Management method, device and equipment of energy storage battery pack, medium and product

Abstract

The embodiment of the invention discloses a management method and device of an energy storage battery pack, equipment, a medium and a product. The management method comprises the following steps: determining an adjustment strategy corresponding to each battery unit according to a working mode corresponding to each battery unit; dynamically adjusting the current of the corresponding battery unit according to each adjustment strategy to determine a current adjustment result; determining a temperature change result according to the preset operation parameters, the environment temperature and the current adjustment result; determining input data corresponding to the energy storage battery pack according to the current adjustment result and the temperature change result; the input data comprises an average state of charge, an average temperature, an average current absolute value and a state of charge standard deviation; and based on the input data, determining a target charge state by using the target neural network model, so as to dynamically manage the charge-discharge process of each battery unit in the energy storage battery pack according to the target charge state. According to the technical scheme, refined and intelligent management of charging and discharging of each unit in the energy storage battery pack is realized.

Description

Technical Field

[0001] This disclosure relates to the field of battery management technology, and in particular to a management method, apparatus, device, medium, and product for energy storage battery packs. Background Technology

[0002] With the widespread application of renewable energy, lithium-ion batteries have become the mainstream energy storage technology in modern power systems, widely used in portable electronic devices, electric vehicles, and grid-scale energy storage systems. Due to their high energy density, low self-discharge rate, and long cycle life, lithium-ion batteries hold an irreplaceable position in various scenarios. However, in actual operation, improper operations such as overcharging and over-discharging can lead to irreversible degradation of battery performance and even safety risks. Therefore, ensuring the safe and efficient operation of lithium-ion batteries relies on real-time monitoring and control of their internal state. Among these, the equalization management of the State of Charge (SoC), dynamic temperature control, and minimization of energy loss are the core functions of the Battery Management System (BMS), and their control accuracy directly affects battery life, energy utilization efficiency, and system operational reliability.

[0003] Traditional BMS design relies on physical experiments, but these experiments are time-consuming, costly, and difficult to cover all complex operating conditions. Therefore, dynamic simulation technology based on mathematical models has become a key means to optimize BMS design. These simulation tools typically use equivalent circuit models (ECMs) or electrochemical models (such as pseudo-two-dimensional models, P2D) to numerically solve for changes in battery voltage, current, system-on-chip (SoC), and temperature, helping engineers evaluate the impact of different control strategies on battery performance.

[0004] In the existing technology, there are various BMS simulation tools and methods for modeling battery behavior and optimizing control strategies. These existing solutions provide a foundation for BMS simulation, but they generally suffer from problems such as single strategy support, simplified temperature models, and insufficient intelligence, and cannot comprehensively evaluate battery performance under the synergy of multiple strategies. Summary of the Invention

[0005] This disclosure provides a management method, apparatus, equipment, medium, and product for energy storage battery packs, enabling refined and intelligent management of the charging and discharging of each unit within the energy storage battery pack.

[0006] In a first aspect, a method for managing an energy storage battery pack is provided, the energy storage battery pack including at least one battery cell, the method comprising:

[0007] The adjustment strategy for each battery cell is determined according to its corresponding operating mode; the adjustment strategy is a strategy for adjusting voltage and / or current.

[0008] According to each adjustment strategy, the current of the corresponding battery cell is dynamically adjusted to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell.

[0009] The temperature change result is determined based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell.

[0010] The input data corresponding to the energy storage battery pack is determined based on the current adjustment results and the temperature change results; the input data includes the average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge.

[0011] Based on the input data, a target state of charge is determined using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

[0012] Secondly, a management device for an energy storage battery pack is provided, the energy storage battery pack including at least one battery cell, the device comprising:

[0013] The adjustment strategy determination module is used to determine the corresponding adjustment strategy for each battery cell based on the operating mode of each battery cell; the adjustment strategy is a strategy for adjusting voltage and / or current.

[0014] The current adjustment result determination module is used to dynamically adjust the current of the corresponding battery cell according to each adjustment strategy in order to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell.

[0015] The temperature change result determination module is used to determine the temperature change result based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell.

[0016] The input data determination module is used to determine the input data corresponding to the energy storage battery pack based on the current adjustment result and the temperature change result; the input data includes average state of charge, average temperature, absolute value of average current and standard deviation of state of charge.

[0017] The management module is used to determine the target state of charge based on the input data using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

[0018] Thirdly, an electronic device is provided, comprising:

[0019] At least one processor; and,

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the energy storage battery pack management method as described in the first aspect above.

[0022] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the management method for the energy storage battery pack as described in the first aspect above.

[0023] Fifthly, a computer program product is provided, the computer program product including a computer program that, when executed by a processor, implements the energy storage battery pack management method as described in the first aspect above.

[0024] This disclosure provides a management method, apparatus, device, medium, and product for an energy storage battery pack. The energy storage battery pack includes at least one battery cell. The method includes: determining a corresponding adjustment strategy for each battery cell based on its corresponding operating mode; the adjustment strategy is a strategy for adjusting voltage and / or current; dynamically adjusting the current of the corresponding battery cell according to each adjustment strategy to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell; determining a temperature change result based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell; determining input data for the energy storage battery pack based on the current adjustment result and the temperature change result; the input data includes average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge; and determining a target state of charge using a target neural network model based on the input data, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge. This technical solution dynamically generates charging current / voltage boundaries and adjusts the current by collecting the working mode, ambient temperature and operating parameters of each battery cell in real time, thereby predicting temperature changes. Finally, it extracts four-dimensional inputs: average SOC, average temperature, average current and SOC standard deviation, and outputs the optimal target SOC by the target neural network model, thus realizing refined and intelligent management of the charging and discharging of each cell in the energy storage battery pack.

