Storage battery state evaluation method and related equipment thereof

Abstract

The invention discloses a storage battery state evaluation method and related equipment thereof, and is applied to the field of automobiles. According to the method, firstly, a storage battery state calculation model is calibrated based on the percentage SOC of the residual capacity of the storage battery, the storage battery state calculation model is used for calculating parameters of the storage battery, and then the calibrated storage battery state calculation model is used for calculating the parameters of the storage battery. And inputting the parameters into a storage battery state evaluation model based on a neural network, and calculating the SOH based on the parameters by using the storage battery state evaluation model. And finally, sending the SOH obtained by calculation to the storage battery state calculation model to update the SOH of the storage battery so as to determine a state evaluation result of the storage battery. According to the invention, the health state of the storage battery can be accurately predicted to a certain extent.

Description

Technical Field

[0001] This application relates to the technical field of automobiles, and particularly relates to a method for evaluating the state of a storage battery and related equipment thereof. Background Art

[0002] As the low-voltage power supply of the whole vehicle, the storage battery can provide energy for vehicle starting or provide energy for the electrical appliances of the whole vehicle when the engine is not running or the high-voltage battery is not working. Effectively predicting the remaining capacity percentage (SOC) and state of health (SOH) of the storage battery and timely feedback to the whole vehicle and the user side is a very important topic in the automotive field. However, in the prior art, the state of health of the storage battery cannot be accurately predicted. Summary of the Invention

[0003] In view of this, the embodiments of this application provide a method for evaluating the state of a storage battery and related equipment thereof, aiming to accurately predict the state of health of the storage battery.

[0004] In a first aspect, the embodiments of this application provide a method for evaluating the state of a storage battery, and the method includes:

[0005] Calibrating a storage battery state calculation model based on the difference situation between the calculation results of the remaining capacity percentage SOC of the storage battery, where the storage battery state calculation model is used to calculate the parameters of the storage battery, and the parameters include SOC;

[0006] Calculating the parameters of the storage battery by using the calibrated storage battery state calculation model, inputting the parameters into a storage battery state evaluation model based on a neural network, and calculating the state of health SOH based on the parameters by using the storage battery state evaluation model;

[0007] Sending the calculated SOH to the storage battery state calculation model to update the SOH of the storage battery, so as to determine the state evaluation result of the storage battery. Optionally, calibrating the storage battery state calculation model based on the difference situation between the calculation results of the remaining capacity percentage SOC of the storage battery includes:

[0008] Obtaining a first SOC by using a first calculation method;

[0009] Obtaining a second SOC by using a second calculation method;

[0010] Setting a target calibration value, where the target calibration value is used to determine the accuracy of the storage battery state calculation model;

[0011] When the difference situation between the first SOC and the second SOC is that the difference is greater than the target calibration value, calibrating the storage battery state calculation model.

[0012] Optionally, calibrating the battery state calculation model includes:

[0013] Taking the cumulative charge and discharge ampere-hours, current, remaining capacity percentage, and internal resistance of the battery as input parameters of the battery state evaluation model;

[0014] Using the battery state evaluation model to calculate the target health state of the battery based on the input parameters;

[0015] Calibrating the battery state calculation model based on the target health state.

[0016] Optionally, using the battery state evaluation model to calculate the target health state based on the input parameters includes:

[0017] Normalizing the input parameters to obtain processed input parameters;

[0018] Initializing the parameters of the battery state evaluation model;

[0019] Calculating the target health state using the battery state evaluation model based on the processed input parameters.

[0020] Optionally, sending the calculated SOH to the battery state calculation model to update the SOH of the battery includes:

[0021] Collecting the parameter data of the battery;

[0022] Taking the parameter data as a sample to establish a mapping relationship between the parameter data of the battery and SOH;

[0023] Using the battery state evaluation model to calculate and determine the current SOH of the battery according to the parameter data.

[0024] Optionally, calibrating the battery state calculation model based on the target health state includes:

[0025] Updating the SOH in the battery state calculation model based on the target health state;

[0026] Based on the updated SOH, rechecking the difference between the first SOC and the second SOC;

[0027] Adjusting the battery state calculation model according to the verification result until the difference is less than the target calibration value, and completing the calibration of the battery state calculation model.

[0028] Optionally, establishing the mapping relationship between the parameter data of the battery and SOH includes:

[0029] Perform a charge and discharge experiment on the storage battery. When the experiment is completed, perform a charging operation on the storage battery until the charge amount of the storage battery meets a preset threshold;

[0030] Determine the remaining capacity and the initial capacity of the storage battery on which the charging operation is completed;

[0031] Determine the SOH based on the remaining capacity and the initial capacity;

[0032] Establish a mapping relationship between the parameter data of the storage battery and the SOH.

[0033] In a second aspect, an embodiment of the present application provides a device for evaluating the state of a storage battery. The device includes: a calibration module, a calculation module, and a result obtaining module;

[0034] The calibration module is configured to calibrate a storage battery state calculation model based on the difference situation between the calculation results of the remaining capacity percentage SOC of the storage battery. The storage battery state calculation model is used to calculate the parameters of the storage battery, and the parameters include SOC;

[0035] The calculation module is configured to calculate the parameters of the storage battery by using the calibrated storage battery state calculation model, input the parameters into a storage battery state evaluation model based on a neural network, and calculate the health state SOH based on the parameters by using the storage battery state evaluation model;

[0036] The result obtaining module is configured to send the calculated SOH to the storage battery state calculation model to update the SOH of the storage battery, so as to determine the state evaluation result of the storage battery.

[0037] In a third aspect, the present application provides an electronic device. The device includes: a processor, a memory, and a system bus;

[0038] The processor and the memory are connected through the system bus;

[0039] The memory is used to store one or more programs. The one or more programs include instructions, and when the instructions are executed by the processor, the processor is caused to execute the method described in the first aspect.

