Method and system for determining state of charge of single battery cell, electronic equipment and medium
By calculating the ohmic internal resistance and polarization internal resistance of the battery cell and using deep learning models, the problem of difficult to accurately determine the charge state of a single battery cell is solved, the stability and service life of the energy storage system are improved, and maintenance costs are reduced .
Patent Information
- Application Number
- CN202411440370.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art is difficult to accurately determine the state of charge of a single battery cell, resulting in high maintenance costs of energy storage systems and poor system stability. Traditional methods cannot reflect the impact of battery cell aging on the state of charge.
By collecting the operating voltage of the battery cell and the operating current of the battery pack, calculating the ohmic internal resistance and polarization internal resistance, and using deep learning models to combine these parameters to estimate the state of charge of the battery, considering the impact of battery aging on the state of charge. .
It improves the accuracy and reliability of the charge state of a single cell, extends the service life of the battery system, reduces maintenance costs, and ensures the stability and reliability of the energy storage system.
Smart Images

Figure CN119087281A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of energy storage, and in particular to a method, system, electronic equipment and medium for determining the state of charge of a single cell. Background Art
[0002] In the field of large-scale energy storage, energy storage systems are usually composed of multiple battery packs connected in series or in parallel, and each battery pack is composed of multiple cells connected in series or in parallel. The performance degradation of a battery pack will affect the entire energy storage system, and the performance degradation of a single cell in the battery pack will affect the overall performance of the battery pack. Therefore, how to determine the SOC (State of Charge) of a single cell and a battery pack is crucial to the stable operation and subsequent maintenance of the energy storage system.
[0003] There are currently two methods in the prior art to calculate the SOC of a battery pack. The first method is the ampere-hour integration method, which calculates the percentage of the change in power by calculating the integral of the charge and discharge current and the corresponding time within a certain period of time to obtain the SOC of the battery pack. This process requires the collection of the working current within the corresponding time. Since the battery cells are combined into a battery pack for application in actual applications, only the working current of the entire battery pack can be collected, rather than the working current of each single cell. Therefore, the SOC calculated by this method cannot reflect the true state of the single cell. When a large energy storage system needs to be repaired due to cell damage, the entire battery pack needs to be replaced, resulting in high maintenance costs.
[0004] The second method is the open circuit voltage method, which uses the relatively fixed functional relationship between the open circuit voltage and SOC when the battery is stationary for a long time. By collecting the voltage across the single cell and directly looking up the table using the voltage and capacity relationship chart, the SOC is estimated based on the open circuit voltage. Although the final method can determine the state of charge of the single cell, in actual working conditions, the battery is generally not stationary for a long time, and the voltage across the cell will change with the working conditions. The slight voltage change in the cell may cause a rapid change in SOC; and as the number of cell cycles increases, chemical crystals will be generated inside the cell, and its capacity will decay. Therefore, the SOC obtained by this method has a large deviation and poor accuracy. Summary of the invention
[0005] The purpose of the present invention is to provide a method, system, electronic device and medium for determining the state of charge of a single battery cell. By determining the ohmic internal resistance and polarization internal resistance of the battery cell that can reflect the aging state of the battery, the influence of battery aging on the state of charge of the battery cell is further considered to ensure the accuracy and reliability of the final state of charge result; by accurately estimating the state of charge of the single battery cell, the service life of the battery system is extended, the maintenance cost is reduced, and the stability and reliability of the entire energy storage system are ensured.
[0006] In order to solve the above technical problems, the present invention provides a method for determining the state of charge of a single cell, comprising:
[0007] Determine operating parameters of the target battery cell, wherein the operating parameters of the target battery cell include an operating voltage of the target battery cell and an operating current of a battery pack in which the target battery cell is located;
[0008] Determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0009] The operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell are input into a pre-established deep learning model to obtain an estimation result of the state of charge of the target battery cell.
[0010] Optionally, the operating parameter of the target battery cell further includes an average voltage of a battery pack where the target battery cell is located; and the method for determining the state of charge of a single battery cell further includes:
[0011] Collecting the operating voltage of the battery pack where the target battery cell is located;
[0012] The average voltage of the battery pack where the target battery cell is located is determined based on the operating voltage of the battery pack where the target battery cell is located and the number of battery cells in the battery pack.
[0013] Optionally, determining the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located includes:
[0014] Constructing a second-order RC equivalent circuit model of the target battery cell, and performing a charge and discharge test on the target battery cell;
[0015] Determine in real time the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test;
[0016] The ohmic internal resistance and polarization internal resistance of the target battery cell are calculated according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test.
[0017] Optionally, the step of calculating the ohmic internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test includes:
[0018] Determine the minimum voltage and maximum voltage of the target battery cell during the charge and discharge test;
[0019] determining a difference between the maximum voltage and the minimum voltage;
[0020] The ratio of the difference to the operating current of the battery pack where the target battery cell is located is determined as the ohmic internal resistance of the target battery cell.
[0021] Optionally, calculating the polarization internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test includes:
[0022] Constructing a state response function of the operating voltage of the target battery cell;
[0023] The operating voltage of the target battery cell and the operating current of the battery pack in which the corresponding target battery cell is located are substituted into the state response function of the operating voltage of the target battery cell to calculate the polarization internal resistance of the target battery cell.
[0024] Optionally, also include:
[0025] Build a deep learning model with the operating parameters, ohmic internal resistance and polarization internal resistance of the single cell as input and the state of charge of the single cell as output;
[0026] The operating parameters, ohmic internal resistance and polarization internal resistance of a single cell with a known state of charge are used as a training set to train the deep learning model.