[0025] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this disclosure, nor is it intended to limit the scope of the embodiments of this disclosure. Other features of the embodiments of this disclosure will become readily apparent from the following description. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart of a management method for an energy storage battery pack provided in Embodiment 1 of this disclosure;

[0028] Figure 2 This is a schematic diagram of adjusting the state of charge according to a PID controller and a CC strategy provided in Embodiment 1 of this disclosure;

[0029] Figure 3 This is a schematic diagram of a process for adjusting the state of charge according to a CV strategy provided in Embodiment 1 of this disclosure;

[0030] Figure 4 This is a method provided in Embodiment 1 of the present disclosure according to CC A schematic diagram illustrating the process of adjusting the state of charge using the CV strategy;

[0031] Figure 5 This is a schematic diagram of a process for adjusting the state of charge according to a DRL strategy provided in Embodiment 1 of this disclosure;

[0032] Figure 6 This is a schematic diagram of the structure of a management device for an energy storage battery pack provided in Embodiment 2 of this disclosure;

[0033] Figure 7 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of this disclosure. Detailed Implementation

[0034] To enable those skilled in the art to better understand the solutions of the embodiments of this disclosure, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the protection scope of the embodiments of this disclosure.

[0035] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0036] Example 1

[0037] Figure 1 This is a flowchart illustrating a management method for an energy storage battery pack according to Embodiment 1 of this disclosure. This embodiment is applicable to situations involving the management of energy storage battery packs. The method can be executed by a management device for the energy storage battery pack, which can be implemented in hardware and / or software. This management device can be configured in an electronic device, including but not limited to computers, PCs, electronic devices, and servers, which are devices with data processing capabilities. Figure 1 As shown, the method includes:

[0038] S110. Determine the corresponding adjustment strategy for each battery cell based on the operating mode of each battery cell; the adjustment strategy is a strategy for adjusting voltage and / or current.

[0039] In this embodiment, a corresponding adjustment strategy can be determined based on the operating mode of each battery cell in the energy storage battery pack. This adjustment strategy can be a strategy for adjusting voltage and / or current; for example, the adjustment strategy may include a proportional adjustment. integral Differential controller (Proportional) Integral Derivative Controller (PID controller), Constant Current Controller Current (CC), constant voltage (Constant) Voltage (CV), constant current Constant pressure CurrentConstant Voltage, CC CV) and deep reinforcement learning ( Strategies such as constant current (DC) and constant voltage (CV) are employed. A PID controller can be a linear controller. It determines the control deviation based on the given value and the actual output value, and then uses a linear combination of the proportional (P), integral (I), and derivative (D) of the deviation to form the control quantity, thereby controlling the controlled object. A constant current (CC) strategy maintains a constant current during charging or discharging. For example, in battery charging, a constant current charging method keeps the charging current at a fixed value. A constant voltage (CV) strategy maintains a constant voltage during charging or discharging. For example, in battery charging, once the battery voltage reaches a certain value, a constant voltage charging method is used, keeping the charging voltage constant. Constant current-constant voltage (CC) CV (Constant Current) strategy can be a method that combines constant current and constant voltage control strategies. During charging, constant current charging is used first, and then constant voltage charging is switched when the battery voltage reaches a certain value. DRL (Deep Learning) strategy can be a strategy that combines deep learning and reinforcement learning. It uses deep neural networks to approximate the value function or policy function in reinforcement learning, and learns the optimal control strategy through interaction with the environment.

[0040] As described above, the operating modes can include charging mode and discharging mode. Charging mode can be understood as an external power source inputting current into the battery cell, and the total energy of the battery cell increases. Discharging mode can be understood as the battery cell outputting power to the load, and the total energy of the battery cell decreases.

[0041] S120. According to each adjustment strategy, dynamically adjust the current of the corresponding battery cell to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell.

[0042] Specifically, after the adjustment strategy corresponding to each battery cell is determined, the current of the corresponding battery cell is dynamically adjusted according to the adjustment strategy of each battery cell, and the current adjustment result can be determined. The current adjustment result can include the adjusted current corresponding to each battery cell.

[0043] S130. Determine the temperature change result based on the preset operating parameters, ambient temperature, and current adjustment result; the temperature change result includes the temperature change data corresponding to each battery cell.

[0044] In this embodiment, preset operating parameters can be obtained. These preset operating parameters can be pre-set parameters, including heat generation coefficient, heat dissipation coefficient, and internal resistance of each battery cell. The preset operating parameters may also include parameters such as initial SoC, voltage, internal resistance, and ambient temperature. The preset operating parameters can provide initial conditions for the energy storage battery pack, enabling each battery cell in the pack to begin operation.