[0040] In a fourth aspect, an embodiment of the present application provides a vehicle, and the vehicle is configured with the electronic device described in the third aspect.

[0041] In a fifth aspect, an embodiment of the present application provides a computer storage medium. Code is stored in the computer storage medium, and when the code is run, the device running the code implements the method described in any one of the foregoing first aspects.

[0042] The present application provides a method, device, electronic device, vehicle, and storage medium for evaluating the state of a storage battery. When executing the method, first, the storage battery state calculation model is calibrated based on the state of charge percentage (SOC) of the storage battery. The storage battery state calculation model is used to calculate the parameters of the storage battery. Then, the calibrated storage battery state calculation model is used to calculate the parameters of the storage battery. The parameters are input into the storage battery state evaluation model based on a neural network. The storage battery state evaluation model calculates the state of health (SOH) based on the parameters. Finally, the calculated SOH is sent to the storage battery state calculation model to update the SOH of the storage battery, so as to determine the state evaluation result of the storage battery. In this way, by correcting the charge and discharge power at different SOC states to show its actual influence on SOH, combined with the internal resistance, SOC, and current of the storage battery, a neural network algorithm is used to more accurately predict the SOH of the storage battery online. Specifically, the accuracy of the storage battery state calculation model can be determined through the deviation of the storage battery SOC. When the deviation exceeds the set value, the relevant change data of the storage battery after calculation and processing is sent to the cloud to calibrate its own model. The calibrated storage battery state calculation model is used to calculate the relevant parameters of the storage battery, and the calculation results are input into the storage battery state evaluation model based on a neural network. Using this model, the SOH and its changes can be predicted. Since the influence of the charge and discharge amount of the storage battery on SOH is different under different SOCs, by combining the evaluation of SOC to predict and update the health state of the vehicle's storage battery, the monitoring accuracy of the storage battery state can be improved, and vehicle storage battery failures can be warned or avoided in a timely manner. Description of the Drawings

[0043] To more clearly illustrate the technical solutions in the embodiments or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0044] Figure 1 It is a flowchart of a method for evaluating the state of a storage battery provided by an embodiment of the present application;

[0045] Figure 2 It is a flowchart of a method for calibrating a storage battery state calculation model provided by an embodiment of the present application;

[0046] Figure 3 It is a flowchart of the training and verification algorithm of a storage battery state evaluation model provided by an embodiment of the present application;

[0047] Figure 4System flowchart for evaluating the state of a storage battery in an application scenario provided by an embodiment of the present application;

[0048] Figure 5 Schematic structural diagram of a device for evaluating the state of a storage battery provided by an embodiment of the present application;

[0049] Figure 6 Schematic structural diagram of a vehicle system provided by an embodiment of the present application. Detailed implementation manners

[0050] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] In the present application, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0052] In the research on related technologies, it is found that as the low-voltage power supply of the whole vehicle, the storage battery can provide energy for vehicle starting or provide energy for the electrical appliances of the whole vehicle when the engine is not running or the high-voltage battery is not working. The normal operation of the storage battery is very important for the whole vehicle. Therefore, effectively predicting the state of charge percentage (SOC) and state of health (SOH) of the storage battery and timely feedback to the whole vehicle and the user side is a very important topic in the automotive field. At present, the method for predicting the remaining SOC of the storage battery is generally obtained by adding or subtracting the ampere-hour integral of the current for a period of time from the initial capacity. However, as an electrochemical energy storage unit, the SOH of the storage battery will gradually decrease with use, and different usage conditions have different effects on the SOH, resulting in a large deviation in the predicted current SOC of the storage battery after the vehicle has been used for a period of time. For the detection of the health degree, it can be carried out by separately performing charge and discharge tests on the storage battery. Although this method can obtain the SOH of the storage battery more accurately, it cannot meet the purpose of detecting and calibrating the SOC during the vehicle use process. In addition, some people also evaluate the impact on the SOH of the storage battery by statistically counting the charge and discharge power of the storage battery within a period of time through life curves or big data models, etc. However, due to the different effects of the charge and discharge power of the storage battery on the SOH under different SOCs, and this method ignores the evaluation of the SOC state of the storage battery, resulting in low accuracy.

[0053] Based on this, the present application proposes a method, device, electronic device, vehicle and storage medium for evaluating the state of a storage battery. It can correct the charge and discharge power at different SOC states to show its actual impact on the SOH, and combine the internal resistance, SOC and current of the storage battery, and use a neural network algorithm to more accurately predict the SOH of the storage battery online.

[0054] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0055] Figure 1 For the flowchart of a method for evaluating the state of a storage battery provided by an embodiment of the present application, see Figure 1 As shown, a method for evaluating the state of a storage battery provided by an embodiment of the present application includes:

[0056] S11: Calibrate the storage battery state calculation model based on the difference situation between the calculation results of the state of charge percentage SOC of the storage battery, and the storage battery state calculation model is used to calculate the parameters of the storage battery.

[0057] A battery state calculation model provided in this embodiment aims to continuously update and obtain accurate state of charge (SOC) and state of health (SOH) of the battery. This model determines the accuracy of the model by comparing the SOCs calculated by two methods. After the deviation exceeds the set value, the relevant changed data of the battery after calculation and processing is sent to the cloud to calibrate its own model.

[0058] In step S11, it is mentioned that "the battery state calculation model is calibrated according to the difference situation between the calculation results of SOC". Figure 2 The figure is a flowchart of a method for calibrating a battery state calculation model provided in an embodiment of this application. As Figure 2 shown, this method specifically includes:

[0059] S111: Obtain the first SOC by using the first calculation method.