[0027] Optionally, the deep learning model includes an input layer, an intermediate layer and an output layer, the intermediate layer includes a multi-layer network, and the number of neurons in each layer of the network is less than a preset value.
[0028] In order to solve the above technical problems, the present invention also provides a system for determining the state of charge of a single cell, comprising:
[0029] An operating parameter determination unit, used to determine the operating parameters of the target battery cell, wherein the operating parameters of the target battery cell include the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0030] An internal resistance calculation unit, used to determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0031] An estimation unit is used to input the operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell into a pre-established deep learning model to obtain an estimation result of the state of charge of the target battery cell.
[0032] In order to solve the above technical problems, the present invention further provides an electronic device, comprising:
[0033] Memory for storing computer programs;
[0034] A processor is used to implement the steps of the method for determining the state of charge of a single cell as described above.
[0035] To solve the above technical problems, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for determining the charge state of a single battery cell as described above are implemented.
[0036] The present invention provides a method for determining the state of charge of a single cell. A deep learning model for determining the state of charge of a single cell according to the working parameters, ohmic internal resistance and polarization internal resistance of the single cell is constructed in advance to estimate the state of charge of the cell. When applied, the working parameters, ohmic internal resistance and polarization internal resistance of the single cell are input to directly obtain the estimated result of the state of charge of the cell, which can effectively determine the state of charge of the single cell in the battery pack. At the same time, when determining the state of charge of the cell, not only the influence of the working parameters of the single cell on the state of charge of the cell is considered, but also the influence of battery aging on the state of charge of the cell is further considered by determining the ohmic internal resistance and polarization internal resistance of the cell that can reflect the aging state of the battery, thereby ensuring the accuracy and reliability of the final state of charge result. By accurately estimating the state of charge of the single cell, the service life of the battery system is extended, the maintenance cost is reduced, and the stability and reliability of the entire energy storage system are ensured.
[0037] The present invention also provides a system for determining the state of charge of a single cell, an electronic device, and a computer-readable storage medium, which have the same beneficial effects as the method for determining the state of charge of a single cell mentioned above. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the prior art and the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0039] Figure 1 A schematic flow chart of a method for determining the state of charge of a single cell provided by the present invention;
[0040] Figure 2 A schematic flow chart of another method for determining the state of charge of a single cell provided by the present invention;
[0041] Figure 3 A schematic diagram of the structure of an equivalent circuit of a single cell provided by the present invention;
[0042] Figure 4A schematic diagram of a waveform variation of the working voltage of a single cell provided by the present invention;
[0043] Figure 5 A schematic diagram of an enlarged portion of the waveform of the working voltage of a single battery cell provided by the present invention;
[0044] Figure 6 A schematic diagram of the structure of a deep learning model provided by the present invention;
[0045] Figure 7 A schematic diagram of the structure of an intermediate layer of a deep learning model provided by the present invention;
[0046] Figure 8 A schematic diagram for comparing the estimated results of the state of charge of a single cell provided by the present invention;
[0047] Fig. 9 A schematic diagram of the structure of a system for determining the state of charge of a single cell provided by the present invention;
[0048] Fig.10 A schematic structural diagram of an electronic device with a single battery cell provided by the present invention. DETAILED DESCRIPTION
[0049] The core of the present invention is to provide a method, system, electronic device and medium for determining the state of charge of a single battery cell. By determining the ohmic internal resistance and polarization internal resistance of the battery cell that can reflect the aging state of the battery, the influence of battery aging on the state of charge of the battery cell is further considered to ensure the accuracy and reliability of the final state of charge result; by accurately estimating the state of charge of the single battery cell, the service life of the battery system is extended, the maintenance cost is reduced, and the stability and reliability of the entire energy storage system are ensured.
[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] Please refer to Figure 1 , Figure 1 A schematic diagram of a method for determining the state of charge of a single cell provided by the present invention; please refer to Figure 2 , Figure 2 A schematic flow chart of another method for determining the state of charge of a single cell provided by the present invention; To solve the above technical problem, the present invention provides a method for determining the state of charge of a single cell, comprising:
[0052] S11: Determine operating parameters of the target battery cell, where the operating parameters of the target battery cell include an operating voltage of the target battery cell and an operating current of a battery pack where the target battery cell is located;
[0053] It is not difficult to understand that the state of charge of a battery cell is related to its working parameters such as voltage and current during operation. The working parameters such as the working voltage of the target battery cell and the working current of the battery pack where the target battery cell is located can accurately reflect the power consumption of the target battery cell and the battery pack where it is located. Therefore, it is necessary to first collect and determine the working parameters of the target battery cell in real time. The data of these working parameters can not only be used to monitor the basic state of the battery, but also provide a basis for subsequent calculations. Therefore, the accuracy and real-time nature of the collection of these working parameters must be ensured. Specifically, the BMS (BATTERY MANAGEMENT SYSTEM) system of the battery pack itself can be used to collect the voltage data of the single battery cell and the current data of the battery pack where it is located. The BMS system can collect the voltage data of the single battery cell and the current data of the battery pack where it is located in real time. The working parameters of the target battery cell can also be determined by other means. This application is not particularly limited here, and is not limited to the direct collection method of the BMS system proposed in this embodiment.