[0045] For example, 10 battery cells can be defined, each with a capacity of 1000 amp-hours and an internal resistance of 0.01 ohms. The simulation time can be set to 10 hours (36,000 seconds), with a time step of 0.1 seconds, for a total of 360,000 steps, and the ambient temperature can be set to 25 degrees Celsius. A linear relationship between open-circuit voltage and state of charge is defined to reflect the characteristics of lithium-ion battery voltage from 3.2 volts to 4.2 volts. A total current demand curve is generated: 200 amps for the first 5 hours of discharge, and -200 amps for the next 5 hours of charging, initializing the state of charge of each cell as follows. The temperature is The simulation uses degrees Celsius to represent the initial state differences between individual battery cells. Preset operating parameters allow setting the basic simulation conditions; data is stored in an efficient array to ensure consistency and computational efficiency. These preset operating parameters can include battery characteristics, time settings, and current requirements for later use.

[0046] Following the above description, after the preset operating parameters are determined, the temperature change data corresponding to each battery cell within a set time step can be determined based on the preset operating parameters, ambient temperature, and current adjustment results. Here, the ambient temperature can be the temperature of the environment where the battery cell is located, and the temperature change data refers to the increase or decrease in battery cell temperature within each time step. This can be calculated by comprehensively considering the two main factors of heat generation and heat dissipation. The specific formula can be expressed as:

[0047]

[0048] in, The temperature rise could be caused by the heat generated when current flows through the battery's internal resistance. This could be due to a decrease in temperature caused by heat exchange between the battery and the environment. Specifically, The calculation formula can be expressed as:

[0049]

[0050] Where I can be the current (the adjusted current in the current adjustment result), R can be the battery internal resistance (unit: ohms), and the heat generation coefficient can be... , It can be the time step (unit: seconds). It can be represented as:

[0051]

[0052] Wherein, the heat dissipation coefficient is , This is the current battery temperature (units can be...). ), It's the ambient temperature. The temperature change at each step. It can be a dynamic value that reflects the changing trend of battery temperature under current operating conditions (current, internal resistance, ambient temperature, etc.). By updating the temperature change at each time step in real time, the thermal effects of the battery under high-power scenarios can be accurately simulated, providing crucial temperature information for the battery management system to ensure safe battery operation and performance optimization.

[0053] S140. Determine the input data corresponding to the energy storage battery pack based on the current adjustment results and temperature change results; the input data includes the average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge.

[0054] Specifically, after obtaining the current adjustment results and temperature change results, the state of charge, voltage, temperature, energy loss, and standard deviation of the state of charge for each battery cell can be determined based on these results.

[0055] As described above, the input data for the energy storage battery pack can be determined by the state of charge, voltage, temperature, energy loss, and standard deviation of state of charge for each battery cell. These input data include the average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge.

[0056] S150. Based on the input data, the target state of charge is determined using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

[0057] Specifically, after obtaining the input data, the input data can be fed into the target neural network model, which determines the target state of charge of the energy storage battery pack. The target neural network model is a pre-trained neural network that may include input nodes, hidden layers, and output nodes.

[0058] For example, the target neural network model can be a feedforward neural network that includes 4 input nodes (average state of charge, average temperature, average absolute current, and standard deviation of state of charge), two hidden layers (64 and 32 nodes respectively, using the ReLU activation function), and 1 output node (using the Sigmoid activation function to output the optimal state of charge (target state of charge)).

[0059] As described above, the target state of charge (SOC) can be the optimal state of discharge (SOC). After obtaining the optimal SOC, the charging and discharging processes of each battery cell in the energy storage battery pack can be dynamically managed based on the target SOC. For example, by using the optimal SOC as the target SOC, the charging and discharging commands of each battery cell in the battery pack can be adjusted in real time according to the target SOC, ensuring that each battery cell always operates around that SOC value.

[0060] This embodiment provides a management method for an energy storage battery pack, including: determining a corresponding adjustment strategy for each battery cell based on its operating mode; the adjustment strategy being a strategy for adjusting voltage and / or current; dynamically adjusting the current of the corresponding battery cell according to each adjustment strategy to determine the current adjustment result; the current adjustment result including the adjusted current of each battery cell; determining a temperature change result based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result including temperature change data corresponding to each battery cell; determining input data for the energy storage battery pack based on the current adjustment result and the temperature change result; the input data including average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge; and determining a target state of charge using a target neural network model based on the input data, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge. The above technical solution achieves refined and intelligent management of the charging and discharging of each cell within the energy storage battery pack.

[0061] As an optional implementation of this embodiment, the energy storage battery pack management method provided in this embodiment further includes:

[0062] 1) Obtain the current information corresponding to each battery cell.

[0063] Specifically, it can obtain the current information of each battery cell in the energy storage battery pack, and the current information can be the total current of each battery cell.

[0064] 2) Based on the current information, determine the operating mode corresponding to each battery cell; the operating mode includes charging mode and discharging mode.

[0065] Specifically, after obtaining the current information corresponding to each battery cell, the individual cells can be determined based on the current information.

[0066] Each battery cell corresponds to a specific operating mode, which may include charging mode and discharging mode. For example, if the current information is the total current of all battery cells, for any given battery cell, if the total current of the battery cell is greater than or equal to 0, the operating mode of the battery cell can be considered to be discharging mode; if the current of the battery cell is less than 0, the operating mode of the battery cell can be considered to be charging mode.