[0060] The SOC of a battery refers to the percentage of the current remaining capacity of the battery to the rated capacity. It can be calculated by the following formula: SOC = current energy storage / rated capacity. The value of SOC is usually between 0 and 100%, indicating the charge and discharge state of the battery. The SOC of the battery can change over time. If the battery continuously releases energy, then the SOC will gradually decrease until it reaches the limit of the battery's storable energy. In addition, as the service life of the battery increases, its capacity will decay, which will also affect the value of SOC. Therefore, the SOC of the battery is a dynamically changing index, which reflects the current charge and discharge state of the battery.

[0061] The methods for calculating SOC can include but are not limited to the following:

[0062] (1) AC impedance method

[0063] The internal resistance of the battery is divided into ohmic internal resistance and polarization internal resistance, and the electrical model is a complex impedance model composed of a resistor and a capacitor. This method is to apply a sinusoidal constant current signal across the battery and measure the voltage response at both ends of the battery to calculate the complex impedance value.

[0064] (2) Open-circuit voltage method

[0065] When a lead-acid battery is in a stable state (usually referring to after the battery has been static for a long time), there is a good linear relationship between the open-circuit voltage of the battery and the remaining capacity, and this linear relationship is less affected by the external temperature and battery aging. Based on the linear relationship between the state of charge of the battery and the open-circuit voltage, the SOC can be determined.

[0066] (3) Battery terminal voltage method

[0067] This method performs on-line detection of the battery terminal voltage and then finds the corresponding remaining capacity according to the discharge curve.

[0068] (4) Ampere-hour integration method

[0069] Also known as the AH method, its principle is to integrate the charge and discharge current over time, calculate the charge and discharge amount of the battery, and then obtain the remaining capacity based on the total capacity of the battery. This detection method requires high-speed sampling of the current, and at the same time, issues such as temperature compensation and charge and discharge efficiency need to be considered.

[0070] In the first calculation method in the embodiments of the present application, it can be the open-circuit voltage method, and the value of the first SOC can be calculated using the open-circuit voltage method. It can be understood that the first calculation method can also be other methods capable of calculating SOC other than the open-circuit voltage method. The specific method for calculating SOC can be selected by those skilled in the art according to the actual situation and application scenario, and is not limited herein.

[0071] S112: Obtain the second SOC using the second calculation method.

[0072] In the second calculation method in the embodiments of the present application, it can be the ampere-hour integration method, and the value of the second SOC can be calculated using the ampere-hour integration method. It can be understood that the second calculation method can also be other methods capable of calculating SOC other than the ampere-hour integration method. The specific method for calculating SOC can be selected by those skilled in the art according to the actual situation and application scenario, and is not limited herein. The above-mentioned first calculation method and second calculation method can be different, and the above is only one example.

[0073] It should also be noted that the "first" and "second" mentioned in the embodiments of the present application are only for distinction, and do not reflect meanings such as priority, importance level, and sequence.

[0074] S113: Set a target calibration value, and the target calibration value is used to determine the accuracy of the battery state calculation model.

[0075] The target calibration value can be understood as the maximum allowable deviation value between the calculated SOC values. When the difference between the SOC values exceeds the target calibration value, it can be considered that there is a problem with the accuracy of the battery state calculation model, and the battery state calculation model needs to be calibrated. The target calibration value can be set by those skilled in the art according to the actual situation and application scenario, and is not limited herein.

[0076] S114: When the difference between the first SOC and the second SOC is greater than the target calibration value, calibrate the battery state calculation model.

[0077] The difference between the first SOC and the second SOC may include: ① The difference between the first SOC and the second SOC is greater than the target calibration value; ② The difference between the first SOC and the second SOC is less than the target calibration value. When the difference is greater than the target calibration value, the battery state calculation model needs to be calibrated. When the difference is less than the target calibration value, the battery state calculation model does not need to be calibrated. It should also be noted that the above-mentioned target calibration value can be set by those skilled in the art according to the actual situation and application scenario, and is not limited here.

[0078] The specific method of "calibrating the battery state calculation model" mentioned in step S114 can be: First, use the cumulative charge and discharge ampere-hours, current, remaining capacity percentage, and internal resistance of the battery as the input parameters of the battery state evaluation model. Then, use the battery state evaluation model to calculate the target health state of the battery based on the input parameters. Finally, calibrate the battery state calculation model based on the target health state.

[0079] Among them, different cumulative charge and discharge ampere-hours Q, current I, remaining capacity percentage SOC, and internal resistance R of the battery i have different effects on the change rate of SOH. Therefore, the situations of the above parameters corresponding to each certain time interval within a period of time can be recorded. The parameter situations at different times will correspond to different change rates of SOH, thereby affecting the value of SOH.

[0080] The main function of the above-mentioned battery state evaluation model is to receive various parameters of the battery input by the controller, fit and update the SOH of the battery according to these parameters, and return it to the controller. The battery state evaluation model can be a neural network calculation model, including three parts: an input layer, a hidden layer, and an output layer. The input layer contains 4 neurons [Q, I, SOC, R i , and the output layer is the SOH change rate [SOH']. The hidden layer is used to construct a non-linear mapping relationship between the input and the output.

[0081] Figure 3 It is a flowchart of the training and verification algorithm for a battery state evaluation model provided by an embodiment of the present application. As Figure 3 shown, the method specifically includes:

[0082] First, the training samples (the cumulative charge and discharge ampere-hours, current, remaining capacity percentage, and internal resistance data of the storage battery mentioned above) are normalized and then input into the storage battery state evaluation model based on a neural network. Then, the parameters of the storage battery state evaluation model are initialized. The specific parameters may include: the maximum number of iterations, learning accuracy, number of hidden layer nodes, initial weights, thresholds, and initial learning rate, etc. Calculate the input and output values of each layer, and calculate the error E(q) of the output layer. Determine whether E(q) is less than the preset threshold ε. If not, calculate the error gradient, and after modifying the weights and thresholds according to the error gradient, perform the calculation process again; if not, end.

