A multi-stage charging control method and system for lithium batteries

Abstract

This invention discloses a multi-stage charging control method and system for lithium batteries, relating to the field of battery management. The method includes: conducting charging tests on the lithium battery model to be analyzed to obtain charging test data; configuring a charging time prediction network based on pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data; setting pre-charge gradient ramp-up constraint step size and constant current gradient descent constraint step size; constructing an unknown population A; constructing a historical population B; receiving the desired charging time selected by the user, performing dual-population optimization to generate a pre-charge current optimized gradient sequence and a constant current optimized gradient sequence; and executing multi-stage charging control of the lithium battery. This method solves the technical problem of existing lithium battery charging control systems failing to achieve a balance between charging efficiency and battery safety, achieving a balance between the two.

Description

Technical Field

[0001] This application relates to the field of battery management, and in particular to a method and system for multi-stage charging control of lithium batteries. Background Technology

[0002] As a crucial energy source for modern electronic devices, the control of the charging process for lithium batteries is vital for ensuring charging efficiency, battery safety, and extending battery life. Traditional lithium battery charging methods primarily utilize constant current / constant voltage charging, which consists of three stages: trickle charging, constant current charging, and constant voltage charging. In the constant current charging stage, the battery is charged with a fixed current until the voltage reaches a preset threshold, after which it transitions to the constant voltage charging stage. This method offers advantages in simplicity and reliability, effectively reducing charging time. However, as charging progresses, the acceptable charging capacity of lithium batteries gradually decreases. Especially in the later stages of charging, continuing to use a large charging current may cause air bubbles to form inside the battery, potentially leading to battery damage or reduced lifespan. Therefore, how to ensure charging efficiency while minimizing the risk of battery damage has become a critical issue that current lithium battery charging technology urgently needs to address.

[0003] Currently, lithium battery charging control faces the technical challenge of failing to achieve a balance between charging efficiency and battery safety. Summary of the Invention

[0004] This application provides a multi-stage charging control method and system for lithium batteries. It employs techniques such as charging tests on the lithium battery model to be analyzed, configuring a charging time prediction network, setting pre-charge gradient ramp constraint step size and constant current gradient descent constraint step size, constructing an unknown population A and a historical population B, and performing dual-population optimization. This achieves the technical effect of finding the optimal charging strategy that can both ensure charging efficiency and reduce the risk of battery damage by precisely controlling the change of charging current, thus achieving a balance between charging efficiency and battery safety.

[0005] This application provides a multi-stage charging control method for lithium batteries, including:

[0006] Charging tests are performed on the lithium battery model to be analyzed to obtain charging test data. The charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data is the current rising gradient, and the constant current gradient sequence recording data is the current falling gradient.

[0007] Configure a charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data;

[0008] Set the pre-charge gradient climb constraint step size and the constant current gradient sink constraint step size;

[0009] Based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, construct an unknown population A;

[0010] Based on the charging test data, a historical population B is constructed;

[0011] The system receives the desired charging duration selected by the user and performs dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence.

[0012] Multi-stage charging control of the lithium battery is performed based on the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

[0013] In one possible implementation, a charging test is performed on the lithium battery model to be analyzed to obtain the charging test data, and then the following processing is performed:

[0014] Configure the pre-charge cutoff voltage, constant current cutoff voltage, and constant voltage cutoff current;

[0015] Set the number of pre-charge current gradients and the number of constant current gradients, wherein the number of pre-charge current gradients is ≥2 and the number of constant current gradients is ≥2;

[0016] Based on the pre-charge cutoff voltage, the constant current cutoff voltage, the constant voltage cutoff current, the number of pre-charge current gradients, and the number of constant current gradients, the lithium battery model to be analyzed is subjected to at least 50,000 charging tests to obtain the charging test data.

[0017] In a possible implementation, based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging duration recording data, a charging duration prediction network is configured to perform the following processing:

[0018] The charging time recording data includes pre-charge time recording data and constant current charging time recording data;

[0019] Based on the pre-charge current gradient sequence recorded data, a first input vector is constructed, and based on the constant current gradient sequence recorded data, a second input vector is constructed.

[0020] The precharge duration prediction subnetwork is trained under supervision using the first input vector and the precharge duration recording data.

[0021] The constant current charging duration prediction subnetwork is trained under supervision using the second input vector and the constant current charging duration recording data.

[0022] A fully connected layer is constructed, which is used to sum the preset constant voltage duration, the first output value of the precharge duration prediction subnetwork, and the second output value of the constant current charging duration prediction subnetwork and output them through the output layer to complete the configuration of the charging duration prediction network.

[0023] In a possible implementation, an unknown population A is constructed based on the pre-charged gradient ascent constraint step size and the constant current gradient descent constraint step size, and the following processing is performed:

[0024] Construct an individual distance evaluation function:

[0025]

[0026]

[0027] Where D1 represents the first deviation distance of the pre-charge current gradient sequence, D2 represents the second deviation distance of the constant current gradient sequence, and I 1,k1 I represents the k-th sequence gradient current value of the first body in the pre-charge current gradient sequence. 1,k2 U(I) represents the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence. 1,k1 The preset cutoff voltage, U(I), represents the gradient current value of the k-th sequence of the first body in the pre-charge current gradient sequence. 1,k2 The preset cutoff voltage characterizes the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence, N is the number of pre-charge current gradients, M is the number of constant current gradients, and I... 2,j1 I represents the gradient current value of the j-th sequence of the first volume in the constant current gradient sequence. 2,j2 U(I) represents the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j2 The preset cutoff voltage, U(I), characterizes the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j1 The preset cutoff voltage characterizing the j-th sequence gradient current value of the first body in the constant current gradient sequence;

[0028] Wherein, the first deviation distance between any two individuals in the unknown population A is greater than or equal to the first deviation distance threshold, and the second deviation distance is greater than or equal to the second deviation distance threshold.

