Charge and discharge capacity correction method and device, storage medium, and electronic device

Abstract

Disclosed in the present disclosure are a charge and discharge capacity correction method and device, a storage medium, and an electronic device. The method comprises: acquiring X*Y sets of test data generated, under Y types of temperature variation parameters, for each battery among X batteries to be tested; determining loss of full discharge and / or loss of full charge corresponding to each set of test data, concatenating the loss of full discharge and / or the loss of full charge as target values with the corresponding first key features to generate a training dataset, and fitting model parameters on the basis of on the training dataset to obtain a nonlinear prediction model; predicting, on the basis of second key features of a target battery and the nonlinear prediction model, target loss of full discharge and / or target loss of full charge corresponding to the target battery under a current state of charge and a current temperature variation condition; and using the target loss of full discharge to correct the initial discharge capacity of the target battery upon a temperature variation and / or using the target loss of full charge to correct the initial charge capacity of the target battery upon the temperature variation.

Description

Method, device, storage medium and electronic device for correcting chargeable and dischargeable capacity

[0001] The present disclosure claims priority from a Chinese patent application No. 202410413820.7 filed on April 8, 2024, and entitled "Method, device, storage medium and electronic device for correcting chargeable and dischargeable capacity", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present disclosure relates to the field of batteries, and in particular, to a method, device, storage medium and electronic device for correcting chargeable and dischargeable capacity. BACKGROUND

[0003] In the related art, the battery capacity of a battery migrated to a low temperature will have a certain capacity loss relative to the battery capacity at normal temperature. Specifically, there are two stages of performance. The first stage is the first charge and discharge stage: due to the influence of temperature change, the battery is in an unstable state, and the charge capacity loss LFC (Loss-of-Full-Charge, full charge loss, LFC for short) and the discharge capacity loss LFD (Loss-of-Full-Discharge, full discharge loss, LFD for short) are not equal. Therefore, the chargeable SOC at this time is not equal to 1-SOC (State of Charge, battery state of charge, SOC for short). The second stage: after several charge and discharge cycles at low temperature, the battery reaches a stable state. At this time, the charge and discharge capacity losses gradually converge (LFC = LFD = (usable capacity at high temperature - usable capacity at low temperature)). That is, the actual usable capacity of the battery is sensitive to the ambient temperature. The usable capacity of the battery migrated to a low temperature will have a large loss, which will cause the uncorrected battery state of charge at low temperature to be unable to accurately indicate the usable capacity of the battery. Conversely, when the temperature changes from low to high, the usable capacity will increase. Therefore, many SOC correction methods for temperature changes have been proposed.

[0004] However, most temperature change SOC correction methods focus on the second stage and ignore the correction of the first stage. The second stage requires recording the full charge or full discharge capacity to correct the SOC. Therefore, the related art cannot quickly determine the LFC and LFD of a battery with single cell differences at different temperature changes and different initial capacities in the first stage, resulting in low correction accuracy of the first chargeable and dischargeable capacity of the battery with single cell differences after temperature change. SUMMARY

[0005] Embodiments of the present disclosure provide a method and device for correcting the first-time chargeable and dischargeable capacity of a battery, a storage medium and an electronic device, to at least solve the technical problem that the first-time chargeable and dischargeable capacity of a battery with single-cell difference cannot be corrected correctly after temperature change in the related art.

[0006] According to an aspect of embodiments of the present disclosure, a method for correcting the first-time chargeable and dischargeable capacity of a battery is provided, including: obtaining X*Y groups of test data generated by test information of each of X batteries to be tested under Y types of temperature change parameters, the test data including a first key feature of the battery to be tested, the first key feature including a starting ambient temperature, a migration ambient temperature and battery parameter information, the battery parameter information including state of charge information, X and Y being positive integers; determining a full discharge loss and / or a full charge loss corresponding to each of the X*Y groups of test data, and respectively splicing the full discharge loss and / or the full charge loss as a target value with the first key feature corresponding thereto to generate a training data set, fitting parameters of a model based on the training data set to obtain a nonlinear prediction model; predicting a target full discharge loss and / or a target full charge loss of a target battery under a current state of charge and a current temperature change condition according to a second key feature of the target battery and the nonlinear prediction model; correcting the first-time dischargeable capacity of the target battery after temperature change by using the target full discharge loss, and / or correcting the first-time chargeable capacity of the target battery after temperature change by using the target full charge loss.

[0007] According to another aspect of the embodiments of the present disclosure, a device for correcting chargeable and dischargeable capacity is also provided, comprising: an acquisition module configured to acquire X*Y groups of test data generated by test information of each of X batteries to be tested under Y types of temperature variation parameters, wherein the test data comprises a first key feature of the battery to be tested, the first key feature comprises a starting ambient temperature, a migration ambient temperature and battery parameter information, the battery parameter information comprises state of charge information, and X and Y are positive integers; a determination module configured to determine a full discharge loss and / or a full charge loss corresponding to each group of test data in the X*Y groups of test data, and splice the full discharge loss and / or the full charge loss as a target value with the first key feature corresponding thereto to generate a training data set, fit parameters of a model based on the training data set to obtain a nonlinear prediction model; a prediction module configured to predict a target full discharge loss and / or a target full charge loss of a target battery under a current state of charge and a current temperature variation condition according to a second key feature of the target battery and the nonlinear prediction model; and a correction module configured to correct a first dischargeable capacity of the target battery after temperature change by using the target full discharge loss, and / or correct a first chargeable capacity of the target battery after temperature change by using the target full charge loss.

[0008] According to still another aspect of the embodiments of the present disclosure, a computer readable storage medium is also provided, and the computer readable storage medium stores a computer program, wherein the computer program is configured to execute the above-mentioned method for correcting chargeable and dischargeable capacity when running.

[0009] According to still another aspect of the embodiments of the present disclosure, an electronic device is also provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-mentioned method for correcting chargeable and dischargeable capacity through the computer program.

[0010] According to still another aspect of the embodiments of the present disclosure, a computer program product is also provided, comprising a computer program, wherein the computer program is executed by a processor to implement the steps in any of the above-mentioned embodiments of the method for correcting chargeable and dischargeable capacity. BRIEF DESCRIPTION OF DRAWINGS

[0011] FIG. 1 is a hardware structure block diagram of a smart car terminal according to a method for correcting chargeable and dischargeable capacity according to an embodiment of the present disclosure;

[0012] FIG. 2 is a flowchart of a method for correcting chargeable and dischargeable capacity according to an embodiment of the present disclosure;

[0013] FIG. 3 is a schematic diagram of LFC and LFD of different pairs of monomer batteries under different quantile regression according to an embodiment of the present disclosure;

[0014] FIG. 4 is a schematic diagram of using separate non-linear fitting algorithms in each divided interval according to an embodiment of the present disclosure;

[0015] FIG. 5 is a schematic diagram of predicting new data by linearizing different quantile regressions with LFC and LFD according to an embodiment of the present disclosure;

[0016] FIG. 6 is a structural block diagram of a capacity correction device according to an embodiment of the present disclosure;

[0017] FIG. 7 is a computer system structural block diagram of an electronic device according to an embodiment of the present disclosure;

[0018] FIG. 8 is an electronic device configured to implement the above-described capacity correction method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0019] In order to make the person skilled in the art better understand the present disclosure scheme, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, not all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present disclosure.

[0020] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] The method embodiments provided by the embodiments of the present disclosure can be executed in a smart car terminal, a mobile terminal or similar computing devices. Taking the case of running on a smart car terminal, FIG. 1 is a hardware structure block diagram of a smart car terminal of a method for correcting the chargeable and dischargeable capacity according to an embodiment of the present disclosure. As shown in FIG. 1, the smart car terminal can include one or more (only one is shown in FIG. 1) processors 102 (the processor 102 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 configured to store data. In an exemplary embodiment, the above-mentioned smart car terminal can further include a transmission device 106 configured to have a communication function and an input and output device 108. Those skilled in the art can understand that the structure shown in FIG. 1 is only schematic, which does not limit the structure of the above-mentioned smart car terminal. For example, the smart car terminal can further include more or fewer components than those shown in FIG. 1, or have a different configuration with the same function or more than the function shown in FIG. 1.