[0067] As an optional implementation of this embodiment, the adjustment strategy includes a constant current adjustment strategy, a constant voltage adjustment strategy, a constant current-constant voltage adjustment strategy, a deep learning reinforcement strategy, and a proportional-integral-derivative PID controller; the step of determining the corresponding adjustment strategy for each battery cell based on the operating mode of each battery cell includes:

[0068] 1) For any battery cell, if the operating mode of the battery cell is the discharge mode, then the proportional-integral-derivative PID controller is determined as the adjustment strategy corresponding to the battery cell.

[0069] Specifically, for any single battery cell in an energy storage battery pack, a corresponding adjustment strategy can be determined based on the cell's operating mode. If the battery cell's operating mode is discharge mode, a PID controller can be used as the corresponding adjustment strategy. The PID controller can be an automatic adjustment device that simultaneously corrects errors based on three aspects: current error, accumulated past error, and future error trend. Its purpose is to quickly and stably maintain the actual value at the target value. For example, a proportional controller can be used... integral A differential (PID) controller balances the state of charge (SOC) of each unit, with a proportional gain of 50, an integral gain of 0.05, and a derivative gain of 10. It calculates the deviation of each unit's SOC from the average value, tracking the accumulation (integral) and rate of change (derivative) of the deviation. The current is adjusted based on the deviation to ensure that units with high SOC discharge more and units with low SOC discharge less. The current for each unit is based on the total current divided equally and then adjusted, thus limiting the current to within a certain range. Between 300 and 300 amperes.

[0070] 2) If the battery cell is in the charging mode, then based on the working state of the battery cell, the constant current adjustment strategy, the constant voltage adjustment strategy, the constant current-constant voltage adjustment strategy, and / or the deep learning reinforcement strategy are determined as the adjustment strategies corresponding to the battery cell.

[0071] Specifically, if the battery cell is in charging mode, then constant current adjustment strategy, constant voltage adjustment strategy, constant current-constant voltage adjustment strategy and / or deep learning reinforcement strategy can be used as the corresponding adjustment strategy for the battery cell according to the working state of the battery cell.

[0072] For example, a constant current regulation strategy (CC) allows each unit to be charged with a constant current. Charging stops when the state of charge reaches a preset current threshold to prevent overcharging. A constant voltage regulation strategy (CV) assumes a target voltage of 4.2 volts. Based on the difference between the current voltage and the target voltage, proportional-integral control is used to adjust the current, limiting it to... 100 to 0 amperes. Constant current-constant voltage adjustment strategy (CC) CV): Initially based on The system operates at 100 amps. When the state of charge exceeds a preset current threshold or the voltage exceeds a preset voltage threshold, it switches to constant voltage mode, adjusting the current individually for each cell. Deep reinforcement learning (DRL) dynamically adjusts the current based on the state of charge deviation and the temperature difference with ambient temperature, prioritizing the balancing of cells with low state of charge and controlling overheating.

[0073] It is known that these strategies can be applied dynamically, specified in simulation parameters or automatically switched by the system based on the current battery state (such as SoC level, temperature) to achieve dynamic current distribution and prevent overcharging. For example, CC is suitable for initial fast charging, CV is used for later voltage maintenance, and CC... CV is a combination, while DRL uses reinforcement learning for intelligent optimization. These adjustment strategies can be viewed as a kind of control constraint, because each strategy defines the boundaries and rules of current / voltage to ensure that the charging process is safe, efficient, and does not exceed the physical limits of the battery.

[0074] Figure 2 This embodiment provides a schematic diagram of adjusting the state of charge based on a PID controller and a CC strategy, as shown below. Figure 2 As shown, the curves of different colors correspond to Unit 1 respectively. Unit 10 can be used to represent different battery cells. The initial SoC value differs for each battery cell, ranging from 0.62 to 0.85 (initial data: This reflects the initial inconsistency of the battery cells. During the discharge phase (0 to 18000 seconds): the curve changes from the initial value, showing that constant current discharge (200 amps) causes a decrease in SoC. The curves converge, indicating that the PID controller adjusts the current ( The current is gradually balanced across the SoCs from 300 to 300 amps. During the charging phase (18,000 to 36,000 seconds): the curve begins to level off after 18,000 seconds, indicating constant current charging. The charge level stopped at 100 amps, conforming to the overcharge protection mechanism of the CC strategy. The SoC stabilized between 0.75 and 0.85, without further increase, reflecting the effectiveness of the adjustment strategy logic. The optimal SoC is 0.815.

[0075] Figure 3 This embodiment provides a schematic diagram of a process for adjusting the state of charge according to a CV strategy, as shown below. Figure 3 As shown, the curves of different colors correspond to Unit 1 respectively. Unit 10 can be used to represent different battery cells. The state of charge (SoC) starts from an initial value (range 0.62 to 0.85), which varies depending on the individual cells. Under a constant voltage (CV) control strategy, the SoC of the 10 battery cells changes over time (seconds). The optimal SoC value is 0.654. ​​The curves represent the SoC change trajectory of different battery cells, gradually decreasing from an initial high SoC value and stabilizing after approximately 17,000 seconds. This demonstrates the dynamic optimization effect of PID control on inter-cell consistency and indicates the influence of discharge behavior and temperature effects on SoC equalization in CV mode within a multi-strategy collaborative system.

