Model training method and device and battery power state updating method and device

Abstract

The invention provides a model training method and device and a battery power state updating method and device, and relates to the technical field of battery detection, and the method comprises the steps: obtaining operation parameters of a first vehicle in a plurality of time periods in a driving process, the plurality of time periods comprising a first group of time periods, the first group of time periods comprises a plurality of first time periods with the same duration, a first power battery of the first vehicle is in a first state in the first group of time periods, and the first state is one of a discharging state and a charging state during energy feedback; the power state SOP of the first power battery in each first time period is obtained, and the SOP of the first power battery in each first time period enables the first power battery not to reach the first preset protection voltage in each first time period; and the first machine learning model is trained by taking the operation parameters of the first vehicle in each first time period as input and taking the SOP of the first power battery in each first time period as output.

Description

Technical Field

[0001] The present disclosure relates to the technical field of battery detection, and in particular to a model training method and device, and a battery power status update method and device. Background Art

[0002] The power state (SOP) of a battery refers to the current upper limit or power upper limit when the battery reaches the preset protection voltage of the battery after charging or discharging for a certain period of time. Summary of the invention

[0003] In the related art, when the power battery of a vehicle is charged or discharged within a preset time period and reaches a preset protection voltage, the SOP of the power battery needs to be updated to protect the power battery. However, during the driving process of the vehicle, the updated SOP of the power battery of the vehicle is inaccurate.

[0004] After analysis, it is found that when updating the SOP of the power battery of a vehicle, only the electrical parameters at a certain moment (for example, the moment when a preset protection voltage is reached) are considered to calculate the updated SOP of the power battery of the vehicle.

[0005] In this way, during the driving process of the vehicle, the SOP of the power battery is calculated only by using the electrical parameters at a certain moment, which cannot accurately reflect the relationship between the dynamic changes of the electrical parameters of the power battery and the SOP of the power battery during the driving process of the vehicle, resulting in inaccurate SOP of the power battery of the updated vehicle.

[0006] In order to solve the above problems, the embodiments of the present disclosure propose the following solutions.

[0007] According to one aspect of an embodiment of the present disclosure, a model training method is provided, comprising: obtaining operating parameters of a first vehicle in multiple time periods during driving, the multiple time periods including a first group of time periods, the first group of time periods including multiple first time periods of the same length, wherein a first power battery of the first vehicle is in a first state during the first group of time periods, the first state being one of a discharge state and a charging state during energy feedback; obtaining a power state SOP of the first power battery in each first time period, wherein the SOP of the first power battery in each first time period makes the first power battery fail to reach a first preset protection voltage in each first time period; and training a first machine learning model by taking the operating parameters of the first vehicle in each first time period as input and the SOP of the first power battery in each first time period as output.

[0008] In some embodiments, the first group of time periods also includes multiple second time periods of the same length, wherein the lengths of the multiple second time periods are different from the lengths of the multiple first time periods, and the method also includes: obtaining the SOP of the first power battery in each second time period, wherein the SOP of the first power battery in each second time period is such that the first power battery does not reach the first preset protection voltage in each second time period; and training a second machine learning model by taking the operating parameters of the first vehicle in each second time period as input and the SOP of the first power battery in each second time period as output.

[0009] In some embodiments, the multiple time periods also include a second group of time periods, the second group of time periods includes multiple third time periods, wherein the first power battery is in a second state during the second group of time periods, the duration of the multiple third time periods is the same as the duration of the multiple first time periods, and the second state is another of the discharge state and the charging state, and the method also includes: obtaining the SOP of the first power battery in each third time period, wherein the SOP of the first power battery in each third time period makes the first power battery fail to reach the first preset protection voltage in each third time period; and taking the operating parameters of the first vehicle in each third time period as input and the SOP of the first power battery in each third time period as output, respectively, to train the third machine learning model.

[0010] In some embodiments, the second group of time periods also includes multiple fourth time periods of the same length, wherein the length of the multiple fourth time periods is different from the length of the multiple third time periods, and the method also includes: obtaining the SOP of the first power battery in each fourth time period, wherein the SOP of the first power battery in each fourth time period is such that the first power battery does not reach the first preset protection voltage in each fourth time period; and training a fourth machine learning model by taking the operating parameters of the first vehicle in each fourth time period as input and the SOP of the first power battery in each fourth time period as output.

[0011] In some embodiments, the operating parameters of each time period include at least one of the following: the state of charge SOC of the first power battery at the beginning of each time period; the rate of change of the SOC of the first power battery in each time period; the state of health SOH of the first power battery in each time period; and the mileage of the first vehicle in each time period.

[0012] In some embodiments, the operating parameters of each time period include the mileage of the first vehicle in each time period and at least one of the following: the SOC of the first power battery at the beginning of each time period; the rate of change of the SOC of the first power battery in each time period; and the SOH of the first power battery in each time period.

[0013] In some embodiments, the operating parameter of each time period further includes at least one of the following: the temperature of the first power battery at the beginning of each time period; and the rate of change of the temperature of the first power battery in each time period.

[0014] In some embodiments, when the first state is the charging state, the operating parameters of each time period also include at least one of the charging voltage of the first power battery at the beginning of each time period and the rate of change of the charging voltage of the first power battery in each time period; when the first state is the discharging state, the operating parameters of each time period also include at least one of the discharge voltage of the first power battery at the beginning of each time period and the rate of change of the discharge voltage of the first power battery in each time period.

[0015] In some embodiments, the starting point and end point of each time period in the first group of time periods are the moments when the SOP of the first vehicle changes, wherein: when the first state is the discharging state, each time period in the first group of time periods is between the moment when the first vehicle starts and the moment when the first vehicle starts braking; and when the first state is the charging state, each time period in the first group of time periods is between the moment when the first vehicle starts braking and the moment when the first vehicle ends braking.