[0022] The memory 104 can be configured to store computer programs, for example, software programs of application software and modules, such as the computer program corresponding to the method for correcting the chargeable and dischargeable capacity in the embodiments of the present disclosure. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, that is, implements the above-mentioned method. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the smart car terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0023] The transmission device 106 is configured to receive or send data via a network. The specific examples of the above-mentioned network can include a wireless network provided by a communication provider of the smart car terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC) which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, RF) module configured to communicate with the Internet in a wireless manner.

[0024] In the present embodiment, a method for correcting the chargeable and dischargeable capacity is provided, which is applied to the above-mentioned smart car terminal. FIG. 2 is a flowchart of the method for correcting the chargeable and dischargeable capacity according to an embodiment of the present disclosure, which includes the following steps:

[0025] In step S202, X*Y sets of test data are generated by obtaining test information of each of the X batteries under Y types of temperature variation parameters, the test data including a first key feature of the battery to be tested, the first key feature including a starting environmental temperature, a migration environmental temperature, and battery parameter information, the battery parameter information including state of charge information, X and Y being positive integers.

[0026] It should be noted that in actual application, after the test is completed, each battery to be tested corresponds to a set of test data under each type of temperature variation parameter, and the set of test data includes a first test data for calculating full discharge loss and / or a second test data for calculating full charge loss. In the embodiment of the present disclosure, when the battery to be tested needs to calculate both full discharge loss and full charge loss under a certain type of temperature variation parameter, different experimental steps are set for the battery to be tested under the type of temperature variation parameter to obtain the first test data and the second test data. When the battery to be tested needs to calculate any one of the full discharge loss or the full charge loss under a certain type of temperature variation parameter, only the first test data or the second test data is obtained for the battery to be tested under the type of temperature variation parameter. The first test data includes a first key feature of the battery to be tested for determining the full discharge loss, and the second test data includes a first key feature of the battery to be tested for determining the full charge loss.

[0027] The state of charge information in the embodiment of the present disclosure includes information for calculating or representing the state of charge of the battery, which can be a state of charge value, a battery voltage, or the like, and further determines the chargeable and dischargeable capacity of the battery through the state of charge information. The battery parameter information can further include a battery voltage, a battery temperature, a battery model, and the like, which can be determined according to requirements.

[0028] As an optional implementation, the above battery parameter information includes an initial state of charge, a migration state of charge, a migration voltage, and a battery temperature, and X*Y sets of test data are generated by obtaining test information of each of the X batteries to be tested under Y types of temperature variation parameters through the following steps S11 to S13:

[0029] In step S11, for each battery to be tested, at least two charge and discharge cycles are performed at a test starting environmental temperature of each type of temperature variation parameter.

[0030] It can be understood that the battery test is continuously carried out, and the test starting environment temperature of a certain test may be transferred from the last temperature change parameter test, and therefore, a plurality of full charging and discharging cycles are required to make the battery in the stable second stage and avoid interference of other tests on the current test. The charging and discharging cycle operation mainly records the first capacity charged in the test starting temperature of the current battery to be tested, records the first capacity charged in the process, and discharges the capacity of the current battery to be tested at the same test starting temperature. The first capacity and the second capacity of the current battery to be tested at the same test starting temperature are compared, and if the two capacities are similar, it indicates that the current battery to be tested has reached the stable second stage, and the subsequent charging and discharging process is not affected by the possible migration process. Thus, other interference in the test process is excluded before the test, and the accuracy of the test data obtained subsequently is improved.

[0031] In step S12, the battery to be tested is charged for a first preset time or discharged for a second preset time, so that the battery to be tested reaches a preset initial state of charge, and the initial state of charge is used to determine the first chargeable capacity and / or the first dischargeable capacity of the battery to be tested.

[0032] After excluding other interference of the battery to be tested in the test condition by step S11, the battery to be tested is charged for a first preset time at the initial environment temperature to reach a preset initial state of charge, and the first chargeable capacity of the battery to be tested is determined, or discharged for a second preset time to reach a preset initial state of charge, and the first dischargeable capacity of the battery to be tested is determined.

[0033] In step S13, the battery to be tested is transferred from the test starting environment temperature to the corresponding test migration environment temperature, and the migration state of charge, the migration voltage and the battery temperature of the battery to be tested are obtained, the migration state of charge is used to determine the second chargeable capacity and / or the second dischargeable capacity of the battery to be tested.

[0034] After the battery to be tested is transferred from the test starting environment temperature to the corresponding test migration environment temperature, the initial state of charge corresponding to the test migration environment temperature needs to be determined again due to the change of temperature, and the migration state of charge of the battery to be tested is obtained. The above migration voltage is the release voltage of the battery to be tested to the outside when the migration state of charge is reached, and the battery temperature is the corresponding internal temperature of the battery to be tested at the test migration environment temperature.

[0035] Further, in order to ensure the accuracy of the migration state of charge of the battery to be tested, after the battery to be tested is migrated from the test starting ambient temperature to the corresponding test migration ambient temperature, the dynamic resting of the battery to be tested at the test migration ambient temperature is used to ensure that the overall temperature of the battery to be tested is consistent with the test migration ambient temperature, thereby improving the accuracy of the test. The resting time corresponding to the dynamic resting is dynamically changed and is inversely proportional to the temperature change of the battery to be tested, for example, when the temperature difference of the ambient temperature of the battery to be tested during the test is greater, the corresponding resting time is longer.

[0036] wherein the first chargeable capacity is the capacity charged into the battery to be tested from the preset initial state of charge when the battery to be tested is fully charged in the starting environment; the first dischargeable capacity is the capacity discharged from the battery to be tested in the preset initial state of charge when the battery to be tested is emptied in the starting environment. The second chargeable capacity is the chargeable capacity of the battery to be tested in the migration environment (i.e. the recorded charge capacity when the battery is fully charged after migration); the corresponding chargeable capacity is recorded when the battery to be tested is fully charged after being charged for a first preset time or discharged for a second preset time to the preset initial state of charge and then migrated to the test migration ambient temperature. The second dischargeable capacity is the dischargeable capacity of the battery to be tested in the migration environment (i.e. the recorded discharge capacity when the battery is emptied after migration); the corresponding dischargeable capacity is recorded when the battery to be tested is emptied after being charged for a first preset time or discharged for a second preset time to the preset initial state of charge and then migrated to the test migration ambient temperature.

[0037] Optionally, before the X*Y sets of test data are generated by acquiring the test information of each of the X batteries to be tested under the Y types of temperature variation parameters, the method further comprises: determining a test requirement corresponding to the X batteries to be tested; and determining a temperature range interval and a gradient interval corresponding to the Y types of temperature variation parameters according to the test requirement. The gradient interval is used to indicate a temperature change value between different types of temperature variation parameters. The temperature range interval can be determined according to an actual use condition. For example, the temperature range interval is 25 degrees Celsius to -30 degrees Celsius. Within the temperature range interval, the temperature range is divided according to a gradient range and a gradient interval to determine multiple types of temperature variation parameters. For example, the gradient range is 25 degrees Celsius, and the gradient interval is 1 degree. The temperature variation parameters include: 25 degrees Celsius and 0 degrees Celsius, 24 degrees Celsius and -1 degrees Celsius, 0 degrees Celsius and -25 degrees Celsius, 0 degrees Celsius and 25 degrees Celsius, and so on. The starting environmental temperature of the temperature variation parameter of 25 degrees Celsius and 0 degrees Celsius is 25 degrees Celsius, and the migration environmental temperature is 0 degrees Celsius, indicating the migration of the battery from high temperature to low temperature during operation. The starting environmental temperature of the temperature variation parameter of 0 degrees Celsius and 25 degrees Celsius is 0 degrees Celsius, and the migration environmental temperature is 25 degrees Celsius, indicating the migration of the battery from low temperature to high temperature during operation. The above is only an example and does not limit the present disclosure. In addition, the gradient interval corresponding to the temperature variation parameters can be flexibly determined. For example, the interval of the starting environmental temperature between different types of temperature variation parameters is 5 degrees Celsius, which is determined according to the actual test requirement. In order to make the data more detailed, the interval of the starting environmental temperature between different types of temperature variation parameters can be adjusted to 1 degree Celsius, that is, the corresponding test data is determined once every 1 degree Celsius.

