Lithium battery degradation data intelligent generation method based on improved generative adversarial network

Abstract

The invention relates to a lithium battery degradation data intelligent generation method based on an improved generative adversarial network, and belongs to the field of battery degradation data generation and equipment health protection, and the method comprises the following steps: obtaining lithium battery historical degradation data, including discharge voltage and capacity data; carrying out normalization on the capacity data, carrying out one-to-one matching correspondence on the discharge voltage and the capacity data, constructing a feature-label pair, and forming a training data set; respectively training the LSTM and the improved GAN network by using the training data set to obtain a battery capacity prediction model and a degradation data generation model; collecting a current cyclic discharge voltage curve of the lithium battery, and predicting the current capacity of the battery by using the battery capacity prediction model; according to the physical characteristics of the target battery, designing and executing a battery capacity generation mechanism program, and setting an expected personalized battery capacity; and inputting a preset target capacity into the degradation data generation model to generate synthesized degradation data.

Description

Technical Field

[0001] The present invention belongs to the field of battery degradation data generation and equipment health protection, and relates to a method for intelligently generating lithium battery degradation data based on an improved generative adversarial network. Background Art

[0002] Due to its advantages such as high energy density, decreasing cost, low self-discharge rate and long life, lithium batteries are widely used in many important fields such as consumer electronics, electric vehicles, drones, smart power station energy storage systems, and have become the energy center of many Internet of Things devices. However, with the increase of charge and discharge cycles, the performance of lithium batteries will gradually degrade, resulting in capacity decay and power drop, which not only affects the reliability of the equipment but also may increase the risk of thermal runaway. Therefore, it is necessary to always understand the health status of lithium batteries to ensure stable operation of the system. Currently, the mainstream lithium battery health status estimation method is based on data-driven prediction method. However, this method has high requirements on the quantity and quality of data. However, obtaining complete and representative battery degradation data faces the problems of long testing time, high cost of obtaining degradation data, and safety risks in the acquisition process. The intelligent generation method of lithium battery degradation data based on the improved generative adversarial network proposed in the present invention generates lithium battery degradation data with specified capacity results according to user needs, providing data support for the lithium battery health status estimation model and lithium battery safety monitoring system. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method for intelligently generating lithium battery degradation data based on an improved generative adversarial network to address the problem of generating lithium battery health status estimation data.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for intelligently generating lithium battery degradation data based on an improved generative adversarial network comprises the following steps:

[0006] Use the battery management system to obtain historical degradation data of lithium batteries, including lithium battery discharge voltage data and lithium battery capacity degradation data;

[0007] Normalize the capacity degradation data, match the discharge voltage with the capacity degradation data one by one, construct feature-label pairs, and form a training dataset.

[0008] The long short-term memory (LSTM) network and the improved generative adversarial network (GAN) were trained using the training dataset to obtain a battery capacity prediction model based on the LSTM network and a degradation data intelligent generation model based on the improved GAN network.

[0009] Collecting the current cycle discharge voltage curve of the lithium battery, and using the battery capacity prediction model to predict the current capacity of the battery;

[0010] Design and execute a personalized battery capacity generation program based on the physical characteristics of the target battery, tamper with the predicted current battery capacity, and generate the desired target capacity;

[0011] The preset target capacity is input into the degradation data intelligent generation model to generate synthetic degradation data.

[0012] Furthermore, the capacity degradation data is normalized, and the formula is as follows:

[0013]

[0014] Among them, c t represents the actual capacity measured during t cycle, c max Indicates the rated maximum capacity of the lithium battery in its initial state. Represents the normalized capacity degradation data.

[0015] Furthermore, the steps for forming the training data set are as follows:

[0016] Collect lithium battery discharge voltage data V={v1,v2,v3,…,v n}, and simultaneously record the normalized capacity degradation data within the corresponding cycle period Form a training dataset for health state estimation.