[0016] According to another aspect of an embodiment of the present disclosure, a method for updating the power state of a battery is provided, comprising: in response to a second vehicle satisfying a first update condition during driving, obtaining operating parameters of a first preset time period of the second vehicle, wherein the first update condition is: a second power battery of the second vehicle is in a first state, and a voltage of the second power battery reaches a first preset protection voltage, wherein the first state is one of a discharge state and a charging state during energy feedback, a current power state SOP of the second power battery is a first SOP, and a duration of the first preset time period is less than or equal to a duration of a current time period corresponding to the first SOP; and inputting the operating parameters of the first preset time period into a machine learning model to estimate a second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and updating the first SOP to the second SOP.

[0017] In some embodiments, the inputting the operating parameters of the first preset time period into the machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and updating the first SOP to the second SOP includes: when the current time period corresponding to the first SOP is the first time period, the operating parameters of the second vehicle of the first preset time period are input into the first machine learning model, wherein the first machine learning model is trained based on the operating parameters of multiple first time periods of the first vehicle during driving and the SOP of the first power battery of the first vehicle during multiple first time periods, and the first power battery is in the first state during the first time period; when the current time period corresponding to the first SOP is the second time period, the operating parameters of the second vehicle of the first preset time period are input into the second machine learning model, wherein the second machine learning model is trained based on the operating parameters of multiple second time periods of the first vehicle during driving and the SOP of the first power battery during multiple second time periods, the duration of the second time period is different from the duration of the first time period, and the first power battery is in the first state during the second time period.

[0018] In some embodiments, when the moment when the voltage of the second power battery reaches the first preset protection voltage is the end point of the current time period corresponding to the first SOP, the first preset time period is the current time period corresponding to the first SOP; when the moment when the voltage of the second power battery reaches the first preset protection voltage is before the end point of the current time period corresponding to the first SOP, the first preset time period is from the starting point of the current time period corresponding to the first SOP to the moment when the voltage of the second power battery reaches the first preset protection voltage.

[0019] In some embodiments, in response to the second vehicle satisfying a second update condition during driving, the operating parameters of the second vehicle for a second preset time period are obtained, wherein the second update condition is: the second power battery of the second vehicle is in a second state, and the voltage of the second power battery reaches a second preset protection voltage, wherein the second state is another of the discharge state and the charging state, the current power state SOP of the second power battery is a third SOP, and the duration of the second preset time period is less than or equal to the duration of the current time period corresponding to the third SOP; and the operating parameters of the second preset time period are input into a machine learning model to estimate the fourth SOP of the second power battery corresponding to the current time period corresponding to the third SOP, and the third SOP is updated to the fourth SOP.

[0020] In some embodiments, the inputting the operating parameters of the second preset time period into the machine learning model to estimate the fourth SOP of the second power battery corresponding to the current time period corresponding to the third SOP, and updating the third SOP to the fourth SOP includes: when the current time period corresponding to the third SOP is the third time period, the operating parameters of the second vehicle in the second preset time period are input into the third machine learning model, wherein the third machine learning model is trained based on the operating parameters of the first vehicle in multiple third time periods during driving and the SOP of the first power battery of the first vehicle in multiple third time periods, and the first power battery is in the second state during the third time period; when the current time period corresponding to the third SOP is the fourth time period, the operating parameters of the second preset time period are input into the fourth machine learning model, wherein the fourth machine learning model is trained based on the operating parameters of the first vehicle in multiple fourth time periods during driving and the SOP of the first power battery in multiple fourth time periods, the duration of the fourth time period is different from the duration of the third time period, and the first power battery is in the second state during the fourth time period.

[0021] In some embodiments, when the moment when the voltage of the second power battery reaches the second preset protection voltage is the end point of the current time period corresponding to the third SOP, the second preset time period is the current time period corresponding to the third SOP; when the moment when the voltage of the second power battery reaches the second preset protection voltage is before the end point of the current time period corresponding to the third SOP, the second preset time period is from the starting point of the current time period corresponding to the third SOP to the moment when the voltage of the second power battery reaches the second preset protection voltage.

[0022] In some embodiments, the operating parameters of the first preset time period include at least one of the following: the state of charge SOC of the second power battery at the beginning of the first preset time period; the rate of change of the SOC of the second power battery in the first preset time period; the state of health SOH of the second power battery in the first preset time period; and the mileage of the second vehicle in the first preset time period.

[0023] In some embodiments, the operating parameters of the first preset time period include the mileage of the second vehicle during the first preset time period and at least one of the following: the SOC of the second power battery at the beginning of the first preset time period; the rate of change of the SOC of the second power battery during the first preset time period; and the SOH of the second power battery during the first preset time period.

[0024] In some embodiments, the operating parameters of the first preset time period further include at least one of the following: the temperature of the second power battery at the beginning of the first preset time period; and the rate of change of the temperature of the second power battery during the first preset time period.

[0025] In some embodiments, when the first state is the charging state, the operating parameters of the first preset time period also include at least one of the charging voltage of the second power battery at the beginning of the first preset time period and the rate of change of the charging voltage of the second power battery during the first preset time period; when the first state is the discharging state, the operating parameters of the first preset time period also include at least one of the discharge voltage of the second power battery at the beginning of the first preset time period and the rate of change of the discharge voltage of the second power battery during the first preset time period.

[0026] According to another aspect of an embodiment of the present disclosure, a model training device is provided, comprising: a module configured to execute the model training method of any one of the above embodiments.

[0027] According to another aspect of the embodiments of the present disclosure, there is provided a device for updating the power status of a battery, comprising: a module configured to execute the method for updating the power status of a battery of any one of the above embodiments.

[0028] According to another aspect of an embodiment of the present disclosure, an electronic device is provided, including: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the method described in any one of the above embodiments based on instructions stored in the memory.

[0029] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, comprising computer program instructions, wherein when the computer program instructions are executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0030] According to another aspect of the embodiments of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0031] In the embodiment of the present disclosure, the operating parameters of the first group of time periods in the multiple time periods during the driving of the first vehicle, and the SOP of the first power battery of the first vehicle in multiple first time periods of the same length in the first group of time periods are obtained, and the operating parameters of the first vehicle in each first time period are used as input and the SOP of the first power battery in each first time period is used as output to train the first machine learning model. Here, the first power battery of the first vehicle is in one of a discharge state and a charging state during energy feedback in the first group of time periods.