[0038] It can be understood that the gradient interval is a temperature difference between the starting environmental temperature or the migration environmental temperature in different temperature variation parameters when the same battery to be tested is tested using different types of temperature variation parameters. The gradient range is the change in temperature when the starting environmental temperature changes to the migration environmental temperature. That is, the gradient range is equal to the temperature difference between the starting environmental temperature and the migration environmental temperature in each type of temperature variation parameter.

[0039] In step S204, the full discharge loss and / or the full charge loss corresponding to each set of test data in the X*Y sets of test data are determined, and the full discharge loss and / or the full charge loss are respectively spliced with the first key feature corresponding thereto to generate a training data set. The parameters of a model are fitted based on the training data set to obtain a nonlinear prediction model.

[0040] In brief, the chargeable capacity and the dischargeable capacity at the starting ambient temperature and the chargeable capacity and the dischargeable capacity at the migration ambient temperature are related to the full charge loss LFC and the full discharge loss LFD, although there are slight differences between different battery individuals, there is an obvious nonlinear change characteristic between the chargeable capacity, the dischargeable capacity and the full charge loss LFC and the full discharge loss LFD; therefore, by collecting the corresponding chargeable capacity, the dischargeable capacity, the full charge loss LFC and the full discharge loss LFD of different to-be-tested batteries, a model can be further fitted to obtain a nonlinear prediction model that can support the prediction of the full charge loss LFC and the full discharge loss LFD.

[0041] Alternatively, the full charge loss is equal to the difference between the chargeable capacity of the battery at the starting ambient temperature and the chargeable capacity at the migration ambient temperature, and the full discharge loss is equal to the difference between the dischargeable capacity of the battery at the starting ambient temperature and the dischargeable capacity at the migration ambient temperature. Taking the test data in the above embodiment as an example, the full charge loss = the first chargeable capacity - the second chargeable capacity, and the full discharge loss = the first dischargeable capacity - the second dischargeable capacity.

[0042] For example, according to the chargeable capacity of the to-be-tested battery A at the first starting ambient temperature and the chargeable capacity of the to-be-tested battery A at the first migration ambient temperature, the full charge loss corresponding to the to-be-tested battery A is determined, and according to the dischargeable capacity of the to-be-tested battery A at the first starting ambient temperature and the dischargeable capacity of the to-be-tested battery A at the first migration ambient temperature, the full discharge loss corresponding to the to-be-tested battery A is determined.

[0043] After the full charge loss and the full discharge loss corresponding to the battery A to be tested are determined, in order to better fit the parameters of the model in each interval according to the above data, the nonlinear prediction model of the current battery A to be tested is determined. At this time, the first key feature of the battery A to be tested included in the test data needs to be spliced with the determined full charge loss and / or full discharge loss, realizing the mutual correspondence between temperature change, battery parameter information and full charge loss and / or full discharge loss. It should be noted that in the splicing process, the key feature of the full discharge loss needs to be spliced with the full discharge loss, and the key feature of the full charge loss needs to be spliced with the full charge loss. For example, the key features corresponding to the full discharge loss of the battery A to be tested are the first initial environmental temperature, the first migration environmental temperature, the dischargeable capacity of the battery A to be tested at the first initial environmental temperature, the dischargeable capacity of the battery A to be tested at the first migration environmental temperature, the end voltage corresponding to the first initial environmental temperature, the start voltage corresponding to the first migration environmental temperature, the negative electrode initial temperature corresponding to the first migration environmental temperature, etc. The key features corresponding to the full charge loss of the battery A to be tested are the first initial environmental temperature, the first migration environmental temperature, the chargeable capacity of the battery A to be tested at the first initial environmental temperature, the chargeable capacity of the battery A to be tested at the first migration environmental temperature, the end voltage corresponding to the first initial environmental temperature, the start voltage corresponding to the first migration environmental temperature, the negative electrode initial temperature corresponding to the first migration environmental temperature, etc.

[0044] Further, after completing the splicing of the full discharge loss and / or full charge loss as the target value with the first key feature corresponding thereto, a plurality of training data will be obtained, and then a training data set is formed. In addition, the battery parameter information in the first key feature includes the initial state of charge, the initial voltage, the migration state of charge, the migration voltage and the battery temperature. The initial state of charge is used to determine the first chargeable capacity and / or the first dischargeable capacity of the battery to be tested. The migration state of charge is used to determine the second chargeable capacity and / or the second dischargeable capacity of the battery to be tested. The initial voltage is used to indicate the end voltage of the battery at the initial environmental temperature, i.e. the output voltage corresponding to the battery when the charging or discharging state is completed at the initial environmental temperature. The migration voltage is used to indicate the start voltage of the battery at the migration environmental temperature, i.e. the output voltage when the battery starts to charge or discharge at the migration environmental temperature.

[0045] Further, when the temperature change from the starting ambient temperature adjustment to the migration ambient temperature is from high temperature to low temperature, the actual working chargeable and dischargeable capacity of the battery to be tested increases due to the temperature rise, and when the temperature change from the starting ambient temperature adjustment to the migration ambient temperature is from low temperature to high temperature, the actual working chargeable and dischargeable capacity of the battery to be tested decreases due to the temperature drop. Therefore, during the test, the actual chargeable and dischargeable capacity of the battery to be tested at different temperatures needs to be collected according to the temperature change of different temperature change parameters, so as to more accurately determine the charge capacity reduction / increase and discharge capacity reduction / increase of the battery to be tested under different conditions. For the convenience of description, the change of the chargeable and dischargeable capacity is summarized as loss in the following text, and the positive and negative of the loss value indicates whether the chargeable and dischargeable capacity decreases or increases. When the loss value is positive, it indicates that the chargeable and dischargeable capacity decreases, and when the loss value is negative, it indicates that the chargeable and dischargeable capacity increases.

[0046] For example, the working condition of the temperature drop operation is described by taking the test starting ambient temperature of 25°C and the test migration ambient temperature of 0°C as an example, and the working condition of the temperature rise operation is described by taking the test starting ambient temperature of 0°C and the test migration ambient temperature of 25°C as an example. Alternatively, as shown below:

[0047] Under the condition that the test starting ambient temperature is 25°C, the first chargeable capacity of the battery to be tested is 75% of the rated capacity. If the battery to be tested is warmed to 0°C, the charging loss is positive due to the temperature drop, and the chargeable capacity of the battery to be tested will be less than 75% of the rated capacity. Similarly, the first dischargeable capacity of the battery to be tested after being warmed to 0°C will be less than 25% of the rated capacity.

[0048] As an alternative example, under the condition that the test starting ambient temperature is 0°C, the second chargeable capacity of the battery to be tested is 75% of the available capacity at 0°C. If the battery to be tested is warmed to 25°C, the charging loss is negative due to the temperature rise, and the chargeable capacity of the battery to be tested will be greater than 75% of the available capacity at 0°C. Similarly, the second dischargeable capacity of the battery to be tested after being warmed to 25°C will be greater than 25% of the available capacity at 0°C.

[0049] It should be noted that in actual application, after being warmed to the test migration ambient temperature, the battery to be tested is in the first stage of inconsistent charge and discharge capacity loss during the first charge and discharge, so the real capacity of the battery to be tested at the test migration ambient temperature cannot be obtained, and therefore the state of charge SOC of the battery to be tested at the test migration ambient temperature can only be defined according to the rated capacity.