[0054] S12: Determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0055] It is understandable that after collecting these working parameters, the ohmic internal resistance and polarization internal resistance of the target battery cell can be calculated. The ohmic internal resistance represents the resistance of the material inside the battery cell itself, while the polarization internal resistance reflects the additional resistance generated by the battery cell during the electrochemical reaction. Therefore, the ohmic internal resistance and polarization internal resistance of the target battery cell determined can accurately characterize whether chemical crystallization has occurred inside the battery cell, effectively identify whether the target battery cell has capacity decay due to factors such as aging, and ensure the accuracy of the estimated results of the state of charge of the single battery cell under conditions such as battery aging. There are many options for calculating the ohmic internal resistance and polarization internal resistance of the battery cell. The ohmic internal resistance and polarization internal resistance can be calculated by constructing a second-order equivalent circuit model of the target battery cell, and based on the voltage data of the single battery cell detected in real time and the current data of the battery pack in which it is located. This application is not particularly limited here.
[0056] S13: Inputting the operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell into a pre-established deep learning model to obtain an estimated result of the state of charge of the target battery cell.
[0057] It should be noted that the present application has constructed in advance a deep learning model with the working parameters, ohmic internal resistance and polarization internal resistance of a single cell as input and the state of charge of a single cell as output. The deep learning model can output the estimated result of the state of charge of the cell corresponding to the current input data when there are data inputs of the working parameters of the target cell, the ohmic internal resistance of the target cell and the polarization internal resistance of the target cell. The deep learning model has been trained in advance and can accurately reflect the SOC changes of single cells under different conditions. Therefore, the operating parameters such as the operating voltage of the single cell of the target cell, the calculated ohmic internal resistance, polarization internal resistance, and the operating current of the battery pack where the target cell is located can be input into the deep learning model. After receiving the above input data, the deep learning model processes and calculates through the neural network structure, and finally obtains the SOC of the single cell. Finally, the deep learning model outputs the SOC value of the single cell.
[0058] It is not difficult to understand that the SOC value output by the pre-built and extensively trained deep learning model is more accurate than the result calculated by the traditional method, and can better reflect the true state of the single battery cell. By introducing the deep learning model, not only the limitations of the traditional method are overcome, but also key parameters such as the ohmic internal resistance and polarization internal resistance of the battery cell are utilized to significantly improve the accuracy of the SOC calculation. Even in the case of battery aging or changes in usage conditions, it can still provide accurate SOC estimation, thus providing a solid technical guarantee for the efficient operation and maintenance of large-scale energy storage systems.
[0059] It should be noted that the method for determining the state of charge of a single cell provided by the present invention can be applied to the BMS system of an energy storage system, and can be used to determine the state of charge of each single cell in each battery pack of the energy storage system, thereby effectively determining the fault condition of the single cell, and is suitable for estimating the SOC of lithium iron phosphate single cells in the field of large-scale energy storage, reducing the maintenance cost of replacing battery packs due to damage to single cells, and improving the stable operation time of large-scale energy storage systems. A processing chip or the like can also be separately set in the energy storage system to complete the entire process of determining the state of charge of the single cell, and this application does not make any special restrictions here. The specific structure and construction method of the deep learning model can also be selected and adjusted according to actual conditions, and this application does not make any special restrictions here. It can be applied to energy storage systems of the type of lithium iron phosphate batteries, and has a wide range of application scenarios. Accordingly, the target cell can be a cell of the type of lithium iron phosphate battery, and this application does not make any special restrictions here.
[0060] The present invention proposes a method for estimating the SOC of lithium iron phosphate single cells in the field of large-scale energy storage. First, it can reduce maintenance costs and effectively reduce the cost of replacing and maintaining battery packs due to damage to single cells. By accurately estimating the SOC of single cells, the service life of the battery system is extended and the maintenance cost is reduced. Secondly, it can improve the stability of the entire energy storage system, and optimize the operation and management of large-scale energy storage systems through accurate SOC estimation. The system can supply power and receive charges more stably, improving the overall stability and reliability of the system. Specifically, the SOC can be estimated through a deep learning model of a 5-layer network, which solves the problem that the BMS system cannot calculate the SOC of single cells and the problem that the application of deep learning methods is limited in the field of large-scale energy storage. By establishing a second-order circuit model of the battery, the internal resistance of the single cell can be calculated, and the internal resistance parameters can be involved in the calculation, which solves the problem that deep learning and neural network methods are subject to the amount of battery aging data and inaccurate estimation.
[0061] The present invention provides a method for determining the state of charge of a single cell. A deep learning model for determining the state of charge of a single cell according to the working parameters, ohmic internal resistance and polarization internal resistance of the single cell is constructed in advance to estimate the state of charge of the cell. When applied, the working parameters, ohmic internal resistance and polarization internal resistance of the single cell are input to directly obtain the estimated result of the state of charge of the cell, which can effectively determine the state of charge of the single cell in the battery pack. At the same time, when determining the state of charge of the cell, not only the influence of the working parameters of the single cell on the state of charge of the cell is considered, but also the influence of battery aging on the state of charge of the cell is further considered by determining the ohmic internal resistance and polarization internal resistance of the cell that can reflect the aging state of the battery, thereby ensuring the accuracy and reliability of the final state of charge result. By accurately estimating the state of charge of the single cell, the service life of the battery system is extended, the maintenance cost is reduced, and the stability and reliability of the entire energy storage system are ensured.
