gradient and asynchronously updating the Actor-Critic network parameters. The Actor network adaptively adjusts the feature fusion weights of different subspaces. The Critic network uses the advantage function to evaluate the feature fusion value in real time. The feature interaction results of all threads are aggregated to obtain the global evaluation feature. The global evaluation features are input into the support vector machine optimized based on the energy valley algorithm to realize the circuit health status evaluation of the battery simulator.

2. The battery simulator circuit health assessment method as claimed in claim 1, wherein: The high-frequency current signal is normalized and the periodic sparse method is used to reconstruct the signal to obtain periodic sparse data, including: The high-frequency current signal is normalized. First, the normalized high-frequency current signal is downsampled and divided into cross-period subsequences. Then, the trend of the cross-period subsequences is predicted through sparse sliding prediction to obtain a prediction matrix. Finally, the prediction matrix is transposed and reconstructed through upsampling to obtain periodic sparse data.

3. The battery simulator circuit health assessment method according to claim 1, wherein A recursive graph method is used to convert periodic sparse data into evaluation feature graphs, including: Firstly, the periodic sparse data is reconstructed in phase space. The distance between each pair of points in the reconstructed phase space is calculated using Euclidean distance to form a distance matrix. Then, the distance matrix is binarized by setting a threshold to generate a recursive graph matrix. Finally, the recursive graph matrix is mapped to a grayscale image to obtain an evaluation feature map.

4. The battery simulator circuit health assessment method according to claim 1, characterized in that The global evaluation features are input into the support vector machine optimized based on the energy valley algorithm to realize the circuit health status assessment of the battery simulator, including: The parameters of the support vector machine are optimized based on the energy valley algorithm. The global optimal solution is efficiently searched in the complex high-dimensional parameter space. The search step size and direction are adaptively adjusted. The global evaluation features are input into the optimized support vector machine to realize the circuit health status assessment of the battery simulator.

Images