[0032] In this way, when training the first machine learning model, the dynamic changes in the operating parameters of the first vehicle during driving and the SOP of the first time period can be captured within a first time period when the first power battery is in the first state. This helps to enable the trained first machine learning model to accurately estimate the SOP of the updated power battery when the SOP of the first power battery in the first state needs to be updated, thereby improving the accuracy of the updated SOP of the power battery.

[0033] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 It is a flowchart of a method for training a model according to some embodiments of the present disclosure.

[0036] Figure 2 is a flowchart of a method for updating the power status of a battery according to some embodiments of the present disclosure.

[0037] Figure 3 is a schematic diagram of a structure of a training device for a model according to some embodiments of the present disclosure

[0038] Figure 4 4 is a schematic diagram of the structure of a device for updating the power status of a battery according to some embodiments of the present disclosure.

[0039] Figure 5 is a schematic diagram of the structure of an electronic device according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0041] Unless specifically stated otherwise, the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0042] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not necessarily drawn according to the actual proportional relationship.

[0043] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0044] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0045] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0046] In addition, in the description of the present disclosure, the terms "first", "second", "third", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance and order. Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in a sequential order, or requiring that all illustrated operations be performed to achieve the desired results. In some cases, multitasking and parallel processing can be advantageous.

[0047] Figure 1 It is a flowchart of a method for training a model according to some embodiments of the present disclosure.

[0048] In step 102, operating parameters of the first vehicle in multiple time periods during the driving process are obtained.

[0049] Here, the multiple time periods include a first group of time periods, and the first group of time periods includes multiple first time periods of the same duration. For example, the first time period is 5 seconds.

[0050] It should be understood that each of the multiple time periods is a fixed time period, that is, a time period with a fixed and unchanging duration.

[0051] The first power battery of the first vehicle is in a first state during a first group of time periods, and the first state is one of a discharge state and a charge state during energy feedback. It should be understood that the discharge state refers to the state when the power battery drives the vehicle to travel, and the charge state during energy feedback refers to the charge state of the power battery when the vehicle performs brake energy recovery.

[0052] In some embodiments, the first vehicle is a vehicle driven by a first power battery (eg, a forklift and / or a heavy truck driven by a power battery). For example, the first vehicle is an electric vehicle; or for another example, the first vehicle is a hybrid vehicle.

[0053] In step 104, the SOP of the first power battery in each first time period is obtained. Here, the SOP of the first power battery in each first time period is such that the first power battery does not reach the first preset protection voltage in each first time period.

[0054] A first time period corresponds to a SOP of a first power battery, that is, the SOP of the first power battery in a first time period remains unchanged.

[0055] As some embodiments, the SOP of the first power battery in each first time period may be manually set. For example, when set based on manual experience, as long as the SOP of the first power battery in each first time period does not reach the first preset protection voltage in each first time period, it is sufficient.

[0056] As some implementation modes, the SOP of the first power battery in each first time period may be expressed in the form of current or power.

[0057] For example, when expressed in the form of current and the first state is the discharge state, the SOP of the first power battery in the first time period may be the average value of the discharge current of the first power battery in the first time period; for another example, when expressed in the form of power and the first state is the discharge state, the SOP of the first power battery in the first time period may be the integral value of the discharge power of the first power battery in the first time period over time.

[0058] It should be understood that the representation form of the SOP of the first power battery when the first state is a charging state during energy feedback is similar to the representation form of the SOP of the first power battery when the first state is a discharging state.

[0059] In step 106, the operating parameters of the first vehicle in each first time period are used as inputs and the SOP of the first power battery in each first time period is used as outputs to train the first machine learning model. For example, the first machine learning model is a long short-term memory network (LSTM) model or a recurrent neural network (RNN) model.

[0060] That is to say, the operating parameters of the first vehicle in each first time period and the SOP of the first power battery in each first time period are respectively used as training data for the first machine learning model.

[0061] In the above embodiment, the operating parameters of the first group of time periods in the multiple time periods during the driving of the first vehicle, and the SOP of the first power battery of the first vehicle in multiple first time periods of the same length in the first group of time periods are obtained, and the operating parameters of the first vehicle in each first time period are used as input and the SOP of the first power battery in each first time period is used as output to train the first machine learning model. Here, the first power battery of the first vehicle is in one of a discharge state and a charging state during energy feedback in the first group of time periods.

[0062] In this way, when training the first machine learning model, the correspondence between the dynamic changes of the operating parameters of the first vehicle during driving and the SOP of the first time period can be captured within a first time period when the first power battery is in the first state. This helps to enable the trained first machine learning model to accurately estimate the SOP of the updated power battery when the SOP of the first power battery in the first state needs to be updated, thereby improving the accuracy of the updated SOP of the power battery.

[0063] In some embodiments, the start point and the end point of each time period in the first set of time periods are the moments when the SOP of the first vehicle changes.

[0064] Here, when the first state is a discharging state, each time period in the first group of time periods is between the moment when the first vehicle starts and the moment when the first vehicle starts braking; when the first state is a charging state, each time period in the first group of time periods is between the moment when the first vehicle starts braking and the moment when the first vehicle ends braking.

[0065] For example, when the first state is a discharge state, if the SOP of the first vehicle does not change between the time when the first vehicle starts and the time when the first vehicle starts braking, then the time when the first vehicle starts to the time when the first vehicle starts braking is a time period.

[0066] For another example, when the first state is the discharge state, if the SOP of the first vehicle changes between the time when the first vehicle starts and the time when the first vehicle starts braking, then the time period between any two adjacent times when the SOP of the first vehicle changes between the time when the first vehicle starts and the time when the first vehicle starts braking is a time period.

[0067] It should be understood that the SOP of the first vehicle when the first vehicle has just started and has not yet started is different from the SOP when the first vehicle has just started and has started.