[0062] Based on the above embodiments:
[0063] As an optional embodiment, the operating parameter of the target battery cell further includes an average voltage of the battery pack where the target battery cell is located; and the method for determining the state of charge of the single battery cell further includes:
[0064] Collect the operating voltage of the battery pack where the target cell is located;
[0065] The average voltage of the battery pack where the target battery cell is located is determined based on the operating voltage of the battery pack where the target battery cell is located and the number of batteries in the battery pack.
[0066] It is not difficult to understand that in order to more accurately determine the voltage of a single cell, the average voltage of the battery pack where the target cell is located can be further calculated to reflect the situation of the single cell in the battery pack. At the same time, in the process of estimating the SOC of a single cell using a deep learning model, the average voltage of the battery pack where the target cell is located can also be input into the deep learning model to increase the input features of the deep learning model, thereby improving the accuracy and reliability of the state of charge of the final single cell. The present application does not make any special restrictions on the specific method for determining the average voltage, which is not limited to being determined based on the entire working voltage of the battery pack and the number of cells set, but can also be determined based on the specific connection method of each single cell in the battery pack during actual application. For example, when all the single cells in the battery pack are in parallel, the working voltage of the battery pack can be directly used as the average voltage; when all the single cells in the battery pack are in series, the specific voltage division relationship between them needs to be considered.
[0067] It should be further explained that in addition to the operating voltage of the target battery cell, the operating current of the battery pack where the target battery cell is located, and the average voltage of the battery pack where the target battery cell is located, the present application also takes into account that temperature has a significant impact on the performance of the battery, so the temperature data of the single battery cell is also crucial for the accurate calculation of the SOC, so the operating parameters of the target battery cell may also include the operating temperature of the target battery cell. In the process of estimating the SOC of a single battery cell using a deep learning model, it is also necessary to input the operating temperature of this target battery cell into the deep learning model to increase the input features of the deep learning model, thereby improving the accuracy and reliability of the charge state of the final single battery cell. The present application does not make any special restrictions on the specific method for determining the operating temperature of the target battery cell, which can be achieved by adding a temperature sensor and the like.
[0068] Specifically, we can further consider the influence of the target cell's operating parameters such as the average voltage of the battery pack where the target cell is located and the operating temperature of the target cell on its state of charge and battery performance, increase the input features of the deep learning model, and thus effectively improve the accuracy and reliability of the final state of charge of the single cell.
[0069] Please refer to Figure 3 , Figure 3 This is a structural schematic diagram of an equivalent circuit of a single cell provided by the present invention; please refer to Figure 4 , Figure 4 This is a schematic diagram of the waveform change of the working voltage of a single battery cell provided by the present invention; please refer to Figure 5 , Figure 5A schematic diagram of a partial waveform enlargement of the working voltage of a single cell provided by the present invention; as an optional embodiment, the ohmic internal resistance and polarization internal resistance of the target cell are determined based on the working voltage of the target cell and the working current of the battery pack where the target cell is located, including:
[0070] Construct a second-order RC equivalent circuit model of the target battery cell and perform charge and discharge tests on the target battery cell;
[0071] Determine in real time the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test;
[0072] The ohmic internal resistance and polarization internal resistance of the target cell are calculated according to the operating voltage of the target cell and the operating current of the battery pack in which the target cell is located during the charge and discharge test.
[0073] It is not difficult to understand that for the BMS system, since it cannot collect the current data of the single cell, the ampere-hour integration method and the open circuit voltage method cannot be used to calculate the SOC of the single cell. Although the Kalman filter method can estimate the SOC, it is essentially based on the correction of the ampere-hour integration method, so it is also not suitable for the SOC estimation of the single cell. To solve this problem, the present invention proposes a SOC estimation method based on a deep learning model, which gets rid of the dependence of the traditional SOC calculation on the current data of the single cell, and estimates the SOC of the single cell by constructing and training a deep learning model in advance. The advantage of the deep learning model lies in its powerful data processing and pattern recognition capabilities, but its disadvantage is that it cannot fully consider the aging factors of the battery. As the number of battery cycles increases, the internal resistance of the cell increases with the number of cycles, and the SOC of the battery will gradually decrease. Therefore, the present invention proposes to model the lithium iron phosphate battery cell, calculate the ohmic internal resistance and polarization internal resistance of the single cell by establishing a second-order RC equivalent circuit model of the single cell and performing charge and discharge tests on the single cell, thereby improving the deep learning SOC estimation model, so that the final estimated SOC result can take into account the battery aging factor.
[0074] It is understandable that when modeling lithium iron phosphate batteries, considering that the BMS systems in the field of large-scale energy storage mostly use STM32 as processors, their computing performance is relatively weak. Therefore, the present invention uses a second-order equivalent circuit model to model single cells. The first-order circuit model is not enough to show the characteristics of lithium iron phosphate batteries, and the third-order and above models have too high hardware requirements for the BMS system and are difficult to deploy in practice. The second-order model strikes a balance between complexity and computing performance, can accurately reflect battery characteristics, and adapt to existing hardware conditions.
[0075] Specifically, if Figure 3As shown, the second-order equivalent circuit model includes two capacitors and two resistors. Capacitor C1 represents the electrochemical polarization capacitance of the battery cell, capacitor C2 represents the concentration difference polarization capacitance of the battery cell, and resistor R C1 Represents the electrochemical polarization resistance of the battery cell, resistance R C2 represents the concentration difference polarization resistance of the battery cell, the resistance R0 represents the ohmic internal resistance of the battery cell, I represents the operating current of the battery pack where the target battery cell is located, and U OC Represents the open circuit voltage of the battery. Then, by real-time monitoring of the changes in operating parameters such as voltage and temperature of the target battery during the charge and discharge test, combined with a deep learning model, the battery SOC can be accurately estimated. This model not only takes into account the real-time status of the battery, but also incorporates the impact of battery aging on internal resistance, making SOC estimation more accurate and reliable.