[0050] Optionally, the battery temperature includes a negative electrode initial temperature, which is used to indicate the initial temperature of the negative electrode of the battery to be tested when the battery is in a discharge or charging state in a migration environment temperature, and is mainly used to identify the difference between different single batteries. It should be noted that because different single batteries migrate to the same temperature environment, there is a difference in the negative electrode temperature due to the difference in internal reaction. The above is only an example and does not limit all the above key features.

[0051] For ease of understanding, taking the battery A to be tested as an example, the training data set determined is as follows in Table 1:

[0052] Table 1

[0053] Optionally, after the training data set is determined, the model parameters can be fitted through the training data set, and the parameters of the model in each interval are fitted. The training data set is classified according to the full charge loss and the full discharge loss, and then the parameters of the model are fitted according to the first training data set corresponding to the classified full charge loss to generate a first nonlinear prediction model related only to the full charge loss; the parameters of the model are fitted according to the second training data set corresponding to the classified full discharge loss to generate a second nonlinear prediction model related only to the full discharge loss. In the subsequent process of using the nonlinear prediction model, when the full charge loss needs to be predicted, the first nonlinear prediction model is used, and when the full discharge loss needs to be predicted, the second nonlinear prediction model is used.

[0054] As an optional embodiment, after the training data set is determined, the parameters of the model are fitted based on the training data set to obtain the nonlinear prediction model. Since the first key feature in the training data set contains multi-dimensional information, high-dimensional information is not easy to visualize, therefore, the multi-dimensional feature information corresponding to the first key feature in the training data set is abstracted into one dimension as the x-axis. LFC or LFD is the target value, so it is used as the y-axis. Understand: y = f(x), f is the prediction model of the fitting parameter. Further, a plurality of quantiles are set according to the target values corresponding to the full discharge loss and the full charge loss, wherein the change range of the full discharge loss and the full charge loss between different quantiles, so as to divide the target values into a plurality of distribution intervals according to the quantiles, and then fit separate regression parameters in each interval. Linear regression of multiple distribution intervals is used to approximate the overall nonlinear regression. Finally, the parameters of the model are fitted based on the training data set to obtain the nonlinear prediction model, and the nonlinear prediction model at least includes a first prediction sub-model corresponding to the full discharge loss and / or a second prediction sub-model corresponding to the full charge loss.

[0055] As an optional embodiment, FIG. 3 is a schematic diagram of LFC and / or LFD corresponding to different batteries to be tested in different quantile regression according to an embodiment of the present disclosure; FIG. 4 is a schematic diagram of using a separate nonlinear fitting algorithm in each divided interval according to an embodiment of the present disclosure; that is, after dividing the target value (equivalent to the full discharge loss and / or full charge loss in the above embodiment) into multiple intervals according to quantiles, in order to enhance the nonlinear fitting capability of the algorithm proposed in the present disclosure, a separate nonlinear fitting algorithm can also be used in each divided interval, wherein the data points in FIGS. 3 and 4 are used to correspond to the target value and the first key feature of each pair of completed splicing in the training data set, and the curves in FIGS. 3 and 4 are used to indicate the mapping relationship between the first key feature and LFC or LFD. It should be noted that the curves in FIGS. 3 and 4 above can be curves corresponding to the full charge loss and / or curves corresponding to the full discharge loss, which can be flexibly controlled according to actual display requirements, and the present disclosure does not make too many limitations. It can be understood that the LFC and LFD of different batteries, different battery starting capacities at different temperatures, and different battery target temperatures are fitted by the quantile regression algorithm; thereby obtaining a nonlinear prediction model. Wherein the x-axis represents the multi-dimensional feature information corresponding to the first key feature in the training data set abstracted into one dimension, and the y-axis is the target value LFC or LFD.

[0056] Step S206, according to the second key feature of the target battery and the nonlinear prediction model, the target full discharge loss and / or the target full charge loss corresponding to the target battery under the current state of charge and the current temperature condition are predicted;

[0057] Optionally, the second key feature can include: a target starting environmental temperature of the target battery, and a target dischargeable capacity or a target chargeable capacity corresponding to the target battery at the target starting environmental temperature; a target migration environmental temperature of the target battery, and a target dischargeable capacity or a target chargeable capacity corresponding to the target battery at the target migration environmental temperature, etc. It can be understood that after fitting the parameters of the regression model in each quantile interval according to the X*Y test data and further determining the corresponding nonlinear prediction model, for the target battery with known second key feature, the nonlinear prediction model can be used for prediction, so as to determine the estimated target value in the nonlinear prediction model that meets the second key feature, and achieve the purpose of determining the full discharge loss and / or the full charge loss corresponding to the target battery according to the second key feature.

[0058] It can be understood that the full charge and full discharge LFD and LFC calculation in the related art does not exclude the difference of single cells. Due to the error of production and assembly, there is inconsistency between different single cells, which leads to a certain difference between different single cells under the same experimental conditions in practice. And the related test results show that the same single cell battery, the same initial temperature, and different initial capacities will obtain very inconsistent LFD and LFC, and the LFD and LFC obtained under the conditions of full charge and full discharge do not have generalization. In order to improve the determination of the slight difference between different single cells LFD and LFC, and to improve the subsequent prediction efficiency; when establishing the above prediction model in practice, the method of machine learning is used to predict the LFD and LFC corresponding to different initial capacities, different migration temperatures, and different single cell batteries, and then correct the first charge and discharge SOC of the battery after temperature change. And in order to adapt to the offline, low-power embedded execution environment, the present disclosure also proposes a low-complexity nonlinear regression algorithm for the above purpose. Therefore, the low temperature and capacity loss in the related art are extended to temperature change and capacity change, so that the first chargeable and dischargeable SOC after any temperature change can be accurately corrected.

[0059] Optionally, the above nonlinear regression algorithm is composed of quantile regression and KNN algorithm. It should be noted that quantile regression can fit the nonlinear characteristics of data and adapt to offline, low-power embedded execution environment, but it cannot predict new samples. And the generalization of KNN algorithm regression is poor, and it cannot well fit the nonlinear characteristics. Optionally, the nonlinear prediction model is obtained by fitting the multiple sets of test data obtained by testing through the quantile regression algorithm. On the basis of obtaining the nonlinear prediction model, the KNN algorithm is used to select the target key feature with similar feature distance in the multiple key features of the fitting data set with reference to the second key feature of the target battery, and the target sub-prediction model with smaller data error is selected from the multiple sub-prediction models contained in the nonlinear prediction model according to the target key feature. Then, the second key feature is input into the target sub-prediction model with smaller data error to predict the results of multiple LFC and LFD, and multiple LFC or multiple LFD are processed respectively, such as: mean processing, weighted sum and mean processing, to obtain the final LFC or LFD.

[0060] Optionally, when the computing power of the execution environment is strong, the above nonlinear regression algorithm (i.e. including quantile regression + KNN algorithm for data processing) as a whole can be replaced by using a more complex model.

[0061] In summary, by proposing a low-complexity nonlinear regression algorithm, the nonlinear characteristics of the battery temperature change process are fitted with low computational cost. While shortening the overall prediction time, the prediction accuracy is improved, so that the subsequent target full discharge loss or target full charge loss can be predicted to correct the first chargeable and dischargeable capacity of the battery after temperature change, that is, based on the determination of the first chargeable and dischargeable capacity of the target battery, the actual chargeable and dischargeable state of charge of the first chargeable and dischargeable battery corresponding to the target battery is determined by the ratio of the first chargeable and dischargeable capacity to the rated capacity corresponding to the target battery.

[0062] Step S208, using the target full discharge loss to correct the first dischargeable capacity of the target battery after temperature change, and / or using the target full charge loss to correct the first chargeable capacity of the target battery after temperature change. That is, in the case where the target full discharge loss and / or the target full charge loss corresponding to the target battery are predicted according to the second key feature and the nonlinear prediction model, the first dischargeable capacity of the target battery after temperature change can be corrected according to the target full discharge loss, and the first chargeable capacity of the target battery after temperature change can also be corrected according to the target full charge loss; which will be described in detail below.