[0017] Furthermore, the training process of the battery capacity prediction model based on the LSTM network is as follows:

[0018] Take the discharge voltage data V as input, the corresponding capacity As labels for supervised training; the input data is organized into three-dimensional tensors (batch_size, sequence_length, input_size) in batches and fed into the LSTM network. The output of the LSTM layer is linearly transformed through a fully connected layer, and the final output is the capacity estimate. By minimizing the error between the estimate and the label, a regression relationship for the degradation behavior is established. The loss function is:

[0019]

[0020] Among them E ω Represents the loss function value, Y t Indicates the label value of the t-th capacity, Ψ LSTM represents the LSTM model, x t represents the discharge voltage data of the tth cycle, Θw Represents the parameter set of the LSTM model; Θ is updated by the backpropagation algorithm w , when E ω The training is stopped when it is less than the threshold ε.

[0021] Furthermore, the intelligent generation model of degraded data based on the improved GAN network includes:

[0022] The generator is composed of a multi-layer perceptron, which receives the random noise vector z and the conditional information y to generate the synthetic discharge voltage data. Among them, θ g Represents the trainable parameters of the generator;

[0023] The discriminator is composed of an LSTM network and is used to receive input data x. Its formula is:

[0024] D(x;θ d )→[0,1]

[0025] Among them, θ d Represents the trainable parameters of the discriminator. The closer the output value is to 1, the higher the probability that the input data is the real discharge voltage data.

[0026] The value function is:

[0027]

[0028] Where x represents the real discharge voltage data sample, which obeys the real sample distribution P r , z represents random noise, which obeys the uniform distribution z~U(-0.01,0.01), y represents the capacity degradation information as the conditional information of the GAN network, D(·) is the discriminator, D(x|y) represents the score of the discriminator on the input x as real data, G(·) is the generator, G(z|y) is the sample generated by the generator based on random noise z and condition y, λ is the weight parameter, which controls the weight of the gradient penalty term, It is a sample obtained by linear interpolation of the real data x and the generated data G(z), that is, where ∈~U(0,1), Represents the gradient modulus of the discriminator to the input sample;

[0029] The generator loss function is:

[0030]

[0031] in, Represents the synthetic discharge voltage data generated by the generator, satisfying z represents a random noise vector that obeys uniform distribution, y is the capacity degradation condition information, P g To generate data distribution;

[0032] The discriminator loss function is:

[0033]

[0034] Among them, x is the actual discharge voltage data, which obeys the distribution P r , is the interpolation sample, and λ is the gradient penalty coefficient.

[0035] Furthermore, the training process of the degraded data intelligent generation model based on the improved GAN network is as follows:

[0036] Step (1): Randomly initialize the weights of the generator G and the discriminator D;

[0037] Step (2): Fix the generator G and optimize it with the Adam optimizer at a learning rate η D Update the discriminator parameters θ D , minimize the discriminator loss

[0038] Step (3): Fix the discriminator D and use the Adam optimizer with a learning rate η g Update the generator parameters θ G , minimize the generator loss

[0039] Step (4): Repeat steps (2)-(3) until the maximum number of iterations T is reached max ;

[0040] The improved GAN network minimizes the discriminator loss Measure the generated data distribution P g With the real data distribution P r The Wasserstein distance is , where the ratio of the number of discriminator training rounds k to the number of generator training rounds is 2:1, and adversarial training is used to make the data distribution of the generator output close to the real data distribution.

[0041] Furthermore, the collecting of the current cycle discharge voltage curve of the lithium battery and the use of the battery capacity prediction model to predict the current capacity of the battery specifically include:

[0042] The current discharge voltage curve x t Send it to the battery capacity prediction model based on LSTM network to get the current predicted capacity The formula is:

[0043]

[0044] Among them, Θ w Solidify the parameter set for the LSTM model, ΨLSTM Represents the LSTM network mapping relationship including the fully connected layer.