[0076] Specifically, by establishing a second-order battery model of a single cell and combining it with the charge and discharge test process, the internal resistance of the single cell can be effectively calculated. The internal resistance parameters are then used in the deep learning model's SOC calculation process, effectively solving the problem of the impact of the amount of battery aging data on the SOC estimation accuracy.
[0077] As an optional embodiment, calculating the ohmic internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test includes:
[0078] Determine the minimum and maximum voltages of the target cell during the charge and discharge test;
[0079] determining the difference between the maximum voltage and the minimum voltage;
[0080] The ratio of the difference to the operating current of the battery pack where the target battery cell is located is determined as the ohmic internal resistance of the target battery cell.
[0081] It is not difficult to understand that when the target battery cell is charged and discharged, its voltage changes as follows Figure 4 As shown in the figure, the ordinate is voltage and the abscissa is time. At the beginning of the first discharge cycle, the terminal voltage of the target cell suddenly drops sharply, and then gradually decreases. At the moment of discharge termination, its terminal voltage suddenly rises sharply, and then gradually rises back. This phenomenon is called the voltage rebound characteristic of the cell. The change in the working voltage of the target cell during the voltage rebound stage is magnified as shown in the figure below. Figure 5 As shown. This voltage rebound phenomenon is caused by the equivalent resistance of the battery cell. Specifically, the ohmic internal resistance R0 directly affects the sudden rise and fall of its terminal voltage. When the current changes suddenly, the ohmic internal resistance will cause the voltage to change sharply in an instant. For example, at the beginning of discharge, due to the sudden increase in current, the ohmic internal resistance will cause the voltage to drop rapidly; at the end of discharge, the current suddenly decreases, and the ohmic internal resistance will cause the voltage to rise rapidly.
[0082] In order to quantitatively analyze this phenomenon, the minimum voltage at the end of discharge can be recorded as the minimum voltage U min , the voltage at the end of the sudden rise is recorded as the maximum voltage U max .but By measuring these two values, we can better understand the impact of ohmic internal resistance on voltage changes, and further use the operating current of the battery pack where the target cell is located to calculate the ohmic internal resistance of the target cell, thereby further optimizing the SOC estimation model of the cell.
[0083] Specifically, since the voltage rebound characteristics of the target battery cell during the charge and discharge test are caused by the ohmic internal resistance, by determining the minimum voltage and maximum voltage of the target battery cell during the charge and discharge test, the ohmic internal resistance can be effectively calculated based on its voltage rebound characteristics, so as to facilitate the subsequent optimization of the deep learning model.
[0084] As an optional embodiment, the polarization internal resistance of the target battery cell is calculated according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test, including:
[0085] Construct a state response function for the operating voltage of the target battery cell;
[0086] The operating voltage of the target battery cell and the operating current of the battery pack in which the corresponding target battery cell is located are substituted into the state response function of the operating voltage of the target battery cell to calculate the polarization internal resistance of the target battery cell.
[0087] It is not difficult to understand that based on the constructed second-order equivalent circuit model of the single cell and the voltage change of the target cell, the state response function of the operating voltage of the target cell can be generated. min The voltage response to the steady state is a zero-state response process, so its terminal voltage is as shown in formula (1):
[0088] ; (1)
[0089] Among them, U is the terminal voltage of the single cell, that is, the working voltage of the single cell; U1 is the capacitor C1, that is, the resistor R C1 The voltage across the two ends, U2 is the capacitor C2, that is, the resistor R C2 The voltage at both ends, t is the moment. At the same time, the working voltage can be changed from U max The voltage change to the stable state of the battery cell is exponentially fitted to obtain formula (2):
[0090] ; (2)
[0091] Among them, y=U, A1 is the fitting value of U1, and A2 is the fitting value of U2. After a long period of static, the polarization phenomenon disappears, that is, at the moment of discharge end, the process in which the working voltage of the battery cell drops sharply to Umin is a zero-shaped response, and its terminal voltage is shown in formula (3):
[0092] ; (3)
[0093] And among them: ; ; ; ; . Among them, R1 is the resistance R C1 The resistance value of R2 resistor R C2 At the same time, during the charge and discharge test, the working voltage of the target battery cell and the working current of the battery pack where the target battery cell is located are also determined in real time. The real-time collected values of the working voltage of the target battery cell and the working current of the battery pack where the target battery cell is located are substituted into formula (3) to obtain the resistance R C1 and resistor R C2 The ohmic internal resistance of a single cell can also be calculated. Thus, the ohmic internal resistance and polarization internal resistance of a single cell can be calculated.
[0094] Specifically, the state response function of the voltage of a single cell can be generated based on the second-order equivalent circuit model of the single cell, and the ohmic internal resistance and polarization internal resistance of the target cell can be calculated accurately and effectively according to the working voltage of the target cell and the working current of the battery pack where the target cell is located determined by real-time acquisition.
[0095] Please refer to Figure 6 , Figure 6 A structural diagram of a deep learning model provided by the present invention; please refer to Figure 7 , Figure 7 A schematic diagram of the structure of an intermediate layer of a deep learning model provided by the present invention; as an optional embodiment, it also includes:
[0096] Build a deep learning model with the operating parameters, ohmic internal resistance and polarization internal resistance of the single cell as input and the state of charge of the single cell as output;
[0097] The operating parameters, ohmic internal resistance and polarization internal resistance of single cells with known charge states are used as training sets to train the deep learning model.