[0083] After the training and verification are completed, the data Q = [Q 1 , Q 2 , …, Q n recorded and calculated by the input controller within time t, I = [I 1 , I 2 , …, I n , SOC = [SOC 1 , SOC 2 , …, SOC n , R i = [R 1 , R 2 , …, R n . The data interval is a fixed value Δt. The change rate of the output SOH [SOH 1 ′, SOH 2 ′, …, SOH′ n can be obtained. Based on this change rate, the target health state SOH can be calculated, and then the target health state SOH is returned to the storage battery state calculation model to update the SOH value of the storage battery state calculation model. Based on the updated SOH, the SOC deviation is re-verified, that is, the storage battery state calculation model is calibrated.

[0084] Through the method of calibrating the storage battery state calculation model mentioned above, the storage battery state evaluation model is trained and verified using the various parameters of the storage battery. After the training and verification are completed, the various parameters of the storage battery input by the controller are received, such as: [Q, I, SOC, R i . The SOH of the storage battery is fitted and updated and returned to the storage battery state calculation model. The storage battery state calculation model uses this SOH to re-verify the deviation of the SOC, realizing the calibration of the storage battery state calculation model.

[0085] It is mentioned above that "calculating the target state of health based on the input parameters using the battery state evaluation model". The specific method can be as follows: First, normalize the input parameters to obtain the processed input parameters. Then, initialize the parameters of the battery state evaluation model. Finally, calculate the target state of health using the battery state evaluation model based on the processed input parameters.

[0086] The reason for normalizing the input parameters is that different parameters vary greatly in value, which will reduce the convergence speed of the weights and thresholds of the battery state evaluation model. The specific normalization method can be the BN normalization calculation method. After normalizing the input parameters, train and validate the battery state evaluation model. After the battery state evaluation model is trained, the change rate of the output SOH [SOH 1 ′, SOH 2 ′,..., SOH′ n can be obtained according to the data recorded and calculated by the vehicle domain controller within time t. Based on this change rate, the target state of health can be calculated.

[0087] Through the method of calculating the target state of health mentioned above, specifically, normalizing the input parameters can improve the convergence speed of the model, thereby improving the running speed and efficiency of the model.

[0088] Through the method of calibrating the battery state calculation model mentioned above, judge the accuracy of the model by comparing the SOC calculated by two methods. After the deviation between the SOCs exceeds the set value, send the calculated and processed battery-related change data to the battery state evaluation model to calibrate its own model.

[0089] The method of "calibrating the battery state calculation model based on the target state of health" mentioned above can also be as follows: First, update the SOH in the battery state calculation model based on the target state of health. Then, recheck the difference between the first SOC and the second SOC based on the updated SOH. Finally, adjust the battery state calculation model according to the check result until the difference is less than the target calibration value, and complete the calibration of the battery state calculation model.

[0090] Specifically, when the vehicle domain controller receives the SOH value returned by the cloud, it updates the SOH value of the state calculation model, and based on the updated SOH, rechecks the SOC deviation. If the deviation value is still greater than the target calibration value, data for a period of time is repeatedly accumulated, and the neural network algorithm is fitted again, or the parameters of the neural network algorithm are adjusted, such as changing the number of hidden layer nodes h and the learning rate α, until a battery state calculation model with an SOC deviation meeting the requirements is obtained, and the calibration of the battery state calculation model is completed.

[0091] Calibrating the battery state calculation model can facilitate improving the accuracy of subsequent battery state assessment.

[0092] S12: Calculate the parameters of the battery using the calibrated battery state calculation model, input the parameters into the battery state assessment model based on the neural network, and use the battery state assessment model to calculate the state of health SOH based on the parameters.

[0093] In step S12, it is mentioned that "send the calculated SOH to the battery state calculation model to update the SOH of the battery to determine the state assessment result of the battery". The specific method can be: first collect the parameter data of the battery, then use the parameter data as a sample to establish the mapping relationship between the parameter data of the battery and SOH. Finally, use the battery state assessment model to calculate and determine the current SOH of the battery according to the parameter data.

[0094] Specifically, for the battery state assessment model based on the neural network, using the data in the battery parameter database as samples, a mapping relationship between the input battery parameter data and the battery SOH is constructed. Then, based on this battery state assessment model, using the battery parameters sent by the battery state calculation model as input data, the SOH and its changes are predicted. Specific battery parameters can include but are not limited to: Q (accumulated charge and discharge ampere-hours of the battery), I (current), SOC (remaining capacity percentage), and R i (internal resistance of the battery), etc.

[0095] The data in the above-mentioned battery parameter database consists of two parts: battery parameter data collected from laboratory tests and calculated by the vehicle. In the laboratory, a charge and discharge test is performed on the battery using a charge and discharge device. The charge and discharge test uses Q (accumulated charge and discharge ampere-hours of the battery), I (current), and SOC as discrete variables, and the battery is tested in interval, charged and discharged at I current, and by charged / discharged to when the change value ΔSOH of SOH. During the charge and discharge test, Q, R i, SOC can be calculated. After the charge and discharge test is completed, after fully charging the battery, the state of health (SOH) is calculated by comparing the remaining capacity C with the initial capacity C0. During the calculation process, the average change rate of SOH can be obtained, and the average change rate of SOH is used as the mapping value of discrete variables Q, I, SOC, R i That is, a function mapping relationship is constructed. Based on the aforementioned mapping relationship, the state evaluation result of the battery can be obtained by using the battery state evaluation model.

[0096] The above-mentioned "establishing the mapping relationship between the parameter data of the battery and SOH" can be implemented as follows: First, conduct a charge and discharge experiment on the battery. When the experiment is completed, perform a charging operation on the battery until the charge of the battery meets a preset threshold. Then determine the remaining capacity and the initial capacity of the battery after the charging operation is completed, and determine SOH based on the remaining capacity and the initial capacity. Finally, establish the mapping relationship between the parameter data of the battery and SOH.