[0045] Furthermore, the design of the battery capacity personalized generation program satisfies the following physical constraints:

[0046] (a) Generated target capacity value C adv satisfy:

[0047] 0 <C adv ≤C rated

[0048] Among them C rated is the rated capacity of the battery;

[0049] (b) The capacity change ΔC between adjacent cycles satisfies:

[0050] |△C|<δ max

[0051] where δ max The maximum allowable capacity fluctuation threshold is preset based on the battery chemical characteristics.

[0052] Furthermore, the execution process of the battery capacity personalized generation program includes:

[0053] According to the predicted capacity Generate expected target capacity The formula is:

[0054]

[0055] Where GEN(·) is the expected capacity generation function set by the user;

[0056] The expected capacity generation function GEN(·) is divided into three categories according to different modification forms: incremental modification, decremental modification, or peak modification. The formulas are:

[0057] Peak modification:

[0058] Incremental modification:

[0059] Reduction modification:

[0060] in, Indicates the preset target capacity. represents the capacity acquired by the LSTM network, ε represents the set modification amplitude, and k(t) represents the continuous modification strategy.

[0061] Furthermore, the generating of synthetic degradation data specifically includes:

[0062] Set the preset target capacity It is fed together with random noise z into the degradation data intelligent generation model based on the improved GAN network to obtain the synthesized discharge voltage data, the formula of which is:

[0063]

[0064] where Θ G Represents the parameters of the improved GAN network.

[0065] The present invention provides a novel method for intelligently generating lithium battery degradation data based on an improved generative adversarial network (GAN), tailored to the specific characteristics of lithium battery degradation data. By improving the GAN network, it accurately generates degradation data corresponding to the capacity, guiding the lithium battery's state of health estimation result to the desired outcome. This method does not require a physical and chemical model of the lithium battery's internal structure, requiring only partial historical data. Compared to existing methods, the present invention generates samples with greater flexibility and generalization capabilities.

[0066] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0068] Figure 1 This is a flow chart of the method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to the present invention;

[0069] Figure 2 Schematic diagram of the structure of the method for intelligently generating lithium battery degradation data based on the improved generative adversarial network according to the present invention;

[0070] Figure 3 Schematic diagram of the improved generative adversarial network structure of the present invention;

[0071] Figure 4 These are the modification results of the cases described in the present invention, where (a) is a peak modification, (b) is an incremental modification, and (c) is a decremental modification. DETAILED DESCRIPTION

[0072] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0073] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0074] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0075] Example 1:

[0076] like Figure 1 As shown, the present invention provides a method for intelligently generating lithium battery degradation data based on an improved generative adversarial network. The improved generative adversarial network used in the present invention is as follows: Figure 2 As shown in the figure, the generator is a multi-layer perceptron, including an input layer, a hidden layer, and an output layer. The input layer dimension is 85, the hidden layer is two layers, each with a dimension of 128, and the output layer dimension is 85. The discriminator is an LSTM network with a time step of sequence length = 85 and the number of features at each time step is featuredim = 2. This method includes the following steps:

[0077] Step 1: Collect historical degradation data of lithium batteries. Collect 2547 discharge voltage data and capacity data. Each discharge voltage data contains 85 sampling points. Normalize the capacity data of each battery. The mathematical expression is:

[0078]

[0079] where c t represents the capacity of the t-th cycle, c max Indicates the maximum capacity of the battery. represents the capacity of the t-th cycle after normalization.

[0080] Step 2: Match the collected discharge voltage data and capacity data one by one, and divide them into training data set and test data set. The training data set contains 2267 discharge voltage data and capacity data, and the test data set contains 280 discharge voltage data and capacity data.

[0081] Step 3: Use the training dataset to train the LSTM network to obtain a model that can accurately predict capacity data based on discharge voltage data.

[0082] Step 4: Use the training dataset to train the improved GAN network to obtain a model that can generate discharge voltage data that matches the capacity condition from random noise and capacity condition.