[0098] It is not difficult to understand that before application, it is necessary to build a deep learning model and use the training set to train the deep learning model to ensure the accuracy of the deep learning model. There are many options for obtaining training set data, and this application does not make any special restrictions here. The working parameters of the single battery cell with a known state of charge can be used as the training set, or the target battery cell can be directly used as the experimental object, and the circuit test under constant experimental conditions is performed on it, and the obtained experimental results are used as the training set data. In order to obtain the training set data, the present invention first builds a cycle test platform, writes the constructed second-order equivalent circuit model of the target battery cell into the BMS system, and saves the data collected by the BMU (Battery Monitor Unit) and the BCU (Battery Control Unit) through the BMS host computer, so as to produce a deep learning model training data set. BMU is mainly responsible for monitoring the basic parameters of the battery cell or battery module, such as voltage, current, temperature, etc. It has high-precision sensors and data acquisition circuits, and can obtain accurate battery status information in real time; BCU can execute the battery control strategy based on the data provided by BMU. It can control the charging and discharging process of the battery, including the regulation of charging current and voltage, setting of discharge cut-off voltage, etc. In order to accurately verify the accuracy of the present invention, the test work of this model of the present invention can be a real machine test.
[0099] It is understandable that for the BMS system, in addition to being unable to collect current data of a single cell, there is also the problem of limited processor storage and computing capacity. In order to reduce the development cost of the BMS system, the BMU on the market usually uses the STM32F103 processor, which does not have a floating-point operation acceleration function. Therefore, when designing the SOC estimation model, it is necessary to fully consider the computing and storage capabilities of the hardware. Most of the traditional neural network algorithms use a 3-to-1 or 5-to-1 design. In order to improve the accuracy of the model, these algorithms often contain a large number of neurons in the middle single layer, resulting in a large number of model parameters, which is difficult to apply to large storage BMS systems. Therefore, the present invention proposes an improved deep learning model that can efficiently and accurately estimate the SOC of a single cell under limited hardware conditions. In order to adapt to the computing and storage capabilities of the STM32F103 processor, the present invention uses a lightweight neural network to design a deep learning model. By reducing the number of neurons and simplifying the network structure, the computational complexity and storage requirements of the model are reduced.
[0100] Specifically, the present invention solves the problem that the neural network cannot consider the impact of electrochemical reactions on the state of charge of the battery cell by establishing a second-order RC model of the battery cell, using the ohmic internal resistance, polarization internal resistance, voltage, temperature, etc. of the battery cell as input variables, thereby solving the problem that the neural network estimates inaccurately over time. Although the deep learning model used is a lightweight network, the final estimated SOC still has high accuracy.
[0101] As an optional embodiment, the deep learning model includes an input layer, an intermediate layer and an output layer, the intermediate layer includes a multi-layer network, and the number of neurons in each layer of the network is less than a preset value.
[0102] It is not difficult to understand that in order to improve the accuracy, traditional neural network models usually add a large number of neurons in the middle layer. As for SOC calculation, it is mainly used to estimate the capacity of the battery, and its influencing factors mainly include working parameters such as voltage, current, and temperature. Therefore, the computational complexity of its model is low, and a small-scale network with a small number of neurons can be used to implement the middle layer, reduce the complexity of the model, and reduce the number of parameters. The neural network algorithms that can be deployed mostly adopt a three-layer structure, namely the input layer, the middle layer, and the output layer. The middle layer mostly adopts a 1-layer or 2-layer network design. In order to ensure the accuracy of the network model, the number of neurons in the middle layer is set very large, which is too bloated. The middle layer of the deep learning model provided in this application is composed of a multi-layer network. Through layer-by-layer nonlinear transformation, it can better capture the complex patterns and hierarchical features in the data, and ensure the accuracy of the final estimation result. By optimizing the network architecture, a multi-layer small-scale network is used in the middle layer instead of a large-scale network, which not only reduces the number of model parameters, but also ensures the accuracy of model estimation.
[0103] As a specific embodiment, the present invention is constructed as follows Figure 6 The 5-layer network shown is used to implement the deep learning model, and the input parameters on the input side include the single cell voltage, single cell temperature, ohmic internal resistance, polarization internal resistance, etc. of the target battery cell; the middle layer is set up with four layers, and the number of neurons in the four middle layers is 6, 8, 4, and 2 respectively, and the output data of the output layer is the SOC value of the target battery cell. The output of each layer of the network can be processed nonlinearly using the Relu function, and the number of parameters of the entire model is only 141. When the middle layer is a single-layer network, it can only go directly from input to output through a layer of nonlinear transformation, which makes it difficult to capture complex patterns and hierarchical features. In this application, a four-layer network is used to implement the middle layer. The first layer of the network (6 nodes) can learn low-level features; the second layer of the network (8 nodes) can combine low-level features to form intermediate features; the third layer of the network (4 nodes) can combine intermediate features to form high-level features. The fourth layer of the network (2 nodes) can further combine features to obtain the final output.
[0104] Each network layer in the middle layer includes a fully connected layer and an activation function. The main function of the fully connected layer is to calculate the linear combination of the input. Each neuron receives all the inputs from the previous layer and calculates the output value through weighted summation and bias terms. The specific calculation formula of the fully connected layer is: y=k·x+b, where: y is the output of the fully connected layer, k is the weight matrix, x is the input vector, and b is the bias vector. The role of the activation function is to introduce nonlinear relationships. By performing a nonlinear transformation on y through the linear combination output of the fully connected layer, the activation function enables the neural network to learn and represent complex nonlinear functions.