[0097] The specific process can be as follows: In the laboratory, a charge and discharge device is used to conduct a charge and discharge test on the battery. The charge and discharge test uses Q, I, SOC as discrete variables, and tests the battery in a certain interval, charges and discharges at current I, and from charge / discharge to the change value ΔSOH of SOH at this time. During the charge and discharge test process, Q, R i , SOC can be calculated by formulas. After the charge and discharge test is completed, after fully charging the battery, the remaining capacity C is compared with the initial capacity C 0 to calculate SOH. Finally, establish the mapping relationship between the parameter data of the battery and SOH. The above-mentioned charge of the battery meeting the preset threshold can be understood as when the battery reaches the full charge state. The specific preset threshold of the charge amount can be determined by those skilled in the art according to the actual situation and application scenarios, and is not limited here.

[0098] S13: Send the calculated SOH to the battery state calculation model to update the SOH of the battery, so as to determine the state evaluation result of the battery.

[0099] By sending the SOH calculated by the battery state evaluation model to the battery state calculation model, the SOH in the battery state calculation model can be updated to realize the prediction of the SOH of the battery, and then determine the current health state of the battery.

[0100] In this embodiment, a method for evaluating the state of a storage battery is proposed. First, the state calculation model of the storage battery is calibrated based on the state of charge percentage (SOC) of the storage battery. The state calculation model of the storage battery is used to calculate the parameters of the storage battery. Then, the calibrated state calculation model of the storage battery is used to calculate the parameters of the storage battery. The parameters are input into the state evaluation model of the storage battery based on a neural network. The state evaluation model of the storage battery is used to calculate the state of health (SOH) based on the parameters. Finally, the calculated SOH is sent to the state calculation model of the storage battery to update the SOH of the storage battery, so as to determine the state evaluation result of the storage battery. In this way, by correcting the charge and discharge power at different SOC states to show its actual impact on SOH, combined with the internal resistance, SOC, and current of the storage battery, a neural network algorithm is used to more accurately predict the SOH of the storage battery online. Specifically, the accuracy of the state calculation model of the storage battery can be determined through the deviation of the SOC of the storage battery. When the deviation exceeds the set value, the relevant change data of the storage battery after calculation and processing is sent to the cloud to calibrate its own model. The calibrated state calculation model of the storage battery is used to calculate the relevant parameters of the storage battery. The calculation results are input into the state evaluation model of the storage battery based on a neural network. The state evaluation model of the storage battery is used to predict the SOH and its changes, and the predicted SOH is sent to the state calculation model of the storage battery to perform real-time update of the SOH of the storage battery. Since the influence of the charge and discharge power of the storage battery on SOH is different under different SOCs, by combining the evaluation of SOC to predict and update the SOH of the vehicle-mounted storage battery, the monitoring accuracy of the state of the storage battery can be improved, and vehicle-mounted storage battery failures can be warned or avoided in a timely manner.

[0101] Figure 4 The system flowchart for evaluating the state of a storage battery in an application scenario provided by an embodiment of the present application is shown in Figure 4 As shown, the system mainly includes: a storage battery sensor, a vehicle domain controller, a storage battery laboratory, and a cloud platform. Among them, the database and the BP neural network calculation model are located on the cloud platform. The storage battery sensor is used to sample voltage and current, and the sampling process can be carried out by means of high-frequency sampling. The vehicle domain controller includes a state calculation model of the storage battery. The main functions of the state calculation model of the storage battery include: ① verifying the accuracy of the SOC; ② calculating the cumulative charge and discharge ampere-hours Q, the SOC of the storage battery, and the internal resistance R of the storage battery i ; ③ updating the SOH. The main function of the storage battery laboratory is to construct the mapping relationship between different Q, I, SOC, R i and the change rate of SOH through charge and discharge tests.

[0102] The storage battery sensor samples voltage and current and sends the sampled data to the vehicle domain controller. The vehicle domain controller calculates Q, I, SOC, and R based on the above sampled datai and the SOH change rate, the Q, I, SOC, and R that meet the deviation requirements in the calculation i and the SOH change rate are supplemented into the database as database samples. At the same time, the calculated Q, I, SOC, and R i are input into the BP neural network calculation model. The BP neural network calculation model outputs the predicted SOH and returns it to the vehicle domain controller for SOH update. The samples for training and validating the BP neural network calculation model include: Q, I, SOC, and R that meet the SOC deviation requirements sent by the vehicle domain controller i and the SOH change rate, as well as the mapping relationship between different Q, I, SOC, and R constructed by the battery laboratory through charge and discharge tests i and the SOH change rate.

[0103] The above-mentioned battery state calculation model aims to continuously update and obtain accurate battery SOC and SOH. This model judges the accuracy of the model by comparing the SOC calculated by two methods. After the deviation exceeds the set value, the relevant change data of the battery after calculation and processing is sent to the cloud to calibrate its own model.

[0104] The above-mentioned database stores the battery parameter data collected from laboratory dynamic test data, and constructs a battery parameter database for cloud calibration of battery SOH.

[0105] The neural network calculation model for predicting battery SOH uses the data in the battery parameter database as samples to construct the mapping relationship between the input battery parameter data and the battery SOH. Then, based on this model, using the battery data sent by the battery state calculation model as input, it predicts the SOH and its changes.

[0106] Figure 5 It is a schematic structural diagram of a device for evaluating battery state provided by an embodiment of the present application, as Figure 5 shown. A device for evaluating battery state specifically includes: a calibration module 100, a calculation module 200, and a result obtaining module 300;

[0107] The calibration module 100 is used to determine whether to calibrate the battery state calculation model based on the difference between the calculation results of the remaining capacity percentage SOC of the battery. The battery state calculation model is used to calculate the parameters of the battery, and the parameters include SOC;

[0108] The calculation module 200 is used to calculate the parameters of the battery using the calibrated battery state calculation model, input the parameters into the battery state evaluation model based on the neural network, and use the battery state evaluation model to calculate the health state SOH based on the parameters;

[0109] The result obtaining module 300 is configured to send the calculated SOH to the battery state calculation model to update the SOH of the battery, so as to determine the state evaluation result of the battery.