[0063] As an optional implementation, the first dischargeable capacity of the target battery after temperature change is corrected using the target full discharge loss, including: correcting the first dischargeable capacity of the target battery after temperature change according to the target dischargeable capacity, the target full discharge loss, and the rated capacity of the target battery.

[0064] Optionally, the first dischargeable capacity of the target battery after temperature change can be corrected by the following formula one:

[0065] Formula one, SOC T2-Dchg = [SOC T1 *C T1 -LFD(T1,T2)] / C rate , wherein SOC T2-Dchg is the state of charge corresponding to the target battery discharge state at T2, that is, the first dischargeable capacity of the target battery after temperature change [SOC T1 *C T1 -LFD(T1,T2)] and the rated capacity C rate of the target battery, and the quotient value between them is the dischargeable state of charge corresponding to the target battery at T2; SOC T1 *C T1 is used to indicate the target dischargeable capacity at T1, LFD(T1,T2) is the target full discharge loss predicted according to the second key feature and the nonlinear prediction model of the target battery, and C rateThe rated capacity of the target battery, T1 is the target starting ambient temperature corresponding to the target battery, and T2 is the target battery corresponding to the migration ambient temperature after the temperature change.

[0066] It should be noted that the SOC of the target battery corresponding to T1 in the formula T1 is defined based on the actual capacity, that is, according to the total chargeable / dischargeable capacity corresponding to the target battery in the second stage when the charge / discharge capacity loss gradually converges. After the temperature change to T2, the target battery is in the first stage when the charge / discharge capacity loss is inconsistent in the first charge / discharge, and the real total chargeable / dischargeable capacity of the target battery corresponding to T2 cannot be effectively obtained. In order to facilitate calculation, the battery state of charge corresponding to the chargeable / dischargeable capacity of the target battery in the first stage under T2 is defined by setting the rated capacity in the formula, so as to ensure the correct correction of the chargeable capacity corresponding to the target battery.

[0067] As an optional implementation, the first chargeable capacity of the target battery after the temperature change is corrected by using the target full charge loss, including: correcting the first chargeable capacity of the target battery corresponding to the temperature change according to the target chargeable capacity, the target full charge loss and the rated capacity; wherein the temperature change is used to represent that the operating environment temperature of the target battery is changed from the target starting ambient temperature to the target migration ambient temperature.

[0068] Optionally, the first chargeable capacity of the target battery after the temperature change can be corrected by the following formula two:

[0069] Formula two, SOC T2-Chg = [(1-SOC T1 )*C T1 -LFC(T1,T2)] / C rate , wherein SOC T2-chg is the state of charge corresponding to the state of charge of the target battery under T2, that is, the first chargeable capacity of the target battery after the temperature change [(1-SOC T1 )*C T1 -LFC(T1,T2)] and the rated capacity C rate of the target battery are the quotient value of the target battery corresponding to the chargeable state of charge under T2; (1-SOC T1 )*C T1 indicates the target chargeable capacity under T1, LFC(T1,T2) is the target full charge loss predicted according to the second key feature and the nonlinear prediction model of the target battery, C rate is the rated capacity of the target battery, T1 is the starting ambient temperature corresponding to the target battery, and T2 is the migration ambient temperature corresponding to the target battery after the temperature change.

[0070] It should be noted that when T1 is less than or equal to T2, the above LFD and LFC are negative numbers, and when T1>T2, then the opposite, LFD and LFC are positive numbers. Further, by using the predicted LFC and LFD to correct the first chargeable and dischargeable SOC after temperature change, the accuracy of the first chargeable and dischargeable SOC display under different temperature changes can be effectively guaranteed.

[0071] Through the above steps, X*Y sets of test data are generated by acquiring the test information of each of the X batteries to be tested under Y types of temperature change parameters; the full discharge loss and / or the full charge loss corresponding to each set of test data in the X*Y sets of test data is determined, and the full discharge loss and / or the full charge loss is respectively spliced with the first key feature corresponding thereto as a target value to generate a training data set. The parameters of the model are fitted based on the training data set to obtain a nonlinear prediction model. Then, based on the nonlinear prediction model, the target full discharge loss and / or the target full charge loss corresponding to the target battery under the current state of charge and the current temperature change condition can be predicted according to the second key feature of the target battery and the nonlinear prediction model, and the first dischargeable capacity of the target battery after the temperature change is corrected by using the target full discharge loss, and / or the first chargeable capacity of the target battery after the temperature change is corrected by using the target full charge loss. In this way, the offline correction of the dischargeable capacity and the chargeable capacity of the target battery under temperature change is accurately realized. The above correction method of the chargeable and dischargeable capacity solves the technical problem in the related art that the LFC and LFD corresponding to the battery with single cell difference under different temperature change states and different initial capacities cannot be quickly determined, so that the state of charge of the battery after the first charge cannot be accurately corrected. In the case where the second key feature corresponding to the target battery is acquired, the target battery is predicted according to the established nonlinear prediction model, and the accuracy of the capacity correction of the target battery after the temperature change is ensured.

[0072] It should be noted that in actual application, the first stage correction is more urgent and difficult, so there are fewer related methods. The second stage correction is relatively simple, and there are more related technologies. However, full charge and full discharge operations are not supported in actual working conditions (the second stage correction method is a commonly used strategy), so the accuracy of the correction is low when correcting the state of charge. The above correction method proposed by the present disclosure acquires the test data corresponding to the battery to be tested under different types of test parameters, fits the model, quickly determines the corresponding LFC and LFD by using the similarity between the new battery and the battery to be tested, and because the data of the fitted model includes test data of multiple batteries with single cell difference under different temperature change states and different initial capacities, the state of charge of the battery that is not fully charged and not fully discharged can also be corrected correctly, and the correction accuracy of the chargeable and dischargeable state of the battery after the first charge and discharge under temperature change is improved.

[0073] Optionally, the above-mentioned embodiments can also be used to correct the battery power state (SOP) after the battery temperature changes. The specific method is the same as the above-mentioned capacity correction process, and only the corresponding fitting model parameters are replaced.

[0074] As an optional embodiment, in order to further improve the accuracy of the prediction, before predicting the target full discharge loss and / or the target full charge loss of the target battery under the current state of charge and the current temperature change condition according to the second key feature of the target battery and the nonlinear prediction model, the following steps need to be performed, including:

[0075] Step S302, calculate the feature distance between the second key feature and the first key feature in the training data set, and obtain X*Y feature distances, wherein the nonlinear prediction model includes M sub-prediction models, and M is a positive integer;

[0076] Step S304, select the first A target feature distances closest to the feature distance from the X*Y feature distances, and determine the target key feature corresponding to each of the first A target feature distances, and obtain A target key features, wherein A is a positive integer.

[0077] That is, according to the feature distance, the target key feature with high correlation with the second key feature is determined from the plurality of first key features included in the training data set. Since the target key feature is information in the training data set, and the nonlinear prediction model is determined based on the fitting model parameters of the training data set; therefore, the determination of the full discharge loss and / or the full charge loss corresponding to the second key feature can be realized by borrowing the nonlinear prediction model corresponding to the target key feature.

[0078] For example, FIG. 5 is a schematic diagram of predicting new data by LFC and / or LFD in different quantile regression lines according to an embodiment of the present disclosure; according to the K-Nearest Neighbor (KNN) algorithm, from the X*Y key features corresponding to the data points on the nonlinear prediction model, A target data points are selected whose spatial proximity distance to the second key feature (new data) satisfies the preset distance condition, that is, from the X*Y data points, A target key features closest to the second key feature are selected, as indicated by the arrow in FIG. 1; wherein the x-axis represents the multi-dimensional feature information corresponding to the first key feature in the training data set, and the y-axis is the target value LFC or LFD.