[0076] Figure 4 This embodiment provides a schematic diagram of a process for adjusting the state of charge according to a CC-CV strategy, as shown below. Figure 4 As shown, the curves of different colors correspond to Unit 1 respectively. Unit 10 can be used to represent different battery cells. The SoC starts from an initial value (range 0.62 to 0.85), which varies depending on the cell. The optimal SoC value is 0.824. The curves represent the SoC trajectory of different battery cells. Starting from the initial SoC value, the curves for different battery cells show a decreasing and increasing trend, rising further and then flattening out, highlighting the CC (Complex Cell) characteristic. Dynamic current distribution and overcharge prevention mechanisms during charging in CV mode. This figure demonstrates the optimization analysis of energy loss and battery performance after integrating a dynamic temperature model into the system.

[0077] Figure 5 This embodiment provides a schematic diagram of a process for adjusting the state of charge according to a DRL strategy, as shown below. Figure 5 As shown, in the initial stage (0 to 5000 seconds): the State of Charge (SoC) starts from an initial value (range 0.62 to 0.85), which varies depending on the individual cells. Under the deep reinforcement learning (DRL) control strategy, the SoC of the 10 battery cells changes over time (in seconds). The optimal SoC value is 0.655. Starting from a relatively high initial SoC value (approximately above 0.6), the curves corresponding to different battery cells show a decreasing and increasing trend, and after 20000 seconds, they begin to simultaneously tend to increase. Figure 5 This demonstrates the intelligent optimization potential of feedforward neural networks in predicting optimal discharge SoCs, and can also be used to compare DRL strategies with traditional control strategies (such as CC, CV, and CC). The technical effects of CV in terms of multi-cell consistency, improved operating efficiency and adaptation to dynamic temperature changes.

[0078] As an optional implementation of this embodiment, the step of determining the input data based on the current adjustment result and the temperature change result includes:

[0079] 1) Calculate the state of charge, temperature and current of each battery cell based on the current adjustment results and temperature change results.

[0080] Specifically, after obtaining the current adjustment results and temperature change results, the state of charge, temperature, and current of each battery cell can be calculated based on the current adjustment results and temperature change results.

[0081] For example, the state of charge (SOC) update is performed as follows: The SOC change is calculated by integrating the current and time step, limited to between 0 and 1. For instance, the updated SOC can be calculated using the SOC of the previous time step, current, time step, and the battery's rated capacity. Voltage calculation: The open-circuit voltage is calculated based on the SOC, then the voltage drop caused by internal resistance is subtracted to obtain the terminal voltage. Energy loss: The Joule heat generated by internal resistance in each step is calculated and accumulated as the total energy loss. Temperature change: The updated temperature can be calculated based on the temperature of the previous time step and the calculated temperature change. SOC standard deviation: The standard deviation of the SOC for each cell is calculated to assess consistency. Data storage: The SOC, voltage, current, temperature, SOC standard deviation, and energy loss are recorded and stored in a structured data container for easy analysis.

[0082] 2) Determine the input data corresponding to the energy storage battery pack based on the state of charge, temperature and current of each battery cell.

[0083] Specifically, after obtaining the state of charge, temperature, and current corresponding to each battery cell, the input data corresponding to the energy storage battery pack can be determined based on the state of charge, temperature, and current corresponding to each individual battery cell. For example, the arithmetic mean of the state of charge of all battery cells is calculated as the average state of charge; the arithmetic mean of the temperature of all battery cells is calculated as the average temperature; the arithmetic mean of the absolute values ​​of the current of all battery cells is calculated as the average absolute value of the current; and the standard deviation of the state of charge of all battery cells is calculated as the standard deviation of the state of charge. The above four together constitute the input dataset of the neural network of this invention.

[0084] As an optional implementation of this embodiment, the method further includes:

[0085] 1) Obtain historical input data; the historical input data includes historical average state of charge, historical average temperature, historical average absolute current, historical state of charge standard deviation and corresponding historical state of charge labels; wherein, the historical state of charge is determined according to preset rules.

[0086] In this embodiment, the preset rule can be a pre-set rule for determining the state of charge. For example, the preset rule can be expressed as:

[0087]

[0088] in, It can represent a specific optimal charge discharge state. It can represent the average state of charge. It can represent the average temperature. It can represent the standard deviation of the average state of charge. After obtaining the historical average state of charge, historical average temperature, historical average absolute current value, and historical standard deviation of the state of charge, the historical state of charge label in each historical input data can be determined according to preset rules.

[0089] 2) Input the historical input data into the neural network model to determine the predicted state of charge.

[0090] Specifically, after obtaining historical input data, the historical input data can be input into the neural network model, and then the predicted value of the state of charge output by the neural network model can be calculated based on the weights in the neural network model.

[0091] 3) Calculate the loss value using a loss function based on the predicted state of charge and the corresponding historical state of charge labels.

[0092] Specifically, after obtaining the predicted state of charge (SOC) value from the neural network model, the deviation between the predicted value and the historical SOC label can be quantified using a loss function, i.e., the loss value. The loss function can be the mean squared error loss function and / or the Huber loss function.

[0093] 4) Update the weights of the neural network model through backpropagation based on the loss value.