[0083] Step 5: Based on the target battery physical information, the capacity personalized generation program designs three modification modes: peak modification, incremental modification, and decrement modification. The peak modification formula is: in represents the expected target capacity, c represents the capacity captured by the LSTM network, and ε represents the set modification amplitude; the incremental modification formula is: Where k(t) represents the strategy of continuous modification; the decremental modification formula is:

[0084] Step 6: Collect the current discharge voltage data v from the lithium battery t , sent to the LSTM network to obtain the current capacity c t , the mathematical expression is c t =Ψ LSTM (v t ;Θ w ).

[0085] Step 7: Based on the current capacity, use different generation modes of the capacity personalization generation program: in peak modification, when t = 100-102, ε = 0.02, when t = 200-202, ε = 0.04; in incremental modification, ε = 0.0003, k(t) = 0.25t; in decremental modification, ε = 0.0007, k(t) = 0.25t, to obtain the expected target capacity.

[0086] Step 8: Set target capacity and random noise z are fed into the improved GAN network to obtain the synthesized discharge voltage data in z~μ(-0.01,0.01); Use the generated discharge voltage data to guide the test results. Figure 4As shown, (a) is peak modification, (b) is incremental modification, and (c) is decremental modification.

[0087] Example 2:

[0088] In this embodiment, a training process of a battery capacity prediction model based on an LSTM network is provided, as follows:

[0089] The discharge voltage data V is used as input, and the corresponding capacity C is used as the label for supervised training. The input data is organized into three-dimensional tensors (batch_size, sequence_length, input_size) in batches and fed into the LSTM network. The output of the LSTM layer is linearly transformed through a fully connected layer, and the final output is the capacity estimate. By minimizing the error between the estimate and the label, a regression relationship between the degradation behavior is established. The loss function is:

[0090]

[0091] Among them E ω Represents the loss function value, Y t Indicates the label value of the t-th capacity, Ψ LSTM represents the LSTM model, x t represents the discharge voltage data of the tth cycle, Θ w Represents the parameter set of the LSTM model; Θ is updated by the backpropagation algorithm w , when E ω The training is stopped when it is less than the threshold ε.

[0092] Example 3:

[0093] In this embodiment, a specific structure of an improved GAN network is provided, such as Figure 3 Shown, including:

[0094] The generator is composed of a multi-layer perceptron, which receives the random noise vector z and the conditional information y to generate the synthetic discharge voltage data. Among them, θ g Represents the trainable parameters of the generator;

[0095] The discriminator is composed of an LSTM network and is used to receive input data x. Its formula is:

[0096] D(x;θ d )→[0,1]

[0097] Among them, θ d Represents the trainable parameters of the discriminator. The closer the output value is to 1, the higher the probability that the input data is the real discharge voltage data.

[0098] The value function is:

[0099]

[0100] Where x represents the real discharge voltage data sample, which obeys the real sample distribution P r , z represents random noise, which obeys the uniform distribution z~U(-0.01,0.01), y represents the capacity degradation information as the conditional information of the GAN network, D(·) is the discriminator, D(x|y) represents the score of the discriminator on the input x as real data, G(·) is the generator, G(z|y) is the sample generated by the generator based on random noise z and condition y, λ is the weight parameter, which controls the weight of the gradient penalty term, It is a sample obtained by linear interpolation of the real data x and the generated data G(z), that is, where ∈~U(0,1), Represents the gradient modulus of the discriminator to the input sample;

[0101] The generator loss function is:

[0102]

[0103] in, Represents the synthetic discharge voltage data generated by the generator, satisfying z represents a random noise vector that obeys uniform distribution, y is the capacity degradation condition information, P g To generate data distribution;

[0104] The discriminator loss function is:

[0105]

[0106] Among them, x is the actual discharge voltage data, which obeys the distribution P r , is the interpolation sample, and λ is the gradient penalty coefficient.