[0079] Further, on the basis of completing the above steps S302 and S304, according to the second key features of the target battery and the nonlinear prediction model, the target full discharge loss of the target battery under the current state of charge and the current temperature change condition is predicted, and the following steps S306-S310 are executed, which are optional:

[0080] Step S306, determine the actual full discharge loss corresponding to each target key feature, and perform the following operation for each feature in the A target key features: input each feature in the A target key features into the M sub-prediction models to obtain M prediction values, wherein the prediction value is the estimated full discharge loss determined by the sub-prediction model.

[0081] That is, according to the prediction type corresponding to the target battery to be predicted, in the case of predicting the target full discharge loss of the target battery under the current state of charge and the current temperature change condition, the M sub-prediction models with the target value of full discharge loss can be determined from the established nonlinear prediction model, and then the A target key features with strong correlation with the second key feature are input into the M sub-prediction models respectively to obtain M prediction values.

[0082] Step S308, for each target key feature, determine the error between each prediction value and the actual full discharge loss to obtain M prediction errors, and select B first type sub-prediction models that meet the preset error condition from the M sub-prediction models according to the M prediction errors; wherein B is a positive integer.

[0083] In short, in the case of obtaining M prediction values, by comparing the prediction error between each estimated full discharge loss and the actual full discharge loss, B sub-prediction models with smaller corresponding error of the prediction error of the target key feature can be selected from the M prediction sub-models according to the prediction error, and A target key features and B sub-prediction models are combined correspondingly, so as to determine the multiple sub-prediction models finally used for predicting the target full discharge loss of the target battery under the state of charge of the battery. It should be noted that the above-mentioned preset error condition can be determined according to the actual use condition, such as setting the error range interval, and the present disclosure does not make too many limitations.

[0084] Step S310, aggregate the B first type sub-prediction models corresponding to the A target key features respectively to obtain A*B first type sub-prediction models for predicting the target full discharge loss of the target battery under the state of charge of the battery (equivalent to the curve indicated by the No. 2 arrow), so as to realize the prediction of the target full discharge loss of the target battery under the state of charge of the battery.

[0085] By the above steps, the suitable sub-prediction model is determined from the nonlinear prediction model according to the type of the target battery to be predicted, so that the prediction of the target full discharge loss can be performed only according to the second key feature of the target battery, and the dischargeable capacity of the target battery is corrected according to the predicted target full discharge loss.

[0086] The prediction of the target full discharge loss is similar to the prediction of the target full charge loss, and some details will not be repeated. Optionally, based on the completion of the above steps S302 and S304, according to the prediction type corresponding to the target battery to be predicted, if the prediction type is that the target full charge loss corresponding to the target battery under the current state of charge and the current temperature change condition needs to be predicted, according to the second key feature of the target battery and the nonlinear prediction model, the target full charge loss corresponding to the target battery under the current state of charge and the current temperature change condition needs to perform the following steps S312-S320, which are optional:

[0087] Step S312, determine the actual full charge loss corresponding to each target key feature, for each feature in the A target key features, the following operations are performed: each feature in the A target key features is input into the M sub-prediction models to obtain M prediction values, wherein the prediction value is the estimated full charge loss determined by the sub-prediction model;

[0088] Step S314, for each target key feature, determine the error between each prediction value and the actual full charge loss, obtain M prediction errors, and select B first type sub-prediction models that meet the preset error condition from the M sub-prediction models according to the M prediction errors; wherein B is a positive integer;

[0089] Step S318, aggregate the B first type sub-prediction models corresponding to the A target key features respectively to obtain A*B first type sub-prediction models;

[0090] Step S320, use the A*B first type sub-prediction models to predict the target full discharge loss corresponding to the target battery under the current state of charge and the current temperature change condition.

[0091] It should be noted that the above is only to predict the full discharge loss. In actual application, the above prediction process can also be used for prediction of the full charge loss. Optionally, to ensure the accuracy of the prediction, the mean of the A*B prediction results can also be taken as the final prediction result.

[0092] It should be noted that, when predicting the target full charge loss and the target full discharge loss of the target battery, the determination of the target full charge loss and the target full discharge loss can be performed respectively, or the determination of the target full charge loss and the target full discharge loss can be performed simultaneously, which can be flexibly set according to the actual computing power of the system or computer in which the prediction model is located, and the present disclosure does not make too many limitations on this.

[0093] Optionally, in order to further improve the accuracy of the prediction, weights can also be set for the A*B prediction results according to the prediction size, the A*B prediction results are weighted and summed according to the weights, and the mean value after the weighted summation is determined. Further, if the cloud service building and data transmission can be realized, the test data determined can be uploaded to the cloud for processing, and after the second key feature related to the battery state of charge of the target battery is obtained, the second key feature can be transmitted to the cloud through data transmission, and the processing result of the second key feature by the cloud is returned to the target battery, so as to correct the battery state of charge of the target battery.

[0094] As an optional implementation, in order to reduce the occurrence of the same sub-prediction model in the A*B sub-prediction models, the A*B sub-prediction models can also be de-duplicated to obtain N sub-prediction models; the second key feature is input into the N sub-prediction models to generate N prediction results; the target full discharge loss and the target full charge loss of the target battery at the battery state of charge are determined based on the N prediction results; wherein N is a positive integer less than M. It can be understood that the value of A*B is not necessarily ≤M, since there can be duplication in the A*B sub-prediction models, by de-duplication, the number of models input in the subsequent prediction process is less than M, thereby reducing the unnecessary calculation amount in the subsequent process and improving the prediction efficiency.

[0095] As an optional implementation, before X*Y groups of test data are generated by obtaining the test information of each battery in X batteries to be tested under Y types of variable temperature parameters, if it is necessary to fit the test data of multiple batteries to be tested, the test data of multiple batteries to be tested can be clustered by using a clustering algorithm to cluster the test data of the same type of batteries to be tested. When fitting the nonlinear prediction model, the test data of the same type of batteries can be selected for fitting, that is, the X batteries to be tested are batteries of the same type. In this way, the fitting accuracy can be improved.

[0096] It can be understood that when there are multiple different types of batteries, in order to ensure the accuracy of subsequent prediction, the data corresponding to the same type of battery can be summarized by using a clustering algorithm to form a data cluster, and then a separate model fitting is performed on each data cluster to form a nonlinear prediction model belonging to a type of battery, so as to ensure the accuracy of the prediction of the same type of target battery in the future.

[0097] That is, when there are multiple battery types of the battery to be tested, clustering screening needs to be performed according to the battery types, so as to determine the nonlinear prediction model corresponding to different battery types, and realize the prediction of different battery types. Optionally, the test data included in any one data cluster after the screening is completed is the test data of the above-mentioned training data set.

[0098] It should be noted that the above-mentioned intelligent automobile terminal can also be replaced by other mobile terminals or cloud servers that can perform the same functions as the intelligent automobile terminal, and the present disclosure does not make too many limitations.

[0099] Optionally, the full charge loss and / or the full discharge loss of each battery to be tested under different temperature change parameters can be calculated by the following formula: when the temperature changes, the full charge loss LFC = the chargeable capacity under the migrated environmental temperature - the chargeable capacity under the starting environmental temperature, and the full discharge loss LFD = the dischargeable capacity under the migrated environmental temperature - the dischargeable capacity under the starting environmental temperature.