[0068] Similarly, when the first state is the charging state, if the SOP of the first vehicle remains unchanged between the moment when the first vehicle starts braking and the moment when the first vehicle ends braking, then the moment when the first vehicle starts braking to the moment when the first vehicle ends braking is a time period.

[0069] If the SOP of the first vehicle changes between the moment when the first vehicle starts braking and the moment when the first vehicle ends braking, then the time period between any two adjacent moments when the SOP of the first vehicle changes between the moment when the first vehicle starts braking and the moment when the first vehicle ends braking is a time period.

[0070] In this way, by utilizing the moment when the first vehicle starts, the moment when braking starts, the moment when braking ends, and the change in the SOP of the first vehicle, the time period when the first power battery is in a discharge state with a constant SOP and the time period when the first power battery is in a charging state with a constant SOP can be accurately divided, and then the operating parameters of the first vehicle in multiple time periods during driving can be accurately obtained, which is helpful to accurately train the machine learning model in the later stage.

[0071] In some embodiments, the operating parameters of each time period include at least one of the following four parameters: the state of charge (SOC) of the first power battery at the beginning of each time period, the rate of change of the SOC of the first power battery in each time period, the state of health (SOH) of the first power battery in each time period, and the mileage of the first vehicle in each time period.

[0072] For example, the operating parameters of each time period include any one, any two, any three or all of the above four parameters.

[0073] In the above embodiment, the operating parameters of each time period include at least one of the above four parameters. In this way, at least one of the above four parameters can be considered when training the first machine learning model, which helps to further improve the accuracy of the SOP of the first power battery of the updated first vehicle estimated by the trained first machine learning model.

[0074] In some embodiments, the operating parameters of each time period include the mileage of the first vehicle in each time period, and the operating parameters of each time period also include at least one of the SOC of the first power battery at the beginning of each time period, the rate of change of the SOC of the first power battery in each time period, and the SOH of the first power battery in each time period.

[0075] For example, the operating parameters of each time period include the mileage of the first vehicle in each time period, and the operating parameters of each time period also include any one, any two or all of the SOC of the first power battery at the beginning of each time period, the rate of change of the SOC of the first power battery in each time period, and the SOH of the first power battery in each time period.

[0076] In this way, the mileage of the first vehicle in each time period and other items related to the first power battery can be considered simultaneously when training the first machine learning model, which helps to further improve the accuracy of the SOP of the updated first power battery of the first vehicle estimated by the trained first machine learning model.

[0077] In some embodiments, the operating parameters of each time period further include at least one of the temperature of the first power battery at the beginning of each time period and the rate of change of the temperature of the first power battery in each time period.

[0078] For example, the operating parameters of each time period include at least one of the above four parameters and at least one of the temperature of the first power battery at the beginning of each time period and the rate of change of the temperature of the first power battery in each time period.

[0079] In this way, when training the first machine learning model, richer operating parameters can be considered, which helps to further improve the accuracy of the SOP of the updated first power battery of the first vehicle estimated by the trained first machine learning model.

[0080] In some embodiments, when the first state is a charging state during energy feedback, the operating parameters of each time period further include at least one of a charging voltage of the first power battery at the beginning of each time period and a rate of change of the charging voltage of the first power battery in each time period.

[0081] For example, when the first state is the charging state during energy feedback, the operating parameters of each time period include, in addition to at least one of the above four parameters, at least one of the charging voltage of the first power battery at the beginning of each time period and the rate of change of the charging voltage of the first power battery in each time period.

[0082] For another example, when the first state is the charging state during energy feedback, the operating parameters of each time period include, in addition to at least one of the above four parameters, at least one of the temperature of the first power battery at the beginning of each time period and the rate of change of the temperature of the first power battery in each time period, and at least one of the charging voltage of the first power battery at the beginning of each time period and the rate of change of the charging voltage of the first power battery in each time period.

[0083] In some embodiments, when the first state is a discharge state, the operating parameters of each time period further include at least one of a discharge voltage of the first power battery at the beginning of each time period and a change rate of the discharge voltage of the first power battery in each time period.

[0084] For example, when the first state is a discharge state, the operating parameters of each time period include, in addition to at least one of the above four parameters, at least one of the discharge voltage of the first power battery at the beginning of each time period and the rate of change of the discharge voltage of the first power battery in each time period.

[0085] For another example, when the first state is the discharge state, the operating parameters of each time period include, in addition to at least one of the above four parameters, at least one of the temperature of the first power battery at the beginning of each time period and the rate of change of the temperature of the first power battery in each time period, and at least one of the discharge voltage of the first power battery at the beginning of each time period and the rate of change of the discharge voltage of the first power battery in each time period.

[0086] In this way, when training the first machine learning model, the type of the first state and the operating parameters corresponding to the first state can be considered, which helps to further improve the accuracy of the SOP of the updated first power battery of the first vehicle estimated by the trained first machine learning model.

[0087] In some embodiments, a correlation analysis method may be used to analyze the correlation between multiple alternatives and the SOP of the first power battery, and determine the operating parameters of each time period from the multiple alternatives. For example, alternatives with a correlation higher than a preset value are used as operating parameters for each time period; for another example, a preset number of alternatives with a high correlation ranking are used as operating parameters for each time period.

[0088] Here, the alternatives include the SOC of the first power battery at the beginning of each time period, the rate of change of the SOC of the first power battery in each time period, the SOH of the first power battery in each time period, the mileage of the first vehicle in each time period, the temperature of the first power battery at the beginning of each time period, and the rate of change of the temperature of the first power battery in each time period.

[0089] As some embodiments, the correlation analysis method includes one of a Pearson correlation coefficient analysis method and a Spearman correlation coefficient analysis method.

[0090] In the above embodiment, the operating parameters for training the first machine learning model can be accurately determined based on the correlation between the alternative items and the SOP, which helps to further improve the accuracy of the SOP of the updated first power battery of the first vehicle estimated by the trained first machine learning model.