[0110] In a possible implementation manner, the calibration module 100 is specifically configured to:

[0111] Obtain a first SOC by using a first calculation method;

[0112] Obtain a second SOC by using a second calculation method;

[0113] Set a target calibration value, where the target calibration value is used to determine the accuracy of the battery state calculation model;

[0114] When the difference between the first SOC and the second SOC is greater than the target calibration value, calibrate the battery state calculation model.

[0115] In a possible implementation manner, the calibration module 100 is specifically configured to:

[0116] Use the cumulative charge and discharge ampere-hours, current, remaining capacity percentage, and internal resistance of the battery as input parameters of the battery state evaluation model;

[0117] Use the battery state evaluation model to calculate a target health state based on the input parameters;

[0118] Calibrate the battery state calculation model based on the target health state.

[0119] In a possible implementation manner, the calibration module 100 is specifically configured to:

[0120] Normalize the input parameters to obtain processed input parameters;

[0121] Initialize the parameters of the battery state evaluation model;

[0122] Calculate the target health state by using the battery state evaluation model based on the processed input parameters.

[0123] In a possible implementation manner, the result obtaining module 300 is specifically configured to:

[0124] Collect parameter data of the battery;

[0125] Use the parameter data as a sample to establish a mapping relationship between the parameter data of the battery and the SOH;

[0126] The state-of-health (SOH) of the battery is calculated and determined according to the parameter data by using the battery state evaluation model.

[0127] In a possible implementation manner, the calibration module 100 is specifically configured to:

[0128] Update the SOH in the battery state calculation model based on the target health state;

[0129] Based on the updated SOH, re-verify the difference between the first state of charge (SOC) and the second SOC;

[0130] Adjust the battery state calculation model according to the verification result until the difference is less than the target calibration value, and then complete the calibration of the battery state calculation model.

[0131] In a possible implementation manner, the result obtaining module 300 is specifically configured to:

[0132] Perform charge and discharge experiments on the battery. When the experiment is completed, perform a charging operation on the battery until the charge amount of the battery meets a preset threshold;

[0133] Determine the remaining capacity and the initial capacity of the battery that has completed the charging operation;

[0134] Determine the SOH based on the remaining capacity and the initial capacity;

[0135] Establish a mapping relationship between the parameter data of the battery and the SOH.

[0136] In this embodiment, a device for evaluating the state of a battery is proposed. The device includes a calibration module, a calculation module, and a result obtaining module. The calibration module is used to determine whether to calibrate the battery state calculation model based on the remaining capacity percentage SOC of the battery. The battery state calculation model is used to calculate the parameters of the battery. The calculation module is used to calculate the parameters of the battery by using the calibrated battery state calculation model, input the parameters into the battery state evaluation model based on a neural network, and calculate the health state SOH based on the parameters by using the battery state evaluation model. The result obtaining module is used to send the calculated SOH to the battery state calculation model to update the SOH of the battery, so as to determine the state evaluation result of the battery. In this way, by correcting the charge and discharge power at different SOC states to show its actual influence on the SOH, and combining the internal resistance, SOC, and current of the battery, a neural network algorithm is used to perform more accurate online prediction of the battery SOH.

[0137] An embodiment of the present application further provides a vehicle. Figure 6A structural schematic diagram of a vehicle system provided by an embodiment of the present application is as follows Figure 6 As shown, the vehicle system specifically includes: a battery and its sensors, a generator, and an engine (fuel vehicle) / a high-voltage and low-voltage conversion module and a power battery (electric vehicle), a vehicle domain controller, the cloud, the vehicle end, and the user end. Among them, the battery state calculation model mainly involves the battery sensors, the vehicle domain controller, and the cloud. The generator, the engine, the high-voltage and low-voltage conversion module, the power battery, the vehicle end, and the user end can be used to implement relevant strategies. An electronic device is configured in this vehicle, and a processor in the electronic device can execute the method for evaluating the battery state in the foregoing embodiments of the present application.

[0138] An embodiment of the present application also proposes a system for evaluating the battery state in an application scenario. Specifically:

[0139] This process mainly includes three parts: battery sensors, a vehicle domain controller, and the cloud and its database.

[0140] 1. Battery sensors

[0141] First, the function of the battery sensors is to detect and transmit battery data. It can monitor battery-related data (voltage V, current I) in real time and send the obtained data to the vehicle domain controller.

[0142] 2. Vehicle domain controller and battery state calculation model

[0143] In this system, the vehicle domain controller mainly provides a battery state calculation model. This model is configured with battery parameters at a certain aging degree of the current battery, including SOH, DOD-OCV curve (discharge depth-open circuit voltage curve). And, this model can calculate two SOCs (SOCI and SOCU) respectively by the open circuit voltage method and the ampere-hour integration method to verify the model accuracy. After a period of time, when the deviation between the two is greater than the calibration value (such as 10%), it is determined that the current model does not conform to the current battery health state, then the [Q, I, SOC, R i (accumulated charge and discharge ampere-hours of the battery, current, remaining capacity percentage, battery internal resistance) during this period is used as the input parameters of the cloud neural network algorithm model and input to the cloud to update [SOH]. In fact, different Q, I, SOC, R i have different effects on the change rate of SOH, so the Q, I, SOC, R i at fixed time intervals Δt during this period are recorded, that is, there is Q = [Q 1 , Q 2 , …, Q n , I = [I 1 , I 2 , …, I n, SOC = [SOC 1 , SOC 2 , …, SOC n , R i = [R 1 , R 2 , …, R n , [Q, I, SOC, R i at different times correspond to different rates of change of SOH [SOH'], which in turn affects the SOH value. The calculation methods of the parameters of this model are as follows:

[0144] ① SOH: In this model, the initial SOH is 100%, and it is updated later through the cloud neural network algorithm.