[0100] In summary, through the above-mentioned embodiments, by accumulating the test data of the battery to be tested under different temperature change starting capacities, the nonlinear prediction model corresponding to the battery to be tested is generated, and then after the second key feature of the target battery is obtained, the spatial proximity distance between the data point corresponding to the second key feature on the nonlinear prediction model and the determined data point can be compared, so as to determine the target curve corresponding to the data point corresponding to the second key feature in the nonlinear prediction model according to the spatial proximity distance, and the target full charge loss and the target full discharge loss corresponding to the current capacity of the target battery are predicted according to the variation trend of the full charge loss LFC and the full discharge loss LFD on the target curve under different starting capacities, thereby realizing the correction of the first chargeable and dischargeable capacity of the battery after the temperature change. By using the above-mentioned correction method of the chargeable and dischargeable capacity, the technical problem that the LFC and the LFD corresponding to the battery with single cell difference under different temperature changes and different starting capacities cannot be quickly determined in the related art is solved, so that the first chargeable and dischargeable capacity of the battery after the temperature change cannot be accurately corrected, and the effect of more accurately determining the first chargeable and dischargeable capacity of the target battery under different temperature changes is achieved.

[0101] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software on a general hardware platform as necessary, and of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present disclosure can be embodied in the form of a software product in essence or in the form of a part of the prior art that contributes to the present disclosure. The computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing an end device (which can be a mobile phone, a computer, a server, or a network device) to execute the method of each embodiment of the present disclosure.

[0102] Fig. 6 is a structural block diagram of a capacity correction device according to an embodiment of the present disclosure; as shown in Fig. 6, comprising:

[0103] The acquisition module 72 is configured to acquire X*Y groups of test data generated by test information of each of the X batteries to be tested under Y types of temperature variation parameters, the test data including a first key feature of the battery to be tested, the first key feature including a starting environmental temperature, a migration environmental temperature, and battery parameter information, the battery parameter information including state of charge information, X and Y being positive integers;

[0104] The determination module 74 is configured to determine a full discharge loss and / or a full charge loss corresponding to each group of test data in the X*Y groups of test data, and splice the full discharge loss and / or the full charge loss as a target value with the first key feature corresponding thereto, to generate a training data set, fit parameters of a model based on the training data set, and obtain a nonlinear prediction model;

[0105] The prediction module 76 is configured to predict a target full discharge loss and / or a target full charge loss corresponding to a target battery under a current state of charge and a current temperature variation condition according to a second key feature of the target battery and the nonlinear prediction model;

[0106] The correction module 78 is configured to correct a first dischargeable capacity of the target battery after a temperature change by using the target full discharge loss, and / or correct a first chargeable capacity of the target battery after a temperature change by using the target full charge loss.

[0107] By the above device, X*Y sets of test data are generated by acquiring test information of each of the X batteries to be tested under Y types of variable temperature parameters; and full discharge loss and / or full charge loss corresponding to each set of test data in the X*Y sets of test data are determined, and the full discharge loss and / or full charge loss are respectively taken as target values, and are spliced with the first key features corresponding thereto to generate a training data set, parameters of a model are fitted based on the training data set, and a nonlinear prediction model is obtained; then, based on the nonlinear prediction model, the target full discharge loss and / or target full charge loss corresponding to the target battery under the current state of charge and the current variable temperature condition can be predicted according to the second key feature of the target battery and the nonlinear prediction model, and the first dischargeable capacity of the target battery after temperature change is corrected by using the target full discharge loss, and / or the first chargeable capacity of the target battery after temperature change is corrected by using the target full charge loss, so that the dischargeable capacity and the chargeable capacity of the target battery can be accurately corrected offline under the variable temperature condition. By using the above dischargeable and chargeable capacity correction method, the technical problem that the LFC and LFD corresponding to the battery with single cell difference under different variable temperature states and different initial capacities cannot be quickly determined in the related art is solved, so that the dischargeable and chargeable capacity of the battery at the first time of charging and discharging after temperature change cannot be correctly corrected.

[0108] In one example embodiment, the above device further includes a distance module configured to, before predicting the target full discharge loss and / or target full charge loss corresponding to the target battery under the current state of charge and the current variable temperature condition according to the second key feature of the target battery and the nonlinear prediction model, calculate a feature distance between the second key feature and the first key feature in the training data set to obtain X*Y feature distances, wherein the nonlinear prediction model includes M sub-prediction models, and M is a positive integer; select the first A target feature distances closest to the feature distances from the X*Y feature distances, and determine a target key feature corresponding to each of the first A target feature distances to obtain A target key features, wherein A is a positive integer.

[0109] In the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined target, and can be implemented entirely or partially by using software, hardware (such as a processing circuit or a memory) or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the functions of the module or unit.

[0110] It should be noted that, for the foregoing method embodiments, for the purpose of simple description, they are all described as a series of action combinations, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application. The above modules can be realized by software or hardware, and for the latter, the realization can be realized by the following ways, but is not limited to: the above modules are located in the same processor; or the above modules are located in different processors in any combination.

[0111] In one example embodiment, a flash memory is also provided, which is configured to execute the steps in any of the above method embodiments.

[0112] Embodiments of the present disclosure also provide a computer readable storage medium, which stores computer programs / instructions, wherein the computer programs / instructions are configured to execute the steps in any of the above method embodiments when running.

[0113] In one example embodiment, the above computer readable storage medium can include, but is not limited to: a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various computer program storage media. The above embodiment numbers of the present disclosure are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0114] FIG. 7 schematically illustrates a computer system structural diagram of an electronic device configured to implement embodiments of the present disclosure. It is noted that the computer system 700 of the electronic device illustrated in FIG. 7 is merely an example and should not impose any limitation on the functions and usage range of embodiments of the present disclosure. As shown in FIG. 7, the computer system 700 includes a central processing unit 701 (CPU), which can perform various appropriate actions and processes according to programs stored in a read-only memory 702 (ROM) or loaded from a storage section 708 into a random access memory 703 (RAM). In the random access memory 703, various programs and data required for system operation are also stored. The central processing unit 701, the read-only memory 702, and the random access memory 703 are connected to each other through a bus 704. An input / output interface 705 (I / O interface) is also connected to the bus 704.

[0115] The following components are connected to the input / output interface 705: an input section 706 including a keyboard, a mouse, and the like; an output section 707 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage section 708 including a hard disk, and the like; and a communication section 709 including a network interface card such as a local area network card, a modem, and the like. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output interface 705 as necessary. A removable media 711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is attached to the drive 710 as necessary, so that a computer program read therefrom is installed in the storage section 708 as necessary.

[0116] In particular, according to embodiments of the present disclosure, the processes described in each of the method flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product including a computer program carried on a computer-readable medium, the computer program containing program codes configured to perform the methods illustrated in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 709, and / or installed from the removable media 711. When the computer program is executed by the central processing unit 701, various functions defined in the system of the present disclosure are performed.

[0117] According to a further aspect of the embodiments of the present disclosure, an electronic device configured to implement the above-mentioned method for correcting the chargeable and dischargeable capacity is further provided. The electronic device of the present embodiment is shown in FIG. 8, which includes a memory 802 storing a computer program and a processor 804 configured to execute the steps in any of the above-mentioned method embodiments by means of the computer program.

[0118] Optionally, in the present embodiment, the above-mentioned electronic device can be located in at least one of the network devices in a computer network. Those skilled in the art can understand that the structure shown in FIG. 8 is only schematic, and the electronic device can also be a device containing the above-mentioned flash memory. FIG. 8 does not limit the structure of the above-mentioned electronic device. For example, the electronic device can further include more or fewer components (such as network interfaces, etc.) than those shown in FIG. 8, or have a different configuration from that shown in FIG. 8.

[0119] In the present embodiment, the memory 802 can be configured to store software programs and modules, such as the program instructions / modules corresponding to the method and device for correcting the chargeable and dischargeable capacity in the embodiments of the present disclosure, and the processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 802, i.e., implements the above-mentioned method for correcting the chargeable and dischargeable capacity. The memory 802 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 802 can further include a memory remotely arranged with respect to the processor 804, which can be connected to the terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. Specifically, the memory 802 can but not limited to be used to contain log information of modeling data. As an example, as shown in FIG. 8, the memory 802 can but not limited to include the modules in the above-mentioned device for correcting the chargeable and dischargeable capacity. In addition, other module units in the above-mentioned device for correcting the chargeable and dischargeable capacity can also be included but not limited to, which will not be described herein.