[0107] The training process of the above-mentioned degraded data intelligent generation model based on the improved GAN network is as follows:

[0108] Step (1): Randomly initialize the weights of the generator G and the discriminator D;

[0109] Step (2): Fix the generator G and optimize it with the Adam optimizer at a learning rate η D Update the discriminator parameters θ D , minimize the discriminator loss

[0110] Step (3): Fix the discriminator D and use the Adam optimizer with a learning rate η g Update the generator parameters θ G , minimize the generator loss

[0111] Step (4): Repeat steps (2)-(3) until the maximum number of iterations T is reached max ;

[0112] The improved GAN network minimizes the discriminator loss Measure the generated data distribution P g With the real data distribution P r The Wasserstein distance is , where the ratio of the number of discriminator training rounds k to the number of generator training rounds is 2:1, and adversarial training is used to make the data distribution of the generator output close to the real data distribution.

[0113] In the above embodiments, references to "this embodiment" in the specification indicate that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily refer to the same embodiment.

[0114] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the present invention are intended to encompass all such alternatives, modifications, and variations that fall within the broad scope of the appended claims.

[0115] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, any one of the methods in this embodiment is implemented.

[0116] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0117] The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes any one of the methods in this embodiment.

[0118] Regarding the computer-readable storage medium in this embodiment, those skilled in the art will appreciate that all or part of the steps in the aforementioned method embodiments can be implemented using hardware associated with the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the aforementioned method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0119] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used for communication, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes the various steps of the above method.

[0120] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0121] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0122] The present invention can be used in a wide variety of general-purpose or special-purpose computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above.

[0123] The present invention may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for intelligently generating lithium battery degradation data based on an improved generative adversarial network, characterized by: The following steps are involved: Use the battery management system to obtain historical degradation data of lithium batteries, including lithium battery discharge voltage data and lithium battery capacity degradation data; Normalize the capacity degradation data, match the discharge voltage with the capacity degradation data one by one, construct feature-label pairs, and form a training dataset. The long short-term memory (LSTM) network and the improved generative adversarial network (GAN) were trained using the training dataset, respectively, to obtain a battery capacity prediction model based on the LSTM network and a degradation data generation model based on the improved GAN network. Collecting the current cycle discharge voltage curve of the lithium battery, and using the battery capacity prediction model to predict the current capacity of the battery; Design and execute a personalized battery capacity generation program based on the physical characteristics of the target battery, predict the current battery capacity, and generate the desired target capacity; The target capacity is input into the degradation data generation model to generate synthetic degradation data.

2. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 1, characterized in that: The capacity degradation data is normalized as follows: Among them, c t represents the actual capacity measured during t cycle, c max Indicates the rated maximum capacity of the lithium battery in its initial state. Represents the normalized capacity degradation data.

3. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 2, characterized in that: The steps for forming the training data set are as follows: Collect lithium battery discharge voltage data V={v1,v2,v3,…,v n }, and simultaneously record the normalized capacity degradation data within the corresponding cycle period Form a training dataset for health state estimation.

4. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 1, characterized in that: The training process of the battery capacity prediction model based on the LSTM network is as follows: Take the discharge voltage data V as input, the corresponding capacity As labels for supervised training; the input data is organized into three-dimensional tensors (batch_size, sequence_length, input_size) in batches and fed into the LSTM network. The output of the LSTM layer is linearly transformed through a fully connected layer, and the final output is the capacity estimate. By minimizing the error between the estimate and the label, a regression relationship for the degradation behavior is established. The loss function is: Among them E ω Represents the loss function value, Y t Indicates the label value of the t-th capacity, Ψ LSTM represents the LSTM model, x t represents the discharge voltage data of the tth cycle, Θ w Represents the parameter set of the LSTM model; Θ is updated by the backpropagation algorithm w , when E ω The training is stopped when it is less than the threshold ε.

5. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 1, characterized in that: The intelligent generation model of degraded data based on the improved GAN network includes: The generator is composed of a multi-layer perceptron, which receives the random noise vector z and the conditional information y to generate the synthetic discharge voltage data. Among them, θ g Represents the trainable parameters of the generator; The discriminator is composed of an LSTM network and is used to receive input data x. Its formula is: D(x;θ d )→[0,1] Among them, θ d Represents the trainable parameters of the discriminator. The closer the output value is to 1, the higher the probability that the input data is the real discharge voltage data. The value function is: Where x represents the real discharge voltage data sample, which obeys the real sample distribution P r , z represents random noise, which obeys the uniform distribution z~U(-0.01,0.01), y represents the capacity degradation information as the conditional information of the GAN network, D(·) is the discriminator, D(x|y) represents the score of the discriminator on the input x as real data, G(·) is the generator, G(z|y) is the sample generated by the generator based on random noise z and condition y, λ is the weight parameter, which controls the weight of the gradient penalty term, It is a sample obtained by linear interpolation of the real data x and the generated data G(z), that is, where ∈~U(0,1), Represents the gradient modulus of the discriminator to the input sample; The generator loss function is: in, Represents the synthetic discharge voltage data generated by the generator, satisfying z represents a random noise vector that obeys uniform distribution, y is the capacity degradation condition information, P g To generate data distribution; The discriminator loss function is: Among them, x is the actual discharge voltage data, which obeys the distribution P r , is the interpolation sample, and λ is the gradient penalty coefficient.

6. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 5, characterized in that: The training process of the degraded data intelligent generation model based on the improved GAN network is as follows: Step (1): Randomly initialize the weights of the generator G and the discriminator D; Step (2): Fix the generator G and optimize it with the Adam optimizer at a learning rate η D Update the discriminator parameters θ D , minimize the discriminator loss Step (3): Fix the discriminator D and use the Adam optimizer with a learning rate η g Update the generator parameters θ G , minimize the generator loss Step (4): Repeat steps (2)-(3) until the maximum number of iterations T is reached max ; The improved GAN network minimizes the discriminator loss Measure the generated data distribution P g With the real data distribution P r The Wasserstein distance is , where the ratio of the number of discriminator training rounds k to the number of generator training rounds is 2:1, and adversarial training is used to make the data distribution of the generator output close to the real data distribution.

7. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 4, characterized in that: The collecting of the current cycle discharge voltage curve of the lithium battery and the prediction of the current capacity of the battery using the battery capacity prediction model specifically include: The current discharge voltage curve x t Send it to the battery capacity prediction model based on LSTM network to get the current predicted capacity The formula is: Among them, Θ w Solidify the parameter set for the LSTM model, Ψ LSTM Represents the LSTM network mapping relationship including the fully connected layer.

8. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 1, characterized in that: The design of the battery capacity personalized generation program meets the following physical constraints: (a) Generated target capacity value C adv satisfy: 0<C adv ≤C rated Among them C rated is the rated capacity of the battery; (b) The capacity change ΔC between adjacent cycles satisfies: |△C|<δ max where δ max The maximum allowable capacity fluctuation threshold is preset based on the battery chemical characteristics.

9. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 8, characterized in that: The execution process of the battery capacity personalized generation program includes: According to the predicted capacity Generate expected target capacity The formula is: Where GEN(·) is the expected capacity generation function set by the user; The expected capacity generation function GEN(·) is divided into three categories according to different modification forms: incremental modification, decremental modification, or peak modification. The formulas are: Peak modification: Incremental modification: Reduction modification: in, Indicates the preset target capacity. represents the capacity acquired by the LSTM network, ε represents the set modification amplitude, and k(t) represents the continuous modification strategy.

10. The method for intelligently generating lithium battery degradation data based on an improved generative adversarial network according to claim 9, characterized in that: The generating of synthetic degradation data specifically includes: Set the preset target capacity It is fed together with random noise z into the degradation data intelligent generation model based on the improved GAN network to obtain the synthesized discharge voltage data, the formula of which is: where Θ G Represents the parameters of the improved GAN network.

Images