[0105] It is understandable that since the major storage manufacturers on the market use STM32F103 as the processor of BMU, it does not have the function of accelerating floating-point operations, and the computing resources are limited. Therefore, the deep learning model constructed by the present invention needs to be able to adapt to the computing power of the processor. This application improves the computing power of the deep learning model by adopting a distributed computing method, and completes the SOC estimation of 52 cells on the STM32F103 processor in only 19 milliseconds. For the BMU system, the self-test and data acquisition of the system need to be completed within 1 second, and the acquisition task is executed every 1 millisecond. In order to ensure the stability of data acquisition, the system cannot wait too long when running tasks. Therefore, the calculation process of the model is carried out step by step, which is divided into 15 steps in total, and each step is executed for about 1 millisecond. The SOC calculation of all the single cells in a battery pack can be completed between 1 millisecond, which fully meets the needs of real-time processing of the BMU system.
[0106] It should be noted that the deep learning model constructed in this application will calculate the fully connected layer and activation function of each layer separately, that is, one layer will be divided into three calculations. The first step of the fully connected layer can calculate k·x first, the second step calculates k·x+b, and the third step calculates the activation function. In this way, the output of the network of the middle layer is obtained. The 5-layer middle layer is divided into 15 steps for calculation, and the computing power of the deep learning model is improved through the process of distributed calculation.
[0107] Please refer to Figure 8 , Figure 8 A schematic diagram for comparing the estimated results of the state of charge of a single battery cell provided by the present invention; in order to verify the effectiveness of the algorithm and the accuracy of SOC estimation, two BMS systems were used in the experiment. The method provided by the present invention was deployed to one of the BMS systems for operation. The other BMS system was independently designed and can collect the current of a single battery cell and then calculate the real SOC of the battery cell using the ampere-hour integration method and the open-circuit voltage method. The two systems tested the same battery cell at the same time, and some data of the difference in the test results are shown in Figure 1. Figure 8As shown in the figure, the X-axis is the number of SOC result comparisons, and the Y-axis is the difference in SOC percentage. The difference between the SOC estimated value and the true value of the deep learning model is calculated to be within 0±1.3%, and this estimation accuracy meets the needs of industrial detection.
[0108] Specifically, this application proposes a deep learning SOC estimation method with a 5-layer network, designs a deep learning model suitable for large-scale energy storage systems, combines a multi-layer network structure and an optimization algorithm, and ensures the accuracy and stability of the deep learning model in complex environments.
[0109] Please refer to Fig. 9 , Fig. 9 The present invention provides a schematic diagram of a system for determining the state of charge of a single cell provided by the present invention. To solve the above technical problems, the present invention also provides a system for determining the state of charge of a single cell, comprising:
[0110] An operating parameter determination unit 11 is used to determine the operating parameters of the target battery cell, where the operating parameters of the target battery cell include the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0111] An internal resistance calculation unit 12, used to determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located;
[0112] The estimation unit 13 is used to input the operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell into a pre-established deep learning model to obtain an estimation result of the state of charge of the target battery cell.
[0113] As an optional embodiment, the operating parameter of the target battery cell further includes an average voltage of the battery pack where the target battery cell is located; and the charge state determination system of the single battery cell further includes:
[0114] A voltage collection unit, used to collect the operating voltage of the battery pack where the target battery cell is located;
[0115] The average voltage calculation unit is used to determine the average voltage of the battery pack where the target battery cell is located based on the operating voltage of the battery pack where the target battery cell is located and the set number of battery cells in the battery pack.
[0116] As an optional embodiment, the internal resistance calculation unit 12 includes:
[0117] A charge and discharge unit, used to construct a second-order RC equivalent circuit model of the target battery cell and perform a charge and discharge test on the target battery cell;
[0118] A voltage change determination unit, used to determine in real time the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test;
[0119] The internal resistance calculation subunit is used to calculate the ohmic internal resistance and polarization internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test.
[0120] As an optional embodiment, the internal resistance calculation subunit includes:
[0121] A voltage extreme value determination unit, used to determine the minimum voltage and maximum voltage of the target battery cell during the charge and discharge test;
[0122] a voltage difference determination unit, configured to determine a difference between the maximum voltage and the minimum voltage;
[0123] The ohmic internal resistance calculation unit is used to determine the ohmic internal resistance of the target battery cell by taking the ratio of the difference to the working current of the battery pack where the target battery cell is located.
[0124] As an optional embodiment, the internal resistance calculation subunit includes:
[0125] A function generating unit, used for constructing a state response function of the operating voltage of the target battery cell;
[0126] The polarization internal resistance calculation unit is used to substitute the operating voltage of the target battery cell and the operating current of the battery pack where the corresponding target battery cell is located into the state response function of the operating voltage of the target battery cell to calculate the polarization internal resistance of the target battery cell.
[0127] As an optional embodiment, the system for determining the state of charge of a single battery cell further includes:
[0128] A model building unit, used to build a deep learning model with the working parameters, ohmic internal resistance and polarization internal resistance of a single cell as input and the charge state of the single cell as output;
[0129] The model training unit is used to train the deep learning model by using the working parameters, ohmic internal resistance and polarization internal resistance of a single cell with a known state of charge as a training set.