[0094] It is known that after obtaining the loss value, the weights of the neural network model can be updated through backpropagation. Specifically, the backpropagation algorithm, based on the chain rule, calculates the partial derivative of the loss function with respect to each weight layer by layer from the output layer to the input layer, obtaining the gradient tensor. The optimizer performs bias correction and momentum smoothing on the gradient based on the selected algorithm (Adam, RMSprop, etc.) to generate an equivalent update. By learning rate The weights are updated using the first moment to complete one parameter iteration.

[0095] 5) Iteratively execute the steps of calculating the model loss value and updating the weights until the difference between the predicted state of charge and the historical state of charge label meets the preset convergence condition.

[0096] It is understood that by repeatedly executing the steps of calculating the loss value and updating the weights of the model, the difference between the predicted state of charge and the historical state of charge label is satisfied with a preset convergence condition. The preset convergence condition may be that the iteration reaches a target number of iterations and / or the loss value tends to stabilize or falls below a preset threshold.

[0097] 6) The neural network model that satisfies the convergence condition is determined as the target neural network model.

[0098] Specifically, the neural network model that satisfies the aforementioned convergence condition can be determined as the target neural network model. For example, 1000 sets of combined data are generated, with the input range being a state of charge (SOC) of 0.3 to 1, a temperature of 25 to 45 degrees Celsius, a current of 0 to 200 amperes, and a standard deviation of 0 to 0.1. The output is adjusted based on low SOC, high temperature, and low consistency to ensure a conservative discharge strategy. The model is trained for 20 epochs using the Adam optimizer and mean squared error loss function, and the trained neural network model is then determined as the target neural network model.

[0099] This embodiment also provides an application process for a management method of an energy storage battery pack, with a total simulation time of 10 hours (36,000 seconds), a time step of 0.1 seconds, and a total of 360,000 steps. Key parameters include the capacity Q of each battery cell. 1000Ah, internal resistance R 0.01Ω, ambient temperature The linear relationship between open-circuit voltage (OCV) and state of charge (SoC) , where a 1.0, b 3.2). The total current demand curve is set to discharge 200A (positive current) for the first 5 hours and charge for the next 5 hours. 200A (negative current) to simulate load variations under actual operating conditions. Initial SoC settings are... The initial temperature can be This is to reflect the inconsistency in the initial state of the battery pack.

[0100] The simulation process includes parameter initialization, control strategy selection, state updates, and data logging. At each time step, the total current determines whether to enter either discharge (greater than or equal to 0) or charging (less than 0) mode. In discharge mode, a PID controller (proportional coefficient Kp) is used. 50. Integral coefficient Ki 0.05, differential coefficient Kd 10) Dynamically adjust the current of each unit based on the SoC deviation (limited to...). (300 to 300A), ensuring SoC balance. In charging mode, four strategies are supported: constant current (CC); constant voltage (CV); constant current-constant voltage (CC). CV); Deep Reinforcement Learning (DRL).

[0101] Simultaneously, heat generation can be calculated based on temperature, and the temperature can be updated in real time. After the simulation, a feedforward neural network (4 nodes in the input layer: average SoC, temperature, absolute current, and SoC standard deviation; 64 and 32 nodes in the hidden layers, ReLU activation; 1 node in the output layer, Sigmoid activation) is used to predict the optimal discharge SoC. This network is trained using 1000 sets of combined data to optimize the mean square error (MSE) loss. The simulation results are analyzed by recording SoC, voltage, current, temperature, SoC standard deviation, and energy loss. Table 1 summarizes the optimal discharge SoC (predicted by the neural network) and total energy loss (cumulative Joule heat) under different charging strategies provided in this embodiment, verifying the performance differences of various adjustment strategies:

[0102] Table 1. Optimal discharge and energy loss for different charging strategies

[0103]

[0104] As shown in Table 1, the DRL strategy achieves lower total energy loss and a lower predicted optimal discharge SoC, indicating its advantages in optimizing energy efficiency and deep discharge; the CV strategy has the lowest energy loss but a higher SoC, making it suitable for scenarios emphasizing safety; CC and CC The CV strategy has a high loss, but the SoC is moderate, making it suitable for fast charging requirements.

[0105] This technical solution is primarily applied to two major scenarios: energy storage systems and electric vehicles. It evaluates and optimizes BMS performance through a simulation framework. In energy storage systems, for grid-scale energy storage (such as large-scale photovoltaic or wind farms) or residential distributed energy storage systems, this solution can simulate and manage charging and discharging optimization under peak-valley electricity pricing, such as the cycle of charging during off-peak hours and discharging during peak hours. By considering temperature fluctuations and load uncertainties, PID control ensures SoC balance, a dynamic temperature model monitors heat accumulation to prevent overheating, and a neural network predicts the optimal SoC to guide operation. In electric vehicles, for electric vehicles (EVs) or hybrid electric vehicles (HEVs), this solution can simulate dynamic operating conditions such as high-speed driving, fast charging, or urban congestion, such as thermal management and SoC balance under high-power discharge. PID control dynamically adjusts the current to avoid unevenness, the temperature model is updated in real time to prevent thermal runaway, and the neural network predicts the optimal SoC based on vehicle data to improve range. This solution supports the integration of vehicle sensor data to achieve embedded BMS optimization, making it suitable for high-performance transportation applications. In the above scenarios, the preferred solution can update and adjust the strategy selection in real time according to the specific working conditions. It can also expand parameters or add new strategies to further adapt to other scenarios such as drones or portable devices.