[0145] ② Ampere-hour of charge and discharge Q: Within time t, if t 0 - t 1 is the charging stage of the battery, and t 1 - t 2 is the discharging stage of the battery, then Q is the cumulative integral sum of the ampere-hour of charge and discharge of the battery, I Charge is the charging current, and I Discharge is the discharging current, as shown in Equation 2-1.

[0146]

[0147] ③ Internal resistance R i : When the current value changes > 200 mA within 1 ms, the battery sensor samples the current and voltage at a frequency of 1 kHz, collects data at n points, and the vehicle domain controller calculates the internal resistance R i of the battery, as shown in Equation 2-2, where ΔU bat refers to the voltage change difference within 1 ms, and ΔI bat refers to the current change difference within 1 ms. The purpose of high-frequency sampling is to reduce the single-sampling time, reduce the battery polarization effect, make the polarization-generated voltage much smaller than the voltage difference change of the battery itself, and thus can be ignored.

[0148]

[0149] ④ SOC: The initial SOC 0 is obtained by substituting the battery terminal voltage U 0 into the DOD-OCV curve. After the vehicle has run for a period of time, SOC can be calculated according to Equation 2-3:

[0150]

[0151] According to the battery terminal voltage U, substituting it into the DOD-OCV curve, we can get:

[0152]

[0153] where a and b are two coefficients (constants) of the DOD-OCV curve.

[0154] 3. Database

[0155] The data in the database consists of two parts: the battery parameter data collected from laboratory tests and the vehicle-level calculations. In the laboratory, charge and discharge tests are performed on the battery using charge and discharge equipment. The charge and discharge tests use Q, I, and SOC as discrete variables, and the battery is charged and discharged at current I within the interval, and the change value ΔSOH of SOH when charging / discharging from to is measured. During the charge and discharge test, Q, R i , and SOC can be calculated using formulas 2-1, 2-2, and 2-3. After the charge and discharge test, after fully charging the battery, the remaining capacity C is compared with the initial capacity C 0 to calculate SOH. The calculation methods of C and C 0 are the ampere-hour integral when discharging the battery at current I 20 after it is fully charged until the cut-off voltage of the battery is reached at current I 20 .

[0156] SOH = C / C 0 × 100% (2-5)

[0157]

[0158] Therefore, the change in SOH during this period t is:

[0159] ΔSOH = SOH 0 - SOH 1 (2-7)

[0160] SOH 0 is the SOH at , and SOH 1 is the SOH at . Further, the average change rate of SOH during this period is:

[0161]

[0162] Using this average change rate of SOH as the mapped value of variable R i , a function mapping relationship is constructed.

[0163] 4. Cloud and Neural Network Calculation Model

[0164] In this system, the cloud mainly constructs a BP neural network calculation model. By receiving various parameters of the battery [Q, I, SOC, R i input by the vehicle domain controller, it fits and updates the SOH of the battery and returns it to the vehicle domain controller. The implementation method is as follows:

[0165] This neural network calculation model consists of three parts: an input layer, a hidden layer, and an output layer. The input layer contains 4 neurons [Q, I, SOC, R i , and the output layer is the SOH change rate [SOH′]. The hidden layer is used to construct the non-linear mapping relationship between the input and the output. The algorithm flow of this neural network calculation model is as follows:

[0166] Among them, the calculation formulas for each layer are as follows (all represented in vector form):

[0167] 1) Input variable normalization

[0168] Perform BN normalization on the input variable X = [Q, I, SOC, R i , because [Q, I, SOC, R i have large numerical differences, which will reduce the convergence speed of the weights and thresholds. The BN normalization calculation method is as follows:

[0169]

[0170]

[0171]

[0172]

[0173] Among them, μ is the mean, σ 2 is the variance, is adjusted to make the sample data conform to a mean of 0 and a standard deviation of 1, γ and β are scale scaling and offset coefficients, m is the number of samples, and ε is a small number set to avoid division by 0.

[0174] 2) Forward propagation function A output calculation

[0175] A [l] = g(W [l] A [1-l] + b [l] ), l = 1, 2, … L (2-13)

[0176] In this formula, W represents the weight coefficient, b represents the threshold of the hidden layer or the output layer, and L represents the number of model layers. Among them, the neuron activation function g(x):

[0177] g(x) = max(0, x), l = 1, 2, … L-1 (2-14)

[0178]

[0179] 3) Calculation of cost function C

[0180]

[0181] In this formula, Y i represents the i-th label vector, A [L](i) represents the i-th output variable of the L-th layer, and m represents the number of output variables.

[0182] 4) Backpropagation calculation

[0183]

[0184]

[0185] 5) Update of weights W and threshold b

[0186]

[0187]

[0188] In this formula, L represents the number of layers of the model, and m represents the number of output variables.

[0189] The training and validation samples of the neural network are sourced from the database. After training and validation are completed, the data Q = [Q 1 , Q 2 , …, Q n , I = [I 1 , I 2 , …, I n , SOC = [SOC 1 , SOC 2 , …, SOC n , and R i = [R 1 , R 2 , …, R n recorded and calculated by the vehicle domain controller within time t are input. The data interval is a fixed value Δt. The change rate of the output SOH [SOH 1 ′, SOH 2 ′, …, SOH′ n can be obtained. Based on this change rate, the SOH after Δt time can be obtained through Equation 2-10:

[0190]

[0191] This value is returned to the vehicle domain controller to complete the update of the SOH.