[0091] In some embodiments, when the first state is a charging state, the plurality of alternative items further include at least one of a charging voltage of the first power battery at the beginning of each time period and a rate of change of the charging voltage of the first power battery in each time period.

[0092] In some embodiments, when the first state is a discharge state, the plurality of alternative items further include at least one of a discharge voltage of the first power battery at the beginning of each time period and a change rate of the discharge voltage of the first power battery in each time period.

[0093] In the above embodiment, the multiple alternatives also include voltage and the rate of change of voltage. In this way, when determining the operating parameters of each time period from the multiple alternatives using the correlation analysis method, different situations of the first state can be considered, thereby more accurately determining the operating parameters for training the first machine learning model, which helps to further improve the accuracy of the SOP of the first power battery of the first vehicle estimated by the trained first machine learning model.

[0094] The inventors have noticed that the SOP of the power battery of some vehicles when charging is inversely proportional to the length of the preset time period corresponding to the charging. That is, the SOP of the power battery when charging in a short time is greater than the SOP when charging in a long time.

[0095] Similarly, the SOP of a power battery when it is discharged in a short period of time is greater than the SOP when it is discharged in a long period of time.

[0096] In order to more accurately capture the correspondence between the duration of the above-mentioned different preset time periods and the value of the SOP of the power battery, so as to further improve the accuracy of the updated SOP of the power battery, the present disclosure also proposes the following technical solutions.

[0097] In some embodiments, the first group of time periods also includes a plurality of second time periods of the same duration. Here, the durations of the plurality of second time periods are different from the durations of the plurality of first time periods.

[0098] It should be understood that the duration of the second time period may be greater or less than the duration of the first time period. For example, the first time period is 5 seconds and the second time period is 30 seconds; or for another example, the first time period is 30 seconds and the second time period is 5 seconds.

[0099] In this case, the SOP of the first power battery in each second time period is also obtained, and the operating parameters of the first vehicle in each second time period are respectively used as input and the SOP of the first power battery in each second time period is used as output to train the second machine learning model. Here, the SOP of the first power battery in each second time period causes the first power battery to fail to reach the first preset protection voltage in each second time period.

[0100] That is to say, the operating parameters of the first vehicle in each second time period and the SOP of the first power battery in each second time period are used as training data for the second machine learning model.

[0101] In the above embodiment, the operating parameters of the first vehicle in each second time period different from the first time period and the SOP of the first power battery in each second time period are used to train a second machine learning model different from the first machine learning model, so as to more accurately capture the correspondence between the lengths of different preset time periods and the SOP of the power battery, which helps to further improve the accuracy of the updated SOP of the power battery.

[0102] In some embodiments, the plurality of time periods further includes a second group of time periods, the second group of time periods includes a plurality of third time periods, and the durations of the plurality of third time periods are the same as the durations of the plurality of first time periods.

[0103] Here, the first power battery is in the second state during the second group of time periods, and the second state is the other of a discharge state and a charge state during energy feedback. That is, when the first state is a discharge state, the second state is a charge state; when the first state is a charge state, the second state is a discharge state.

[0104] In this case, the SOP of the first power battery in each third time period is also obtained, and the operating parameters of the first vehicle in each third time period are respectively used as input and the SOP of the first power battery in each third time period is used as output to train the third machine learning model. Here, the SOP of the first power battery in each third time period causes the first power battery to fail to reach the second preset protection voltage in each third time period.

[0105] That is to say, the operating parameters of the first vehicle in each third time period and the SOP of the first power battery in each third time period are used as training data for the third machine learning model.

[0106] In the above embodiment, the operating parameters of the second group of time periods in the multiple time periods during the driving process of the first vehicle, and the power state SOP of the first power battery of the first vehicle in each third time period of the second group of time periods with the same duration as the first time period are obtained, and the operating parameters of the first vehicle in each third time period are used as input and the SOP of the first power battery in each third time period is used as output, respectively, to train a third machine learning model different from the first machine learning model. Here, the first power battery of the first vehicle is in another state of discharge and charging during energy feedback in the second group of time periods.

[0107] In this way, regardless of whether the first power battery is in a discharging state or a charging state, the corresponding machine learning model can be trained to capture the corresponding relationship between the dynamic changes of the operating parameters in the corresponding time period and the SOP of the time period, which helps to more accurately estimate the SOP of the updated power battery.

[0108] In some embodiments, similar to the case where the first group of time periods also includes multiple second time periods of the same duration, the second group of time periods also includes multiple fourth time periods of the same duration. Here, the duration of the multiple fourth time periods is different from the duration of the multiple third time periods. As some implementation methods, the duration of the fourth time period is equal to the duration of the second time period.

[0109] In this case, the SOP of the first power battery in each fourth time period is also obtained, and the operating parameters of the first vehicle in each fourth time period are respectively used as input and the SOP of the first power battery in each fourth time period is used as output to train the fourth machine learning model. Here, the SOP of the first power battery in each fourth time period causes the first power battery to fail to reach the second preset protection voltage in each fourth time period.

[0110] That is to say, the operating parameters of the first vehicle in each fourth time period and the SOP of the first power battery in each fourth time period are used as training data for the fourth machine learning model.

[0111] In the above embodiment, the operating parameters of the first vehicle in multiple fourth time periods different from the third time period and the SOP of the first power battery in each fourth time period are used to train a fourth machine learning model different from the third machine learning model, so that the correspondence between the lengths of different preset time periods and the values ​​of the SOP of the power battery can be more accurately captured, which helps to further improve the accuracy of the SOP of the power battery of the updated vehicle.

[0112] It should be understood that the first machine learning model, the second machine learning model, the third machine learning model and the fourth machine learning model are different from each other.

[0113] The present disclosure also proposes a method for updating the power status of a battery.

[0114] Figure 2 is a flowchart of a method for updating the power status of a battery according to some embodiments of the present disclosure.

[0115] In step 202, in response to the second vehicle satisfying a first update condition during driving, operating parameters of the second vehicle in a first preset time period are acquired.