[0029] In a possible implementation, the desired charging duration selected by the user is received. Then, through the charging duration prediction network, dual-population optimization is performed based on the unknown population A and the historical population B to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. The following processing is then performed:

[0030] By using the charging time prediction network, the unknown population A is traversed to obtain the predicted charging time for the unknown population.

[0031] By iterating through the predicted charging time of the unknown population and calculating the distance between the expected charging time, the fitness set of the unknown population is obtained.

[0032] By iterating through the charging duration records and calculating the distance between them and the expected charging duration, the historical population fitness set is obtained.

[0033] The fitness of the unknown population fitness set and the fitness of the historical population fitness set are summed to obtain the fitness summation result;

[0034] Traverse the unknown population fitness set and the historical population fitness set, and compare the sum of the fitness values ​​to obtain the individual elimination probability set;

[0035] Based on the set of individual elimination probabilities, half of the individuals are randomly eliminated to generate a selected population C;

[0036] Based on the selected population C, individuals are expanded in a targeted manner to generate an expanded population D;

[0037] The expanded population D is processed by the charging duration prediction network to obtain the minimum global fitness value;

[0038] When the minimum global fitness value is less than or equal to the fitness convergence threshold, the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence are output.

[0039] In possible implementations, the following processing is performed:

[0040] When the minimum global fitness value is greater than the fitness convergence threshold, determine whether the number of updates for the unknown population A meets the update number threshold.

[0041] When the number of updates for the unknown population A does not meet the update number threshold, the inferior individuals in the selected population C are replaced by the superior individuals in the expanded population D, the historical population B is updated, and the unknown population A is updated by the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, and a new loop is executed.

[0042] In possible implementations, the following processing is performed:

[0043] When the number of updates of the unknown population A meets the update number threshold, the individual corresponding to the minimum global fitness is output and set as the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

[0044] In one possible implementation, individuals are selectively expanded based on the selected population C to generate an expanded population D, and the following processing is performed:

[0045] Based on the selected population C, obtain the individual with the lowest fitness and a preset number of individuals with higher fitness in the selected population;

[0046] Based on the selected individual with the lowest fitness as the target direction, the preset number of individuals with higher fitness are randomly perturbed a certain number of times to generate the expanded population D.

[0047] This application also provides a multi-stage charging control system for lithium batteries, including:

[0048] The charging test data acquisition module is used to perform charging tests on the lithium battery model to be analyzed and obtain charging test data. The charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data is the current rising gradient, and the constant current gradient sequence recording data is the current falling gradient.

[0049] A charging time prediction network configuration module is used to configure a charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data.

[0050] A constraint step size setting module is used to set the pre-charge gradient climbing constraint step size and the constant current gradient sinking constraint step size;

[0051] An unknown population A construction module is used to construct an unknown population A based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size.

[0052] A historical population B construction module is used to construct a historical population B based on the charging test data.

[0053] A dual-population optimization module is used to receive the desired charging duration selected by the user terminal, and perform dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence.

[0054] A lithium battery multi-stage charging control module is used to perform multi-stage charging control of the lithium battery according to the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

[0055] The proposed multi-stage charging control method and system for lithium batteries first involves conducting charging tests on the lithium battery model to be analyzed, obtaining charging test data. This data includes pre-charge current gradient sequence recordings, constant current gradient sequence recordings, and charging time recordings. The pre-charge current gradient sequence recordings represent the current rising gradient, and the constant current gradient sequence recordings represent the current falling gradient. Next, based on these data, a charging time prediction network is configured, and then the pre-charge gradient creepage... The system employs a pre-charge gradient ramp constraint step size and a constant current gradient descent constraint step size. Based on these constraints, an unknown population A is constructed, and a historical population B is constructed based on charging test data. Then, the system receives the user's desired charging duration and uses a charging duration prediction network to perform dual-population optimization based on the unknown population A and the historical population B. This generates a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. Finally, multi-stage charging control of the lithium battery is executed based on these two sequences, achieving a balance between charging efficiency and battery safety. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0057] Figure 1 A flowchart illustrating a multi-stage charging control method for a lithium battery provided in an embodiment of this application;

[0058] Figure 2 This is a schematic diagram of a multi-stage charging control system for a lithium battery provided in an embodiment of this application.

[0059] Explanation of reference numerals in the attached figures: 10 for charging test data acquisition module, 20 for charging duration prediction network configuration module, 30 for constraint step size setting module, 40 for unknown population A construction module, 50 for historical population B construction module, 60 for dual population optimization module, and 70 for lithium battery multi-stage charging control module. Detailed Implementation

[0060] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0063] This application provides a multi-stage charging control method for lithium batteries, such as... Figure 1 As shown, the method includes:

[0064] Step S100: Perform a charging test on the lithium battery model to be analyzed to obtain charging test data. This charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data represents the current increase gradient, and the constant current gradient sequence recording data represents the current decrease gradient. Specifically, the charging test data refers to the current change sequence and charging time recorded during the charging process. Select the lithium battery model to be analyzed, prepare the charging test equipment, set the charging test parameters such as the initial charging current, the final charging current, and the charging voltage, and execute the charging test to obtain the pre-charge current gradient sequence recording data. This pre-charge current gradient sequence recording data represents the process where the current gradually increases from a small value to a preset value in the initial stage of charging, characterized by a current increase gradient. Continue executing the charging test to obtain the constant current gradient sequence recording data. This constant current gradient sequence recording data represents the process where the current gradually decreases under a constant voltage at a certain stage of charging, characterized by a current decrease gradient. Record the charging time for the entire charging process.