[0106] Existing battery management methods have the following limitations: First, they only support a single charging or discharging strategy, making it impossible to comprehensively model and compare multiple control strategies, and thus difficult to fully assess their impact on battery performance. Second, they do not fully consider the impact of temperature changes on battery performance and lifespan. Third, they do not utilize machine learning techniques (such as neural networks) to predict optimal operating states (such as discharge SoC). The above-mentioned technical solution provides an efficient and flexible dynamic simulation and performance analysis method for battery management systems (BMS) through multi-strategy collaborative control, PID regulation, dynamic temperature models, and feedforward neural network optimization. This overcomes the limitations of existing technologies, significantly improving simulation accuracy, optimizing battery performance, and reducing energy loss. Specifically, it includes:

[0107] 1) Introducing a PID controller dynamically adjusts the current based on SoC deviation, optimizing SoC consistency among multiple battery cells. This can reduce SoC standard deviation, improve balance, and reduce energy loss (by minimizing Joule heating caused by current unevenness), thus improving battery life under high-load discharge scenarios.

[0108] 2) Integrates four strategies: constant current (CC), constant voltage (CV), constant current-constant voltage (CC-CV), and deep reinforcement learning (DRL), to achieve diversified charge and discharge management under a unified framework, and supports direct comparison of the impact of different strategies on battery performance.

[0109] Example 2

[0110] Figure 6 This is a schematic diagram of the structure of a management device for an energy storage battery pack provided in Embodiment 2 of this disclosure; as shown Figure 6 As shown, the device includes: an adjustment strategy determination module 210, a current adjustment result determination module 220, a temperature change result determination module 230, an input data determination module 240, and a management module 250.

[0111] The adjustment strategy determination module 210 is used to determine the corresponding adjustment strategy for each battery cell based on the operating mode of each battery cell; the adjustment strategy is a strategy for adjusting voltage and / or current.

[0112] The current adjustment result determination module 220 is used to dynamically adjust the current of the corresponding battery cell according to each adjustment strategy in order to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell.

[0113] Temperature change result determination module 230 is used to determine temperature change result based on preset operating parameters, ambient temperature and current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell.

[0114] The input data determination module 240 is used to determine the input data corresponding to the energy storage battery pack based on the current adjustment result and the temperature change result; the input data includes average state of charge, average temperature, absolute value of average current and standard deviation of state of charge.

[0115] The management module 250 is used to determine the target state of charge based on the input data using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

[0116] Embodiment 2 of this disclosure provides a management device for an energy storage battery pack, which realizes refined and intelligent management of the charging and discharging of each unit in the energy storage battery pack.

[0117] Furthermore, the device also includes:

[0118] The current information acquisition module is used to acquire the current information corresponding to each battery cell.

[0119] The operating mode determination module is used to determine the operating mode corresponding to each battery cell based on the current information; the operating mode includes charging mode and discharging mode.

[0120] Furthermore, the adjustment strategies include constant current adjustment strategy, constant voltage adjustment strategy, constant current-constant voltage adjustment strategy, deep learning reinforcement strategy, and proportional-integral-derivative PID controller.

[0121] Furthermore, the adjustment strategy determination module 210 is also used for:

[0122] For any battery cell, if the operating mode of the battery cell is the discharge mode, then the proportional-integral-derivative PID controller is determined as the adjustment strategy corresponding to the battery cell.

[0123] If the battery cell operates in the charging mode, then based on the operating state of the battery cell, the constant current adjustment strategy, the constant voltage adjustment strategy, the constant current-constant voltage adjustment strategy, and / or the deep learning reinforcement strategy are determined as the adjustment strategies corresponding to the battery cell.

[0124] Furthermore, the input data determination module 240 is also used for:

[0125] Calculate the state of charge, temperature, and current of each battery cell based on the current adjustment results and temperature change results;

[0126] The input data for the energy storage battery pack is determined based on the state of charge, temperature, and current of each battery cell.

[0127] Furthermore, the device also includes:

[0128] The historical input data acquisition module is used to acquire historical input data; the historical input data includes historical average state of charge, historical average temperature, historical average absolute current value, historical state of charge standard deviation, and corresponding historical state of charge labels; wherein, the historical state of charge is determined according to preset rules.

[0129] The prediction value determination module is used to input the historical input data into the neural network model to determine the predicted value of the state of charge.

[0130] The loss value determination module is used to calculate the loss value using a loss function based on the predicted state of charge and the corresponding historical state of charge label.

[0131] An update module is used to update the weights of the neural network model through backpropagation based on the loss value;

[0132] The target neural network model determination module iteratively executes the steps of calculating the model loss value and updating the weights until the difference between the predicted state of charge and the historical state of charge label meets the preset convergence condition; the neural network model that meets the convergence condition is determined as the target neural network model.

[0133] Furthermore, the target neural network model includes input nodes, hidden layers, and output nodes.

[0134] The energy storage battery pack management device provided in this disclosure can execute the energy storage battery pack management method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of executing the method.

[0135] Example 3

[0136] Figure 7 A schematic diagram of the structure of an electronic device 10 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments of the present disclosure described and / or claimed herein.

[0137] like Figure 7As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0138] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0139] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microprocessor, etc. Processor 11 performs the various methods and processes described above, such as the management methods for energy storage battery packs.