[0116] Here, the first update condition is: the second power battery of the second vehicle is in the first state, and the voltage of the second power battery reaches the first preset protection voltage.

[0117] The first state is one of a discharge state and a charging state during energy feedback, the current power state SOP of the second power battery is the first SOP, and the duration of the first preset time period is less than or equal to the duration of the current time period corresponding to the first SOP. The current time period is a fixed time period, that is, a time period with a fixed and unchanging duration, for example, 5 seconds or 30 seconds.

[0118] In step 204, the operating parameters of the first preset time period are input into the machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and the first SOP is updated to the second SOP.

[0119] As some implementations, if the first update condition is not met, the update operation is not performed. For example, if the voltage of the second power battery does not reach the first preset protection voltage, the update operation is not performed.

[0120] In the above embodiment, when the second power battery of the second vehicle is in one of a discharge state and a charging state during energy feedback, and the voltage of the second power battery reaches a first preset protection voltage, the operating parameters of the second vehicle for a first preset time period are obtained, and the operating parameters for the first preset time period are input into a machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and the first SOP is updated to the second SOP.

[0121] In this way, by utilizing the dynamic changes of the operating parameters of the second power battery within the first preset time period when it is in the first state, the SOP corresponding to the first preset time period is determined through a machine learning model. When the SOP of the second power battery in the first state needs to be updated, the SOP of the updated second power battery can be accurately estimated, thereby improving the accuracy of the SOP of the updated power battery of the vehicle.

[0122] In some embodiments, when the current time period corresponding to the first SOP is the first time period, the operating parameters of the second vehicle in the first preset time period are input into the first machine learning model.

[0123] Here, the first machine learning model is trained based on the operating parameters of the first vehicle in multiple first time periods during driving and the SOP of the first power battery of the first vehicle in multiple first time periods, and the first power battery is in a first state in the first time period.

[0124] That is, the first machine learning model is the first machine learning model in the model training method of the present invention.

[0125] When the current time period corresponding to the first SOP is the second time period, the operating parameters of the second vehicle in the first preset time period are input into the second machine learning model.

[0126] Here, the second machine learning model is trained based on the operating parameters of the first vehicle in multiple second time periods during driving and the SOP of the first power battery in multiple second time periods. The length of the second time period is different from the length of the first time period, and the first power battery is in a first state during the second time period.

[0127] That is, the second machine learning model is the second machine learning model in the training method of the model disclosed in the present invention.

[0128] In the above embodiment, according to the different time periods corresponding to the first SOP, the SOP of the updated second power battery can be more accurately estimated by using a machine learning model trained based on different time periods, thereby further improving the accuracy of the SOP of the updated power battery.

[0129] In some embodiments, when the moment when the voltage of the second power battery reaches the first preset protection voltage is the end point of the current time period corresponding to the first SOP, the first preset time period is the current time period corresponding to the first SOP.

[0130] In some embodiments, when the moment when the voltage of the second power battery reaches the first preset protection voltage is before the end of the current time period corresponding to the first SOP, the first preset time period is from the starting point of the current time period corresponding to the first SOP to the moment when the voltage of the second power battery reaches the first preset protection voltage.

[0131] For example, when the current time period corresponding to the first SOP is from the 1st second to the 6th second, if the voltage of the second power battery reaches the first preset protection voltage in the 6th second, the first preset time period is the current time period corresponding to the first SOP, that is, from the 1st second to the 6th second.

[0132] For another example, when the current time period corresponding to the first SOP is from the 1st second to the 6th second, if the voltage of the second power battery reaches the first preset protection voltage at the 3rd second, the first preset time period is equal to the 1st second to the 3rd second.

[0133] In the above embodiment, the duration of the first preset time period can be accurately determined through the temporal sequence between the moment when the voltage of the second power battery reaches the first preset protection voltage and the end point of the current time period corresponding to the first SOP, and then the operating parameters of the first preset time period of the second vehicle can be accurately obtained, which helps the machine learning model to more accurately estimate the SOP of the updated second power battery and further improve the accuracy of the SOP of the updated power battery.

[0134] In some embodiments, in response to the second vehicle satisfying a second update condition during driving, the operating parameters of the second vehicle for a second preset time period are obtained, and the operating parameters for the second preset time period are input into a machine learning model to estimate a fourth SOP corresponding to the current time period corresponding to the third SOP of the second power battery, and the third SOP is updated to the fourth SOP.

[0135] Here, the second update condition is: the second power battery of the second vehicle is in the second state, and the voltage of the second power battery reaches the second preset protection voltage.

[0136] The second state is the other of the discharge state and the charging state during energy feedback, the current SOP of the second power battery is the third SOP, and the duration of the second preset time period is less than or equal to the duration of the current time period corresponding to the third SOP.

[0137] As some implementations, if the second update condition is not met, the update operation is not performed. For example, if the voltage of the second power battery does not reach the second preset protection voltage, the update operation is not performed.

[0138] As some implementations, the second preset protection voltage is different from the first preset protection voltage. For example, the preset protection voltage in the discharge state is smaller than the preset protection voltage in the charge state.

[0139] That is, when the first state is a discharge state, the first preset protection voltage is less than the second preset protection voltage; when the first state is a charge state, the first preset protection voltage is greater than the second preset protection voltage.

[0140] In the above embodiment, no matter whether the second power battery of the second vehicle is in a discharge state or a charging state during energy feedback, when the voltage of the second power battery reaches the corresponding preset protection voltage, the dynamic changes of the operating parameters of the second vehicle during driving within a preset time period are utilized to determine the SOP corresponding to the preset time period through a machine learning model, so that when the SOP of the second power battery needs to be updated, the SOP of the updated second power battery can be accurately estimated, thereby improving the accuracy of the SOP of the updated power battery.

[0141] In some embodiments, when the current time period corresponding to the third SOP is the third time period, the operating parameters of the second vehicle in the second preset time period are input into the third machine learning model.

[0142] Here, the third machine learning model is trained based on the operating parameters of the first vehicle in multiple third time periods during driving and the SOP of the first power battery of the first vehicle in multiple third time periods, and the first power battery is in the second state in the third time period.