[0065] In one possible implementation, a charging test is performed on the lithium battery model to be analyzed to obtain charging test data. Step S100 further includes step S110, configuring a pre-charge cutoff voltage, a constant current cutoff voltage, and a constant voltage cutoff current. Specifically, based on the specifications, material characteristics, and safety requirements of the lithium battery to be analyzed, a pre-charge cutoff voltage is determined. This pre-charge cutoff voltage is used during the pre-charge current stage; when the voltage reaches this value, the pre-charge current stage ends, and the next charging stage begins. Similarly, a constant current cutoff voltage is determined; during the constant current charging stage, when the voltage reaches this value, the constant current charging stage ends. A constant voltage cutoff current is set; during the constant voltage charging stage, when the current drops below this value, the battery is considered fully charged, and the charging process ends. Step S120 sets the number of pre-charge current gradients and the number of constant current gradients, wherein the number of pre-charge current gradients is ≥2, and the number of constant current gradients is ≥2. Specifically, based on the charging characteristics and testing requirements of the lithium battery to be analyzed, a pre-charge current gradient number is set. This pre-charge current gradient number refers to the number of different current change levels during the pre-charge current stage, and this number is at least two, meaning there are at least two different current values ​​used in the pre-charge stage. Similarly, a constant current gradient number is set. This constant current gradient number refers to the number of different current change levels during the constant current charging stage, and this number is also at least two, to meet the requirement of performing multi-gradient current testing during the constant current charging stage. In step S130, based on the pre-charge cutoff voltage, the constant current cutoff voltage, the constant voltage cutoff current, the pre-charge current gradient number, and the constant current gradient number, the lithium battery model to be analyzed is subjected to at least 50,000 charging tests to obtain the charging test data. Specifically, using charging testing equipment, a charging test scheme is designed based on the set pre-charge cutoff voltage, constant current cutoff voltage, constant voltage cutoff current, and the number of gradients for the pre-charge current and constant current. Following this scheme, at least 50,000 charging tests are conducted on the lithium battery under analysis to cover all possible charging conditions. Data such as the charging current gradient sequence and charging time are recorded for each test. This approach, by setting multiple current gradients, comprehensively understands the behavior of the lithium battery under analysis under different charging conditions, obtaining accurate charging characteristic data. By configuring the cutoff voltage and cutoff current, testing is ensured under the premise of battery safety, avoiding battery damage or safety issues caused by overcharging or over-discharging. The at least 50,000 charging tests ensure that the obtained charging test data has sufficient generality to reflect the average charging behavior of the lithium battery under analysis under different conditions. Therefore, this approach achieves the technical effect of obtaining more accurate and comprehensive charging test data.

[0066] Step S200: Configure a charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data. The charging time prediction network is a model based on a neural network, used to predict the charging time of the lithium battery to be analyzed based on a given current gradient sequence. Specifically, the charging test data is preprocessed, such as normalized and denoised, and the preprocessed charging test data is used to train the neural network to predict the charging time under different charging current gradients.

[0067] In one possible implementation, step S200 further includes step S210, whereby the charging duration recording data includes pre-charge duration recording data and constant current charging duration recording data. Specifically, the charging duration of the pre-charge stage and the charging duration of the constant current stage are separated from the charging test data, and the pre-charge duration and constant current charging duration are recorded to form pre-charge duration recording data and constant current charging duration recording data. The pre-charge duration recording data is the time recorded during the pre-charge current stage, from the start of charging to reaching the pre-charge cutoff voltage of the lithium battery to be analyzed; the constant current charging duration recording data is the time recorded during the constant current charging stage, from the end of pre-charge to reaching the constant current cutoff voltage of the lithium battery to be analyzed. Step S220: Based on the pre-charge current gradient sequence recording data, a first input vector is constructed, and based on the constant current gradient sequence recording data, a second input vector is constructed. Specifically, the pre-charge current gradient sequence recording data is encoded according to a certain order or structure to form a first input vector, which is used to input into the pre-charge duration prediction sub-network; the constant current gradient sequence recording data is also encoded according to the same rules or structure to form a second input vector, which is used to input into the constant current charging duration prediction sub-network. Step S230: The pre-charge duration prediction sub-network is trained under supervision using the first input vector and the pre-charge duration recording data. Specifically, the first input vector is used as the input to the pre-charge duration prediction sub-network, and the corresponding pre-charge duration recording data is used as the target output. A supervised learning algorithm (such as backpropagation) is used to train the pre-charge duration prediction sub-network so that it can accurately predict the pre-charge duration. Step S240: The constant current charging duration prediction sub-network is trained under supervision using the second input vector and the constant current charging duration recording data. Specifically, the second input vector is used as the input to the constant current charging time prediction sub-network, and the corresponding constant current charging time recording data is used as the target output. The same supervised learning algorithm is used to train the constant current charging time prediction sub-network so that it can accurately predict the constant current charging time. Step S250: Construct a fully connected layer. The fully connected layer is used to sum the preset constant voltage time, the first output value of the pre-charge time prediction sub-network, and the second output value of the constant current charging time prediction sub-network, and output them through the output layer to complete the configuration of the charging time prediction network. Specifically, a fully connected layer is created. This fully connected layer accepts the outputs of the pre-charge time prediction sub-network and the constant current charging time prediction sub-network as input. In the fully connected layer, the preset constant voltage time is summed with the pre-charge time prediction value and the constant current charging time prediction value. The summed result is output through the output layer as the final charging time prediction value. The preset constant voltage time is the charging time preset during the constant voltage charging stage, determined based on the characteristics and charging requirements of the lithium battery to be analyzed.This implementation reduces the complexity of the problem by decomposing the charging time prediction into two sub-tasks: pre-charging time prediction and constant current charging time prediction. This allows each sub-network to focus on learning the charging time patterns at specific stages, thereby improving the technical effect of prediction accuracy.