[0130] For an introduction to a system for determining the state of charge of a single cell provided by the present invention, please refer to the embodiment of the method for determining the state of charge of a single cell described above, and the present invention will not be repeated here.
[0131] Please refer to Fig.10 , Fig.10 The present invention provides a schematic diagram of the structure of an electronic device with a single battery cell. To solve the above technical problems, the present invention also provides an electronic device, including:
[0132] A memory 21, used for storing computer programs;
[0133] The processor 22 is used to implement the steps of the method for determining the state of charge of a single cell as described above.
[0134] Among them, the processor 22 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 22 can be implemented in at least one hardware form of DSP (Digital Signal Processor), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 22 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit; the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 22 may integrate a GPU (graphics processing unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 22 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.
[0135] The memory 21 may include one or more computer-readable storage media, which may be non-transitory. The memory 21 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 21 is at least used to store the following computer program, wherein, after the computer program is loaded and executed by the processor 22, it can implement the relevant steps of the method for determining the state of charge of a single cell disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 21 may also include an operating system and data, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system may include Windows, Unix, Linux, etc. The data may include, but is not limited to, data of the method for determining the state of charge of a single cell, etc.
[0136] In some embodiments, the electronic device may further include a display screen, an input and output interface, a communication interface, a power supply, and a communication bus.
[0137] It can be understood by those skilled in the art that Fig.10 The structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure.
[0138] For an introduction to an electronic device provided by the present invention, please refer to the embodiment of the method for determining the state of charge of a single cell, which will not be described in detail herein.
[0139] To solve the above technical problem, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for determining the state of charge of a single battery cell as described above are implemented.
[0140] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the method described in each embodiment of the present application. Specifically, the computer-readable storage medium may include but is not limited to any type of disk, including a floppy disk, an optical disk, and a mobile hard disk, or any type of medium or device suitable for storing instructions, data, etc., and the present application does not make any special restrictions here.
[0141] For an introduction to a computer-readable storage medium provided by the present invention, please refer to the embodiment of the method for determining the state of charge of a single cell described above, and the present invention will not be repeated here.
[0142] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. The term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.
[0143] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0144] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
[0145] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for determining the state of charge of a single cell, characterized in that: include: Determine operating parameters of the target battery cell, wherein the operating parameters of the target battery cell include an operating voltage of the target battery cell and an operating current of a battery pack in which the target battery cell is located; Determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located; The operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell are input into a pre-established deep learning model to obtain an estimation result of the state of charge of the target battery cell.
2. The method for determining the state of charge of a single cell according to claim 1, wherein: The operating parameters of the target battery cell also include the average voltage of the battery pack where the target battery cell is located; and the method for determining the state of charge of the single battery cell also includes: Collecting the operating voltage of the battery pack where the target battery cell is located; The average voltage of the battery pack where the target battery cell is located is determined based on the operating voltage of the battery pack where the target battery cell is located and the number of battery cells in the battery pack.
3. The method for determining the state of charge of a single cell according to claim 1, wherein: The step of determining the ohmic internal resistance and the polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located includes: Constructing a second-order RC equivalent circuit model of the target battery cell, and performing a charge and discharge test on the target battery cell; Determine in real time the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test; The ohmic internal resistance and polarization internal resistance of the target battery cell are calculated according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test.
4. The method for determining the state of charge of a single cell according to claim 3, characterized in that: The step of calculating the ohmic internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test includes: Determine the minimum voltage and maximum voltage of the target battery cell during the charge and discharge test; determining a difference between the maximum voltage and the minimum voltage; The ratio of the difference to the operating current of the battery pack where the target battery cell is located is determined as the ohmic internal resistance of the target battery cell.
5. The method for determining the state of charge of a single cell according to claim 3, characterized in that: Calculating the polarization internal resistance of the target battery cell according to the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located during the charge and discharge test, including: Constructing a state response function of the operating voltage of the target battery cell; The operating voltage of the target battery cell and the operating current of the battery pack in which the corresponding target battery cell is located are substituted into the state response function of the operating voltage of the target battery cell to calculate the polarization internal resistance of the target battery cell.
6. The method for determining the state of charge of a single cell according to any one of claims 1 to 5, characterized in that: Also includes: Build a deep learning model with the operating parameters, ohmic internal resistance and polarization internal resistance of the single cell as input and the state of charge of the single cell as output; The deep learning model is trained by using the operating parameters, ohmic internal resistance and polarization internal resistance of a single cell with a known state of charge as a training set.
7. The method for determining the state of charge of a single cell according to claim 6, characterized in that: The deep learning model includes an input layer, an intermediate layer and an output layer. The intermediate layer includes a multi-layer network, and the number of neurons in each layer of the network is less than a preset value.
8. A system for determining the state of charge of a single cell, characterized in that: include: An operating parameter determination unit, used to determine the operating parameters of the target battery cell, wherein the operating parameters of the target battery cell include the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located; An internal resistance calculation unit, used to determine the ohmic internal resistance and polarization internal resistance of the target battery cell based on the operating voltage of the target battery cell and the operating current of the battery pack where the target battery cell is located; An estimation unit is used to input the operating parameters of the target battery cell, the ohmic internal resistance of the target battery cell, and the polarization internal resistance of the target battery cell into a pre-established deep learning model to obtain an estimation result of the state of charge of the target battery cell.
9. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, used to implement the steps of the method for determining the state of charge of a single battery cell as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining the state of charge of a single battery cell as claimed in any one of claims 1 to 7 are implemented.