[0192] 5. Update and Verification of Battery State Calculation Model

[0193] When the vehicle domain controller receives the SOH value returned by the cloud, it updates the SOH value of the state calculation model. Based on the updated SOH, it re-verifies the SOC deviation. If the deviation value is still greater than 10%, accumulate data for a period of time again, perform the fitting of the neural network algorithm again, or adjust the parameters of the neural network algorithm, such as changing the number of hidden layer nodes h and the learning rate α, until a battery state calculation model with a SOC deviation meeting the requirements is obtained.

[0194] After obtaining a battery state calculation model meeting the requirements, the database can save the current input [Q, I, SOC, R i , output [SOH′] to the cloud database, which plays the role of dynamically expanding samples and improving the fitting accuracy.

[0195] Through the above system, the charge and discharge power at different SOC states can be corrected to show its actual impact on SOH. Combining the internal resistance, SOC, and current of the battery, the neural network algorithm is used to perform more accurate online prediction and update of the battery SOH, which can improve the monitoring accuracy of the battery state and timely warn or avoid vehicle battery failures.

[0196] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the devices and methods according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0197] The embodiments of the present application also provide corresponding devices and computer-readable storage media for implementing the solutions provided by the embodiments of the present application.

[0198] Among them, the device includes a memory and a processor. The memory is used to store instructions or code, and the processor is used to execute the instructions or code so that the device executes a method for evaluating the state of a battery according to any embodiment of the present application.

[0199] In practical applications, the computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0200] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0201] The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0202] The computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0203] It should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0204] As described above, it is only a specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by this application should be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A method for evaluating the state of a storage battery, characterized in that, the method includes: Calibrating a storage battery state calculation model based on the difference situation between the calculation results of the remaining capacity percentage SOC of the storage battery, where the storage battery state calculation model is used to calculate the parameters of the storage battery, and the parameters include SOC; Calculating the parameters of the storage battery using the calibrated storage battery state calculation model, inputting the parameters into a storage battery state evaluation model based on a neural network, and calculating the state of health SOH based on the parameters using the storage battery state evaluation model; Sending the calculated SOH to the storage battery state calculation model for updating the SOH of the storage battery to determine the state evaluation result of the storage battery.

2. The method according to claim 1, characterized in that, the calibrating the storage battery state calculation model based on the difference situation between the calculation results of the remaining capacity percentage SOC of the storage battery includes: Obtaining a first SOC using a first calculation method; Obtaining a second SOC using a second calculation method; Setting a target calibration value, where the target calibration value is used to determine the accuracy of the storage battery state calculation model; When the difference between the first SOC and the second SOC is greater than the target calibration value, calibrating the storage battery state calculation model.

3. The method according to claim 2, characterized in that, the calibrating the storage battery state calculation model includes: Taking the cumulative charge and discharge ampere-hours, current, remaining capacity percentage, and internal resistance of the storage battery as input parameters of the storage battery state evaluation model; Calculating the target state of health of the storage battery using the storage battery state evaluation model based on the input parameters; Calibrating the storage battery state calculation model based on the target state of health.

4. The method according to claim 3, characterized in that, the calculating the target state of health using the storage battery state evaluation model based on the input parameters includes: Performing normalization processing on the input parameters to obtain processed input parameters; Initializing the parameters of the storage battery state evaluation model; Calculating the target state of health using the storage battery state evaluation model based on the processed input parameters.

5. The method according to claim 1, characterized in that, the calculating the state of health SOH using the storage battery state evaluation model based on the parameters includes: Collecting the parameter data of the storage battery; Taking the parameter data as samples to establish a mapping relationship between the parameter data of the storage battery and SOH; Calculating and determining the current SOH of the storage battery using the storage battery state evaluation model according to the parameter data.

6. The method according to claim 3, characterized in that, the calibrating the storage battery state calculation model based on the target state of health includes: Updating the SOH in the storage battery state calculation model based on the target state of health; Based on the updated SOH, rechecking the difference between the first SOC and the second SOC; Adjust the battery state calculation model according to the verification result until the difference is less than the target calibration value, and then complete the calibration of the battery state calculation model.

7. The method according to claim 5, wherein, establishing the mapping relationship between the parameter data of the battery and the SOH includes: Performing charge and discharge experiments on the battery. When the experiment is completed, perform a charging operation on the battery until the charge amount of the battery meets a preset threshold; Determine the remaining capacity and the initial capacity of the battery that has completed the charging operation; Determine the SOH based on the remaining capacity and the initial capacity; Establish the mapping relationship between the parameter data of the battery and the SOH.

8. An apparatus for evaluating the state of a battery, wherein, the apparatus includes: a calibration module, a calculation module, and a result obtaining module; The calibration module is configured to calibrate the battery state calculation model based on the difference between the calculation results of the remaining capacity percentage SOC of the battery. The battery state calculation model is used to calculate the parameters of the battery, and the parameters include SOC; The calculation module is configured to calculate the parameters of the battery by using the calibrated battery state calculation model, input the parameters into the battery state evaluation model based on a neural network, and calculate the health state SOH based on the parameters by using the battery state evaluation model; The result obtaining module is configured to send the calculated SOH to the battery state calculation model for updating the SOH of the battery to determine the state evaluation result of the battery.

9. An electronic device, wherein, the device includes: a processor, a memory, and a system bus; The processor and the memory are connected through the system bus; The memory is used to store one or more programs, and the one or more programs include instructions that, when executed by the processor, cause the processor to execute the method for evaluating the state of a battery according to any one of claims 1-7.

10. A vehicle, wherein, the vehicle is configured with the electronic device according to claim 9.

11. A computer-readable storage medium, wherein, an implementation program for implementing the method for evaluating the state of a battery is stored on the computer-readable storage medium. When the implementation program for implementing the method for evaluating the state of a battery is executed by a processor, the steps of the method according to any one of claims 1-7 are implemented.

Images