[0068] Step S300: Set the pre-charge gradient ramp-up constraint step size and the constant current gradient descent constraint step size. Specifically, based on the charging characteristics and safety requirements of the lithium battery to be analyzed, the pre-charge gradient ramp-up constraint step size is set, which is the maximum allowable step size when the current rises during the pre-charge current stage, to ensure charging safety and efficiency. Similarly, based on the charging characteristics and safety requirements of the lithium battery to be analyzed, the constant current gradient descent constraint step size is set, which is the maximum allowable step size when the current decreases during the constant current stage, also to ensure charging safety and efficiency.

[0069] Step S400: Construct an unknown population A based on the pre-charge gradient ramp constraint step size and the constant current gradient descent constraint step size. Specifically, initialize the unknown population A, which is a set containing multiple possible charging current gradient sequences. Each individual in the unknown population A represents a possible charging current gradient sequence, used to optimize the search process. Constrain the individuals in the unknown population A according to the set pre-charge gradient ramp constraint step size and constant current gradient descent constraint step size to ensure they meet the charging requirements.

[0070] In one possible implementation, step S400 further includes step S410, constructing an individual distance evaluation function:

[0071]

[0072] Where D1 represents the first deviation distance of the pre-charge current gradient sequence, D2 represents the second deviation distance of the constant current gradient sequence, and I 1,k1 I represents the k-th sequence gradient current value of the first body in the pre-charge current gradient sequence. 1,k2 U(I) represents the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence. 1,k1 The preset cutoff voltage, U(I), represents the gradient current value of the k-th sequence of the first body in the pre-charge current gradient sequence. 1,k2 The preset cutoff voltage characterizes the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence, N is the number of pre-charge current gradients, M is the number of constant current gradients, and I... 2,j1 I represents the gradient current value of the j-th sequence of the first volume in the constant current gradient sequence. 2,j2 U(I) represents the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j2The preset cutoff voltage, U(I), characterizes the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j1 The preset cutoff voltage characterizes the j-th sequence gradient current value of the first individual in the constant current gradient sequence. Specifically, the individual distance evaluation function is used to calculate the deviation distance between any two individuals in the unknown population A. The individual distance evaluation function consists of two parts: a first deviation distance of the pre-charge current gradient sequence and a second deviation distance of the constant current gradient sequence. The first deviation distance measures the difference in current gradient between any two individuals in the pre-charge current gradient sequence. It is calculated by comparing the sequence gradient current value of each individual and its corresponding preset cutoff voltage. If the difference in sequence gradient current value or cutoff voltage between two individuals is large, the first deviation distance is large. The second deviation distance is calculated similarly for the constant current gradient sequence, measuring the difference between any two individuals in the constant current stage current gradient sequence. The individual distance evaluation function quantifies the similarity or difference between individuals in the unknown population A, and is used for screening and optimization based on these differences. Step S420, wherein the first deviation distance between any two individuals in the unknown population A is greater than or equal to the first deviation distance threshold, and the second deviation distance is greater than or equal to the second deviation distance threshold. Specifically, the first and second deviation distance thresholds are used to evaluate the differences between any two individuals in the unknown population A on the pre-charge current gradient sequence and the constant current gradient sequence, respectively. The specific selection criteria are: the first deviation distance between any two individuals in the unknown population A must be greater than or equal to the first deviation distance threshold, and the second deviation distance between these two individuals must also be greater than or equal to the second deviation distance threshold. This setting ensures that individuals in the unknown population A have sufficient differences on both the pre-charge current gradient sequence and the constant current gradient sequence, thereby helping to explore a wider solution space during the optimization process and avoiding getting trapped in local optima. This implementation constructs a diverse and efficient unknown population A by building an individual distance evaluation function and setting threshold conditions, achieving the technical effect of making the optimization process more efficient and effective.

[0073] Step S500: Construct a historical population B based on the charging test data. Specifically, extract historical charging current gradient sequences from the charging test data, and construct a historical population B based on these historical data. Historical population B is a set containing multiple historical charging current gradient sequences, where each individual represents a historical charging current gradient sequence, used for comparison and optimization with the unknown population A.

[0074] Step S600: Receive the desired charging duration selected by the user. Using the charging duration prediction network, perform dual-population optimization based on the unknown population A and the historical population B to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. Specifically, receive the desired charging duration input by the user. Use the charging duration prediction network to predict the charging duration of individuals in the unknown population A and the historical population B. Based on the prediction results and the desired charging duration, use a dual-population optimization algorithm (such as a genetic algorithm or particle swarm optimization algorithm) to perform optimization search. After multiple iterations, generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence that meet the desired charging duration.