[0143] That is, the third machine learning model is the third machine learning model in the training method of the model disclosed in the present invention.

[0144] In some embodiments, when the current time period corresponding to the third SOP is the fourth time period, the operating parameters of the second preset time period are input into the fourth machine learning model.

[0145] Here, the fourth machine learning model is trained based on the operating parameters of the first vehicle in multiple fourth time periods during driving and the SOP of the first power battery in multiple fourth time periods. The length of the fourth time period is different from the length of the third time period, and the first power battery is in the second state during the fourth time period.

[0146] That is, the fourth machine learning model is the fourth machine learning model in the training method of the model disclosed in the present invention.

[0147] In the above embodiment, according to the different time periods corresponding to the third SOP, the SOP of the updated second power battery can be more accurately estimated by using a machine learning model trained based on different time periods, thereby further improving the accuracy of the SOP of the updated power battery.

[0148] In some embodiments, when the moment when the voltage of the second power battery reaches the second preset protection voltage is the end point of the current time period corresponding to the third SOP, the second preset time period is the current time period corresponding to the third SOP.

[0149] When the moment when the voltage of the second power battery reaches the second preset protection voltage is before the end of the current time period corresponding to the third SOP, the second preset time period is from the starting point of the current time period corresponding to the third SOP to the moment when the voltage of the second power battery reaches the second preset protection voltage.

[0150] It should be understood that the manner of determining the duration of the second preset time period is similar to the manner of determining the duration of the first preset time period.

[0151] In the above embodiment, the duration of the second preset time period can be accurately determined through the temporal sequence between the moment when the voltage of the second power battery reaches the second preset protection voltage and the end point of the current time period corresponding to the third SOP, and then the operating parameters of the second preset time period of the second vehicle can be accurately obtained, which helps the machine learning model to more accurately estimate the SOP of the updated second power battery and further improve the accuracy of the SOP of the updated power battery.

[0152] In some embodiments, the operating parameters of the first preset time period include at least one of the following four parameters: the SOC of the second power battery at the beginning of the first preset time period, the rate of change of the SOC of the second power battery in the first preset time period, the SOH of the second power battery in the first preset time period, and the mileage of the second vehicle in the first preset time period.

[0153] For example, the operating parameters of the first preset time period include any one, any two, any three or all of the above four parameters.

[0154] In this way, when estimating the SOP using a machine learning model, at least one of the above four parameters can be considered, which helps to further improve the accuracy of the SOP of the updated power battery estimated by the machine learning model.

[0155] In some embodiments, the operating parameters of the first preset time period include the mileage of the second vehicle within the first preset time period, and the operating parameters of the first preset time period also include at least one of the SOC of the second power battery at the beginning of each time period, the rate of change of the SOC of the second power battery in the first preset time period, and the SOH of the second power battery in the first preset time period.

[0156] For example, the operating parameters of the first preset time period include the mileage of the second vehicle within the first preset time period, and the operating parameters of the first preset time period also include any one, any two or all of the SOC of the second power battery at the beginning of the first preset time period, the rate of change of the SOC of the second power battery within the first preset time period, and the SOH of the second power battery within the first preset time period.

[0157] In this way, when estimating the SOP using a machine learning model, the mileage of the second vehicle within the first preset time period and other items related to the second power battery can be considered in parallel, which helps to further improve the accuracy of the SOP of the updated power battery estimated by the machine learning model.

[0158] In some embodiments, the operating parameters of the first preset time period further include at least one of the temperature of the second power battery at the beginning of the first preset time period and the rate of change of the temperature of the second power battery during the first preset time period.

[0159] For example, in addition to at least one of the above four parameters, the operating parameters of the first preset time period also include at least one of the temperature of the second power battery at the beginning of the first preset time period and the rate of change of the temperature of the second power battery in the first preset time period.

[0160] In this way, when estimating the SOP using the first machine learning model, more abundant operating parameters can be considered, which helps to further improve the accuracy of the SOP of the updated power battery estimated by the machine learning model.

[0161] In some embodiments, when the first state is a charging state during energy feedback, the operating parameters of the first preset time period also include at least one of a charging voltage of the second power battery at the beginning of the first preset time period and a rate of change of the charging voltage of the second power battery in the first preset time period.

[0162] For example, when the first state is a charging state during energy feedback, the operating parameters of the first preset time period include, in addition to at least one of the above four parameters, at least one of the charging voltage of the second power battery at the beginning of the first preset time period and the rate of change of the charging voltage of the second power battery in the first preset time period.

[0163] For another example, when the first state is the charging state during energy feedback, the operating parameters of the first preset time period include, in addition to at least one of the above four parameters, at least one of the temperature of the second power battery at the beginning of the first preset time period, and the rate of change of the temperature of the second power battery in the first preset time period, and at least one of the charging voltage of the second power battery at the beginning of the first preset time period and the rate of change of the charging voltage of the second power battery in the first preset time period.

[0164] In some embodiments, when the first state is a discharge state, the operating parameters of the first preset time period also include at least one of a discharge voltage of the second power battery at the beginning of the first preset time period and a rate of change of the discharge voltage of the second power battery in the first preset time period.

[0165] For example, when the first state is a discharge state, the operating parameters of the first preset time period include, in addition to at least one of the above four parameters, at least one of the discharge voltage of the second power battery at the beginning of the first preset time period and the rate of change of the discharge voltage of the second power battery in the first preset time period.

[0166] For another example, when the first state is the discharge state, the operating parameters of the first preset time period include, in addition to at least one of the above four parameters, at least one of the temperature of the second power battery at the beginning of the first preset time period and the rate of change of the temperature of the second power battery in the first preset time period, and at least one of the discharge voltage of the second power battery at the beginning of the first preset time period and the rate of change of the discharge voltage of the second power battery in the first preset time period.