[0120] Optionally, the above-mentioned transmission device 806 is configured to receive or send data via a network. Specifically, examples of the above-mentioned network can include wired networks and wireless networks. In one example, the transmission device 806 includes a network adapter (NIC), which can be connected to other network devices and routers through a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 806 is a radio frequency (RF) module configured to communicate with the Internet in a wireless manner.

[0121] Further, the electronic device described above further includes a display 808; and a connection bus 810 configured to connect each of the module components in the electronic device.

[0122] Embodiments of the present disclosure further provide a computer program product, which comprises a computer program, and the computer program, when executed by a processor, implements the steps in any of the method embodiments described above.

[0123] Embodiments of the present disclosure further provide another computer program product, which comprises a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps in any of the method embodiments described above.

[0124] Embodiments of the present disclosure further provide a computer program, which comprises computer instructions stored in a computer readable storage medium; a processor of a computer device reads the computer instructions from the computer readable storage medium, and executes the computer instructions, so that the computer device executes the steps in any of the method embodiments described above.

[0125] The specific examples in the embodiments can refer to the examples described in the above embodiments and exemplary embodiments, and will not be described here again.

[0126] Obviously, those skilled in the art should understand that the modules or steps of the present disclosure described above can be realized by general computing devices, and they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and they can be realized by program codes executable by computing devices, so that they can be stored in storage devices and executed by computing devices, and in some cases, the steps shown or described can be executed in different order, or they can be manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present disclosure is not limited to any specific combination of hardware and software.

[0127] The above only describes preferred embodiments of the present disclosure and is not intended to limit the present disclosure. For those skilled in the art, the present disclosure can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for correcting charge and discharge capacity, comprising: Obtaining X*Y sets of test data generated by test information of each of X batteries to be tested under Y types of variable temperature parameters, wherein the test data includes a first key characteristic of the battery to be tested, the first key characteristic including a starting ambient temperature, a transition ambient temperature, and battery parameter information, the battery parameter information including state of charge information, where X and Y are positive integers; Determining the full discharge loss and / or full charge loss corresponding to each set of test data in the X*Y sets of test data, and using the full discharge loss and / or full charge loss as target values, respectively, and concatenating them with the corresponding first key features to generate a training data set, and fitting the parameters of the model based on the training data set to obtain a nonlinear prediction model; predicting a target full-discharge loss and / or a target full-charge loss corresponding to the target battery under a current state of charge and current variable temperature conditions based on the second key feature of the target battery and the nonlinear prediction model; The target full-discharge loss is used to correct the first dischargeable capacity of the target battery after the temperature change occurs, and / or the target full-charge loss is used to correct the first chargeable capacity of the target battery after the temperature change occurs.

2. The method for correcting the chargeable and dischargeable capacity according to claim 1, wherein: The second key feature includes a target starting ambient temperature of the target battery, and a target dischargeable capacity or a target chargeable capacity of the target battery at the target starting ambient temperature; The correcting the first dischargeable capacity of the target battery after the temperature change by using the target full discharge loss includes: Correcting the initial dischargeable capacity of the target battery after the temperature changes according to the target dischargeable capacity, the target full-discharge loss, and the rated capacity of the target battery; The method of correcting the first chargeable capacity of the target battery after the temperature change by using the target full charge loss includes: Correcting the initial rechargeable capacity of the target battery after the temperature changes according to the target rechargeable capacity, the target full charge loss, and the rated capacity; The temperature change is used to indicate that the operating environment temperature of the target battery is adjusted from the target initial environment temperature to the target migration environment temperature.

3. The method for correcting the chargeable and dischargeable capacity according to claim 1, wherein: The nonlinear prediction model includes M sub-prediction models, where M is a positive integer. Before predicting a target full-discharge loss and / or a target full-charge loss corresponding to the target battery under a current state of charge and current variable temperature conditions based on the second key feature of the target battery and the nonlinear prediction model, the method further includes: Calculate the feature distance between the second key feature and the first key feature in the training data set to obtain X*Y feature distances; Select the first A target feature distances with the closest feature distances from the X*Y feature distances, and determine the target key features corresponding to each target feature distance in the first A target feature distances to obtain A target key features, where A is a positive integer.

4. The method for correcting the chargeable and dischargeable capacity according to claim 3, wherein: Predicting a target full-discharge loss corresponding to the target battery under a current state of charge and current variable temperature conditions based on the second key feature of the target battery and the nonlinear prediction model includes: Determine the actual full-load loss corresponding to each target key feature, and perform the following operation on each of the A target key features: input each of the A target key features into M sub-prediction models to obtain M predicted values, where the predicted value is the estimated full-load loss determined by the sub-prediction model; For each target key feature, determine the error between each predicted value and the actual full-discharge loss to obtain M prediction errors, and select B first-class sub-prediction models that meet preset error conditions from the M sub-prediction models based on the M prediction errors; where B is a positive integer; Summarize the B first-category sub-prediction models corresponding to the A target key features to obtain A*B first-category sub-prediction models; The A*B first-category sub-prediction models are used to predict the target full-discharge loss corresponding to the target battery under the current state of charge and current variable temperature conditions.

5. The method for correcting the chargeable and dischargeable capacity according to claim 4, wherein: The target full discharge loss corresponding to the target battery under the current state of charge and current variable temperature conditions is predicted using the A*B first-category sub-prediction models, including: Deduplication processing is performed on the A*B first-category sub-prediction models to obtain N second-category sub-prediction models, where A*B≤N; Inputting the second key features into the N second-category sub-prediction models respectively to obtain N prediction results; Based on the N prediction results, a target full-discharge loss corresponding to the target battery under the current state of charge and current variable temperature conditions is determined; wherein N is a positive integer less than M.

6. The method for correcting the chargeable and dischargeable capacity according to claim 1, wherein: The battery parameter information includes initial state of charge, initial voltage, migration state of charge, migration voltage, and battery temperature. The generating of X*Y sets of test data based on the test information of each battery in the X batteries to be tested under Y types of temperature varying parameters includes: For each battery to be tested, perform at least two charge-discharge cycle operations on the battery to be tested at the test starting ambient temperature of each type of variable temperature parameter; charging the battery to be tested for a first preset time or discharging it for a second preset time, so that the battery to be tested reaches a preset initial state of charge, wherein the initial state of charge is used to determine a first chargeable capacity and / or a first dischargeable capacity of the battery to be tested; The battery to be tested is migrated from the test starting ambient temperature to a corresponding test migration ambient temperature, and a migration state of charge, a migration voltage, and a battery temperature of the battery to be tested are obtained, wherein the migration state of charge is used to determine a second chargeable capacity and / or a second dischargeable capacity of the battery to be tested.

7. A device for correcting charge and discharge capacity, comprising: an acquisition module configured to acquire test information of each of X batteries to be tested under Y types of variable temperature parameters to generate X*Y sets of test data, wherein the test data includes a first key characteristic of the battery to be tested, the first key characteristic including a starting ambient temperature, a transition ambient temperature, and battery parameter information, the battery parameter information including state of charge information, and X and Y are positive integers; a determination module configured to determine a full discharge loss and / or full charge loss corresponding to each set of test data in the X*Y sets of test data, and use the full discharge loss and / or full charge loss as target values, respectively, and concatenate them with the corresponding first key features to generate a training data set, and obtain a nonlinear prediction model based on the parameters of the model fitted by the training data set; a prediction module configured to predict a target full-discharge loss and / or a target full-charge loss corresponding to the target battery under a current state of charge and current variable temperature conditions based on a second key feature of the target battery and the nonlinear prediction model; The correction module is configured to correct the first dischargeable capacity of the target battery after the temperature change by using the target full-discharge loss, and / or to correct the first chargeable capacity of the target battery after the temperature change by using the target full-charge loss.

8. A computer-readable storage medium comprising a stored program, wherein: When the program is executed, the steps of the method described in any one of claims 1 to 6 are executed.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the steps of the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 6.

Images