[0075] In one possible implementation, step S600 further includes step S610, which involves traversing the unknown population A through the charging time prediction network to obtain the predicted charging time of the unknown population; calculating the distance between the predicted charging time of the unknown population and the expected charging time to obtain the fitness set of the unknown population. Specifically, for each individual in the unknown population A, the charging time prediction network is used to predict its charging time, and each predicted charging time is compared with the expected charging time. The distance between them is calculated (e.g., using absolute error, squared error, etc.), and these distances are used as the fitness of each individual in the unknown population A to form the fitness set of the unknown population. Step S620 involves traversing the charging time record data and calculating the distance between it and the expected charging time to obtain the historical population fitness set. Specifically, each record in the charging time record data (representing the historical population) is traversed, and the actual charging time in each record is compared with the expected charging time. The distance between them is calculated, and these distances are used as the fitness of each individual in the historical population to form the fitness set of the historical population. Step S630: Sum the fitness of the unknown population fitness set and the historical population fitness set to obtain a fitness summation result; traverse the unknown population fitness set and the historical population fitness set, and compare the summation result with the fitness summation result to obtain an individual elimination probability set. Specifically, sum the fitness of the unknown population fitness set and the historical population fitness set to obtain a total fitness summation result. Traverse the unknown population fitness set and the historical population fitness set, and compare the fitness of each individual with the total fitness summation result to obtain an individual elimination probability. These elimination probabilities form an individual elimination probability set, which is used for individual elimination operations. Step S640: Randomly eliminate half of the individuals according to the individual elimination probability set to generate a selected population C; perform targeted individual expansion according to the selected population C to generate an expanded population D. Specifically, according to the individual elimination probability set, randomly eliminate half of the individuals from the unknown population A and the historical population B to generate the selected population C. The selected population C is expanded in a targeted manner, i.e., new individuals are generated according to a certain strategy (such as crossover and mutation operations in genetic algorithms) to form an expanded population D. Step S650: The expanded population D is processed by the charging duration prediction network to obtain the minimum global fitness value. Specifically, the charging duration prediction network is used to predict the charging duration of each individual in the expanded population D, and the distance between each predicted charging duration and the expected charging duration is calculated to obtain the fitness of each individual in the expanded population D. From these fitness values, the minimum global fitness value is found, which is the fitness of the optimal individual. Step S660: When the minimum global fitness value is less than or equal to the fitness convergence threshold, the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence are output.Specifically, if the global fitness minimum is less than or equal to the fitness convergence threshold, it indicates that an optimal solution that meets the requirements has been found. The pre-charge current optimization gradient sequence and the constant current optimization gradient sequence corresponding to the global fitness minimum are then output. This implementation method achieves population renewal by eliminating individuals with poor fitness and expanding with new individuals, increasing the diversity of solutions and achieving the technical effect of facilitating rapid searching for the global optimum.

[0076] In one possible implementation, step S600 further includes step S670, where, when the minimum global fitness value is greater than the fitness convergence threshold, it is determined whether the update count of the unknown population A meets the update count threshold. Specifically, if the minimum global fitness value is still greater than the fitness convergence threshold, it is checked whether the update count of the unknown population A has reached a preset update count threshold. The update count threshold is a control parameter used to limit the maximum update count of the unknown population A, preventing the algorithm from getting stuck in an infinite loop or overcomputing. Step S680, when the update count of the unknown population A does not meet the update count threshold, the inferior individuals of the selected population C are replaced by the superior individuals of the expanded population D, the historical population B is updated, and the unknown population A is updated by the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, and a new loop is executed. Specifically, if the number of updates for the unknown population A has not yet reached the update threshold, the update process begins. First, the inferior individuals in the selected population C are replaced by superior individuals from the expanded population D, updating the historical population B. This means that the superior individuals explored in the expanded population D are used to improve the selected population C, increasing its overall fitness level. Then, the unknown population A is updated according to the pre-filled gradient ascent constraint step size and the constant-current gradient descent constraint step size. After completing the above updates, the process returns to step S650 to begin a new loop, continuing to evaluate the fitness of the new population and attempting to find a better solution. This implementation, by adding a judgment on the number of updates for the unknown population A and updating the population and historical data, allows the algorithm to adjust its strategy more flexibly during the optimization process, achieving the technical effect of improving the efficiency and accuracy of finding the optimal solution.

[0077] In one possible implementation, step S600 further includes step S690, whereby when the number of updates of the unknown population A meets the update count threshold, the individual corresponding to the minimum global fitness is output, designated as the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence. Specifically, when the number of updates of the unknown population A meets (i.e. reaches or exceeds) the update count threshold, it indicates that the algorithm has made a sufficient number of iterations, but the minimum global fitness is still greater than the fitness convergence threshold. At this point, the algorithm cannot continue to find a current gradient sequence that satisfies the desired charging time using the current optimization strategy. In this case, the individual corresponding to the current minimum global fitness is output, i.e., the currently found "optimal" pre-charge current optimization gradient sequence and constant current optimization gradient sequence are output. Although these sequences have not reached the expected fitness convergence threshold, they are the best solutions that the algorithm can find under given conditions. This implementation, by setting an update count threshold, allows the algorithm to stop trying after reaching a certain number of iterations, avoiding infinite loops. This achieves the technical effect of ensuring that the algorithm can provide a usable result within a finite time while avoiding the predicament of infinite loops.