[0140] In some embodiments, the energy storage battery pack management method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the energy storage battery pack management method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the energy storage battery pack management method by any other suitable means (e.g., by means of firmware).

[0141] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] Computer programs for implementing the methods of embodiments of this disclosure may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0143] In the context of embodiments of this disclosure, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0145] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0146] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0147] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the embodiments of this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of the embodiments of this disclosure can be achieved, and this document does not impose any limitations.

[0148] The specific embodiments described above do not constitute a limitation on the scope of protection of the embodiments disclosed herein. Those skilled in the art should understand that various modifications, combinations, sub-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 the embodiments disclosed herein should be included within the scope of protection of the embodiments disclosed herein.

[0149] This disclosure also provides a computer program product, including a computer program and / or instructions, which, when executed by a processor, implements the energy storage battery pack management method provided in any embodiment of this application.

[0150] In implementing a computer program product, computer program code for performing the operations of the embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0151] Note that the above are merely preferred embodiments and the technical principles applied in this disclosure. Those skilled in the art will understand that this disclosure is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the protection scope of this disclosure. Therefore, although the embodiments of this disclosure have been described in detail above, this disclosure is not limited to the above embodiments. More other equivalent embodiments may be included without departing from the concept of this disclosure, and the scope of this disclosure is determined by the scope of the appended claims.

Claims

1. A management method for an energy storage battery pack, characterized in that, The energy storage battery pack includes at least one battery cell, and the method includes: The adjustment strategy for each battery cell is determined according to its corresponding operating mode; the adjustment strategy is a strategy for adjusting voltage and / or current. According to each adjustment strategy, the current of the corresponding battery cell is dynamically adjusted to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell. The temperature change result is determined based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell. The input data corresponding to the energy storage battery pack is determined based on the current adjustment results and the temperature change results; the input data includes the average state of charge, average temperature, absolute value of average current, and standard deviation of state of charge. Based on the input data, a target state of charge is determined using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

2. The method according to claim 1, characterized in that, The method further includes: Obtain the current information corresponding to each battery cell; Based on the current information, the operating mode corresponding to each battery cell is determined; the operating mode includes charging mode and discharging mode.

3. The method according to claim 2, characterized in that, The adjustment strategies include constant current adjustment strategy, constant voltage adjustment strategy, constant current-constant voltage adjustment strategy, deep learning reinforcement strategy, and proportional-integral-derivative PID controller. The step of determining the corresponding adjustment strategy for each battery cell based on its operating mode includes: For any battery cell, if the operating mode of the battery cell is the discharge mode, then the proportional-integral-derivative PID controller is determined as the adjustment strategy corresponding to the battery cell. If the battery cell operates in the charging mode, then based on the operating state of the battery cell, the constant current adjustment strategy, the constant voltage adjustment strategy, the constant current-constant voltage adjustment strategy, and / or the deep learning reinforcement strategy are determined as the adjustment strategies corresponding to the battery cell.

4. The method according to claim 1, characterized in that, The step of determining the input data based on the current adjustment result and the temperature change result includes: Calculate the state of charge, temperature, and current of each battery cell based on the current adjustment results and temperature change results. The input data for the energy storage battery pack is determined based on the state of charge, temperature, and current of each battery cell.

5. The method according to claim 1, characterized in that, The method further includes: Acquire historical input data; the historical input data includes historical average state of charge, historical average temperature, historical average absolute current value, historical state of charge standard deviation, and corresponding historical state of charge labels; wherein, the historical state of charge is determined according to preset rules; The historical input data is fed into a neural network model to determine the predicted state of charge. Based on the predicted state of charge and the corresponding historical state of charge labels, the loss value is calculated using a loss function. Based on the loss value, the weights of the neural network model are updated through backpropagation; The steps of calculating the model loss value and updating the weights are executed iteratively until the difference between the predicted state of charge and the historical state of charge label meets the preset convergence condition; the neural network model that meets the convergence condition is determined as the target neural network model.

6. The method according to claim 5, characterized in that, The target neural network model includes input nodes, hidden layers, and output nodes.

7. A management device for an energy storage battery pack, characterized in that, The energy storage battery pack includes at least one battery cell, and the device includes: The adjustment strategy determination module is used to determine the corresponding adjustment strategy for each battery cell based on the operating mode of each battery cell; the adjustment strategy is a strategy for adjusting voltage and / or current. The current adjustment result determination module is used to dynamically adjust the current of the corresponding battery cell according to each adjustment strategy in order to determine the current adjustment result; the current adjustment result includes the adjusted current of each battery cell. The temperature change result determination module is used to determine the temperature change result based on preset operating parameters, ambient temperature, and the current adjustment result; the temperature change result includes temperature change data corresponding to each battery cell. The input data determination module is used to determine the input data corresponding to the energy storage battery pack based on the current adjustment result and the temperature change result; the input data includes average state of charge, average temperature, absolute value of average current and standard deviation of state of charge. The management module is used to determine the target state of charge based on the input data using a target neural network model, so as to dynamically manage the charging and discharging process of each battery cell in the energy storage battery pack according to the target state of charge.

8. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the energy storage battery pack management method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the management method for the energy storage battery pack as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the management method for an energy storage battery pack as described in any one of claims 1-6.

Images