[0078] In one possible implementation, based on the selected population C, targeted expansion of individuals is performed to generate an expanded population D. Step S640 further includes step S641, obtaining the individual with the lowest fitness and a preset number of individuals with higher fitness from the selected population C. Specifically, the individual with the lowest fitness is selected from the selected population C. This individual represents the optimal solution in the current population and is therefore selected as the target direction to guide subsequent individual expansion. In addition to the individual with the lowest fitness, a certain number of individuals with higher fitness are selected from the selected population C. Although these individuals are not as good as the individual with the lowest fitness, their presence can increase the diversity of the population and help explore a wider solution space in the subsequent expansion process. Step S642, based on the individual with the lowest fitness in the selected population as the target direction, the preset number of individuals with higher fitness are randomly perturbed a certain number of times to generate the expanded population D. Specifically, the individual with the lowest fitness selected in step S641 is taken as the target direction. Subsequent expansion processes attempt to explore this direction to find a better solution. The individuals with higher fitness selected in step S641 are subjected to random perturbation. This perturbation can be achieved by adding noise, performing small-scale mutations, or generating new individuals through a certain strategy. The aim of this process is to introduce new elements while maintaining a certain level of similarity, hoping to generate individuals with better performance. This random perturbation operation is repeated several times to generate a sufficient number of new individuals, forming the expanded population D. The expanded population D introduces new possibilities through random perturbation, providing more choices for the optimization process. This implementation, by selecting the individual with the lowest fitness as the target direction, ensures that the individuals in the expanded population D have consistency in the optimization direction, achieving the technical effect of generating an expanded population D with excellent performance and diversity.

[0079] Step S700: Perform multi-stage charging control of the lithium battery according to the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence. Specifically, the multi-stage charging control involves performing phased charging control of the lithium battery based on the optimized current gradient sequence to achieve a more efficient and safer charging process. Specifically, the generated pre-charge current optimization gradient sequence and constant current optimization gradient sequence are applied to the lithium battery charging process. In the initial stage of charging, the charging current is gradually increased according to the pre-charge current optimization gradient sequence until a preset current value is reached. Subsequently, a constant current charging stage is entered, maintaining a constant current. The charging current is then gradually decreased according to the constant current optimization gradient sequence. The charging process is monitored to ensure that the charging current and voltage are within the allowable range, and necessary adjustments are made according to the actual situation. When the preset charging termination condition is reached, charging is stopped. This application embodiment employs charging tests on the lithium battery model to be analyzed, configures a charging time prediction network, sets pre-charge gradient ramp constraint step size and constant current gradient descent constraint step size, constructs unknown population A and historical population B, and performs dual-population optimization and other technical means to achieve the best charging strategy that can both ensure charging efficiency and reduce the risk of battery damage by finely controlling the change of charging current, thus achieving the technical effect of balancing charging efficiency and battery safety.

[0080] In the above text, refer to Figure 1 A multi-stage charging control method for a lithium battery according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A multi-stage charging control system for a lithium battery according to an embodiment of the present invention is described.

[0081] A lithium battery multi-stage charging control system according to an embodiment of the present invention addresses the technical problem of existing lithium battery charging control systems failing to achieve a balance between charging efficiency and battery safety, thereby achieving a balance between the two. The lithium battery multi-stage charging control system includes: a charging test data acquisition module 10, a charging duration prediction network configuration module 20, a constraint step size setting module 30, an unknown population A construction module 40, a historical population B construction module 50, a dual-population optimization module 60, and a lithium battery multi-stage charging control module 70.

[0082] The charging test data acquisition module 10 is used to perform charging tests on the lithium battery model to be analyzed and obtain charging test data. The charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data is the current rising gradient, and the constant current gradient sequence recording data is the current falling gradient.

[0083] The charging time prediction network configuration module 20 is used to configure the charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data.

[0084] The constraint step size setting module 30 is used to set the constraint step size of the pre-charge gradient climb and the constraint step size of the constant current gradient sink.

[0085] The unknown population A construction module 40 is used to construct the unknown population A according to the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size;

[0086] The historical population B construction module 50 is used to construct historical population B based on the charging test data;

[0087] The dual-population optimization module 60 is used to receive the desired charging duration selected by the user terminal, and perform dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence.

[0088] The lithium battery multi-stage charging control module 70 is used to perform multi-stage charging control of the lithium battery according to the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

[0089] The specific configuration of the charging test data acquisition module 10 will be described in detail below. As mentioned above, to perform charging tests on the lithium battery model to be analyzed and obtain charging test data, the charging test data acquisition module 10 may further include: a configuration unit for configuring the pre-charge cutoff voltage, constant current cutoff voltage, and constant voltage cutoff current; a current gradient quantity setting unit for setting the pre-charge current gradient quantity and the constant current gradient quantity, wherein the pre-charge current gradient quantity is ≥2 and the constant current gradient quantity is ≥2; and a charging test unit for performing charging tests on the lithium battery model to be analyzed at least 50,000 times based on the pre-charge cutoff voltage, the constant current cutoff voltage, the constant voltage cutoff current, the pre-charge current gradient quantity, and the constant current gradient quantity to obtain the charging test data.

[0090] The specific configuration of the charging time prediction network configuration module 20 will be described in detail below. As mentioned above, the charging time prediction network is configured based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data. The charging time prediction network configuration module 20 may further include: a charging time recording data construction unit for the charging time recording data including pre-charge time recording data and constant current charging time recording data; an input vector construction unit for constructing a first input vector based on the pre-charge current gradient sequence recording data and a second input vector based on the constant current gradient sequence recording data; and pre-charge time prediction... The sub-network training unit is used to supervise the training of the pre-charge duration prediction sub-network using the first input vector and the pre-charge duration recording data; the constant current charging duration prediction sub-network training unit is used to supervise the training of the constant current charging duration prediction sub-network using the second input vector and the constant current charging duration recording data; the fully connected layer construction unit is used to construct a fully connected layer, which is used to sum the preset constant voltage duration, the first output value of the pre-charge duration prediction sub-network, and the second output value of the constant current charging duration prediction sub-network and output them through the output layer to complete the configuration of the charging duration prediction network.

[0091] The specific configuration of the unknown population A construction module 40 will be described in detail below. As mentioned above, the unknown population A is constructed based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size. The unknown population A construction module 40 may further include: an individual distance evaluation function construction unit for constructing the individual distance evaluation function.

[0092]

[0093] Where D1 represents the first deviation distance of the pre-charge current gradient sequence, D2 represents the second deviation distance of the constant current gradient sequence, and I 1,k1 I represents the k-th sequence gradient current value of the first body in the pre-charge current gradient sequence. 1,k2 U(I) represents the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence. 1,k1 The preset cutoff voltage, U(I), represents the gradient current value of the k-th sequence of the first body in the pre-charge current gradient sequence. 1,k2 The preset cutoff voltage characterizes the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence, N is the number of pre-charge current gradients, M is the number of constant current gradients, and I... 2,j1 I represents the gradient current value of the j-th sequence of the first volume in the constant current gradient sequence. 2,j2 U(I) represents the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j2The preset cutoff voltage, U(I), characterizes the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. 2,j1 The preset cutoff voltage characterizes the gradient current value of the j-th sequence of the first individual in the constant current gradient sequence, wherein the first deviation distance between any two individuals in the unknown population A is greater than or equal to the first deviation distance threshold, and the second deviation distance is greater than or equal to the second deviation distance threshold.

[0094] The specific configuration of the dual-population optimization module 60 will be described in detail below. As mentioned above, the module receives the desired charging duration selected by the user and performs dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. The dual-population optimization module 60 may further include: an unknown population charging prediction duration acquisition unit for traversing the unknown population A through the charging duration prediction network to obtain the unknown population charging prediction duration; an unknown population fitness set acquisition unit for traversing the unknown population charging prediction duration and calculating the distance between the desired charging duration to obtain the unknown population fitness set; a historical population fitness set acquisition unit for traversing the charging duration record data and calculating the distance between the desired charging duration to obtain the historical population fitness set; and a fitness summing unit for summing the unknown population fitness. The fitness set is compared with the fitness set of the historical population to obtain the fitness summation result; the individual elimination probability set acquisition unit is used to traverse the unknown population fitness set and the historical population fitness set, and compare them with the fitness summation result to obtain the individual elimination probability set; the selected population C generation unit is used to randomly eliminate half of the individuals according to the individual elimination probability set to generate the selected population C; the expanded population D generation unit is used to perform targeted individual expansion according to the selected population C to generate the expanded population D; the global fitness minimum value acquisition unit is used to process the expanded population D through the charging time prediction network to obtain the global fitness minimum value; the output unit is used to output the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence when the global fitness minimum value is less than or equal to the fitness convergence threshold.

[0095] The dual-population optimization module 60 may further include: a judgment unit used to determine whether the number of updates of the unknown population A meets the update number threshold when the minimum global fitness value is greater than the fitness convergence threshold; and a loop execution unit used to replace the inferior individuals of the selected population C with the superior individuals of the expanded population D, update the historical population B, update the unknown population A with the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, and execute a new loop when the number of updates of the unknown population A does not meet the update number threshold.

[0096] The dual-population optimization module 60 may further include: an individual output unit for outputting the individual corresponding to the minimum global fitness value when the number of updates of the unknown population A meets the update number threshold, which is set as the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

[0097] The process of expanding an expanded population D by selectively expanding individuals from the selected population C can be further divided into: an individual acquisition subunit for obtaining individuals with the lowest fitness and a preset number of individuals with higher fitness from the selected population C; and a random perturbation subunit for randomly perturbing the preset number of individuals with higher fitness a certain number of times, with the individuals with the lowest fitness as the target direction, to generate the expanded population D.

[0098] The lithium battery multi-stage charging control system provided in this embodiment of the invention can execute the lithium battery multi-stage charging control method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0099] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0100] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A multi-stage charging control method for lithium batteries, characterized in that, include: Charging tests are performed on the lithium battery model to be analyzed to obtain charging test data. The charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data is the current rising gradient, and the constant current gradient sequence recording data is the current falling gradient. Configure a charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data; Set the pre-charge gradient climb constraint step size and the constant current gradient sink constraint step size; Based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, construct an unknown population A; Based on the charging test data, a historical population B is constructed; The system receives the desired charging duration selected by the user and performs dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. Multi-stage charging control of the lithium battery is performed based on the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence. The unknown population A is constructed based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, including: Construct an individual distance evaluation function: ； ； in, The first deviation distance characterizes the pre-charge current gradient sequence. The second deviation distance characterizes the constant current gradient sequence. The k-th sequence gradient current value characterizing the first volume of the pre-charge current gradient sequence. Characterizing the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence. The preset cutoff voltage characterizing the k-th sequence gradient current value of the first body in the pre-charge current gradient sequence. The preset cutoff voltage characterizes the k-th sequence gradient current value of the second individual in the pre-charge current gradient sequence, where N is the number of pre-charge current gradients and M is the number of constant current gradients. The gradient current value of the j-th sequence of the first volume characterizing the constant current gradient sequence. The gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. The preset cutoff voltage characterizing the gradient current value of the j-th sequence of the second individual in the constant current gradient sequence. The preset cutoff voltage characterizing the j-th sequence gradient current value of the first body in the constant current gradient sequence; Wherein, the first deviation distance between any two individuals in the unknown population A is greater than or equal to the first deviation distance threshold, and the second deviation distance is greater than or equal to the second deviation distance threshold.

2. The lithium battery multi-stage charging control method as described in claim 1, characterized in that, Charging tests were conducted on the lithium battery model to be analyzed to obtain charging test data, including: Configure the pre-charge cutoff voltage, constant current cutoff voltage, and constant voltage cutoff current; Set the number of pre-charge current gradients and the number of constant current gradients, wherein the number of pre-charge current gradients is ≥2 and the number of constant current gradients is ≥2; Based on the pre-charge cutoff voltage, the constant current cutoff voltage, the constant voltage cutoff current, the number of pre-charge current gradients, and the number of constant current gradients, the lithium battery model to be analyzed is subjected to at least 50,000 charging tests to obtain the charging test data.

3. The lithium battery multi-stage charging control method as described in claim 2, characterized in that, Based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging duration recording data, a charging duration prediction network is configured, including: The charging time recording data includes pre-charge time recording data and constant current charging time recording data; Based on the pre-charge current gradient sequence recorded data, a first input vector is constructed, and based on the constant current gradient sequence recorded data, a second input vector is constructed. The precharge duration prediction subnetwork is trained under supervision using the first input vector and the precharge duration recording data. The constant current charging duration prediction subnetwork is trained under supervision using the second input vector and the constant current charging duration recording data. A fully connected layer is constructed, which is used to sum the preset constant voltage duration, the first output value of the precharge duration prediction subnetwork, and the second output value of the constant current charging duration prediction subnetwork and output them through the output layer to complete the configuration of the charging duration prediction network.

4. The lithium battery multi-stage charging control method as described in claim 1, characterized in that, The system receives the desired charging duration selected by the user, and uses the charging duration prediction network to perform dual-population optimization based on the unknown population A and the historical population B, generating a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence, including: By using the charging time prediction network, the unknown population A is traversed to obtain the predicted charging time for the unknown population. By iterating through the predicted charging time of the unknown population and calculating the distance between the expected charging time, the fitness set of the unknown population is obtained. By iterating through the charging duration records and calculating the distance between them and the expected charging duration, the historical population fitness set is obtained. The fitness of the unknown population fitness set and the fitness of the historical population fitness set are summed to obtain the fitness summation result; Traverse the unknown population fitness set and the historical population fitness set, and compare the sum of the fitness values ​​to obtain the individual elimination probability set; Based on the set of individual elimination probabilities, half of the individuals are randomly eliminated to generate a selected population C; Based on the selected population C, individuals are expanded in a targeted manner to generate an expanded population D; The expanded population D is processed by the charging duration prediction network to obtain the minimum global fitness value; When the minimum global fitness value is less than or equal to the fitness convergence threshold, the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence are output.

5. The lithium battery multi-stage charging control method as described in claim 4, characterized in that, Also includes: When the minimum global fitness value is greater than the fitness convergence threshold, determine whether the number of updates for the unknown population A meets the update number threshold. When the number of updates for the unknown population A does not meet the update number threshold, the inferior individuals in the selected population C are replaced by the superior individuals in the expanded population D, the historical population B is updated, and the unknown population A is updated by the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size, and a new loop is executed.

6. The lithium battery multi-stage charging control method as described in claim 5, characterized in that, Also includes: When the number of updates of the unknown population A meets the update number threshold, the individual corresponding to the minimum global fitness is output and set as the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

7. The lithium battery multi-stage charging control method as described in claim 4, characterized in that, Based on the selected population C, individuals are selectively expanded to generate an expanded population D, including: Based on the selected population C, obtain the individual with the lowest fitness and a preset number of individuals with higher fitness in the selected population; Based on the selected individual with the lowest fitness as the target direction, the preset number of individuals with higher fitness are randomly perturbed a certain number of times to generate the expanded population D.

8. A multi-stage charging control system for lithium batteries, characterized in that, The system is used to implement the multi-stage charging control method for lithium batteries according to any one of claims 1-7, the system comprising: The charging test data acquisition module is used to perform charging tests on the lithium battery model to be analyzed and obtain charging test data. The charging test data includes pre-charge current gradient sequence recording data, constant current gradient sequence recording data, and charging time recording data. The pre-charge current gradient sequence recording data is the current rising gradient, and the constant current gradient sequence recording data is the current falling gradient. A charging time prediction network configuration module is used to configure a charging time prediction network based on the pre-charge current gradient sequence recording data, the constant current gradient sequence recording data, and the charging time recording data. A constraint step size setting module is used to set the pre-charge gradient climbing constraint step size and the constant current gradient sinking constraint step size; An unknown population A construction module is used to construct an unknown population A based on the pre-charged gradient climbing constraint step size and the constant current gradient sinking constraint step size. A historical population B construction module is used to construct a historical population B based on the charging test data. A dual-population optimization module is used to receive the desired charging duration selected by the user terminal, and perform dual-population optimization based on the unknown population A and the historical population B through the charging duration prediction network to generate a pre-charge current optimization gradient sequence and a constant current optimization gradient sequence. A lithium battery multi-stage charging control module is used to perform multi-stage charging control of the lithium battery according to the pre-charge current optimization gradient sequence and the constant current optimization gradient sequence.

Images