[0167] In this way, when estimating the SOP using the first machine learning model, the type of the first state and the operating parameters corresponding to the first state can be considered, which helps to further improve the accuracy of the SOP of the updated power battery estimated by the machine learning model.

[0168] It should be understood that the operating parameters of the second preset time period are similar to the operating parameters of the first preset time period, and are not described in detail here.

[0169] Next, in combination with some embodiments of the present disclosure, it is introduced how to train a machine learning model and how to use the trained machine learning model to update the SOP of a power battery.

[0170] In step S1, data of multiple driving cycle segments of a first vehicle uploaded by a battery management system (BMS) is obtained from a data platform, and the data is cleaned. Here, the data of each driving cycle segment includes operating parameters of the first vehicle. For example, the data platform is located in the cloud.

[0171] As some embodiments, the data of the driving cycle segments may be divided according to the speed of the first vehicle. .here, Indicates the number of driving cycle segments. represents the data for the first driving cycle segment, The data of the second driving cycle segment is represented, and so on. For example, the period between two adjacent moments when the vehicle speed is 0 is a driving cycle segment.

[0172] As some embodiments, data cleaning includes removing driving cycle segments that have too many missing values ​​and / or too many outliers in the data. For example, the data of the driving cycle segment after data cleaning is , M is less than or equal to N.

[0173] In step S2, based on the data of the driving cycle segment after data cleaning , extracting the characteristic data segment of the SOP when the first power battery is discharged, and the characteristic data segment of the SOP when the first power battery is charged during energy feedback.

[0174] Next, the SOP of the first power battery during discharge is 5 seconds. Switch to a 30-second discharge , and then discharge for 30 seconds Switch to persistent Take this as an example for description.

[0175] As some embodiments, the data of the driving cycle segment after data cleaning can be In the embodiment, a characteristic data segment of the SOP of the first power battery during discharge is extracted.

[0176] For example, when extracting the discharge of the first power battery In the case of the characteristic data segment, the SOP of the first power battery can be changed from Switch to The moment (which can be called the switching moment) is recorded as , and Data within the last 5 seconds As feature data segment.

[0177] Here, the data of the driving cycle segment In, satisfied The characteristic data segment is recorded as ,in, To satisfy Number of feature data, Less than or equal to M. Indicates the first satisfaction The characteristic data segment, The second satisfaction characteristic data segment, and so on.

[0178] For example, when extracting the discharge of the first power battery In the case of the characteristic data segment, the SOP of the first power battery can be changed from Switch to The time is recorded as ,Pick Time to Data at the moment As feature data segment.

[0179] Here, the data of the driving cycle segment In, satisfied The characteristic data segment is recorded as ,in, To satisfy Number of feature data, Less than or equal to M. Indicates the first satisfaction The characteristic data segment, The second satisfaction characteristic data segment, and so on.

[0180] Next, the SOP of charging the first power battery is 5 seconds. Switch to 30-second charging , and then charge for 30 seconds Switch to persistent Take this as an example for description.

[0181] As some embodiments, the data of the driving cycle segment after data cleaning can be In the embodiment, the characteristic data segment of the SOP of charging during energy feedback of the first power battery is extracted.

[0182] For example, when extracting the charge of the first power battery In the case of the characteristic data segment, the SOP of the first power battery can be changed from Switch to The moment (which can be called the switching moment) is recorded as , and Data within the last 5 seconds As feature data segment.

[0183] Here, the data of the driving cycle segment In, satisfied The characteristic data segment is recorded as ,in, To satisfy Number of feature data, Less than or equal to M. Indicates the first satisfaction The characteristic data segment, The second satisfaction characteristic data segment, and so on.

[0184] For example, when extracting the charge of the first power battery In the case of the characteristic data segment, the SOP of the first power battery can be changed from Switch to The time is recorded as ,Pick Time to Data at the moment As feature data segment.

[0185] Here, the data of the driving cycle segment In, satisfied The characteristic data segment is recorded as ,in, To satisfy Number of feature data, Less than or equal to M. Indicates the first satisfaction The characteristic data segment, The second satisfaction characteristic data segment, and so on.

[0186] In step S3, a discharge SOP estimation model is trained based on the characteristic data segment of the SOP when the first power battery is discharged, and a charge SOP estimation model is trained based on the characteristic data segment of the SOP when the first power battery is charged.

[0187] As some embodiments, based on the discharge of the first power battery The characteristic data segment is used to determine the operating parameters for training the first discharge SOP estimation model. For example, the first discharge SOP estimation model is the first machine learning model mentioned above.

[0188] As other embodiments, based on the discharge of the first power battery The characteristic data segment is used to determine the operating parameters for training the second discharge SOP estimation model. For example, the second discharge SOP estimation model is the second machine learning model mentioned above.

[0189] As some embodiments, based on the charging of the first power battery The characteristic data segment determines the operating parameters used to train the first charging SOP estimation model. For example, the first charging SOP estimation model is the third machine learning model described above.

[0190] As other embodiments, based on the charging of the first power battery The characteristic data segment determines the operating parameters used to train the second charging SOP estimation model. For example, the second charging SOP estimation model is the fourth machine learning model described above.

[0191] It should be understood that the above has introduced how to determine the operating parameters for each time period and the specific process of training each machine learning model, which will not be repeated here.

[0192] In step S4 , the SOP of the second power battery is dynamically updated in response to the first update condition and the second update condition.

[0193] Next, the SOP of the second power battery during discharge is 5 seconds. Switch to a 30-second discharge , and then discharge for 30 seconds Switch to persistent , and the first state is a discharge state and the first preset protection voltage is 2.5V. That is, the first update condition is that the second power battery of the second vehicle is in a discharge state and the voltage of the second power battery reaches 2.5V.

[0194] As some embodiments, the second power battery is discharged at (i.e., the first SOP) The moment when the discharge reaches 2.5V is recorded as , and Data within the last 5 seconds As the first SOP correction feature data segment. Here, 5 seconds before The time period is the first preset time period.

[0195] For example, you can extract The operating parameters of the second vehicle are obtained, and the operating parameters are used as input to the first discharge SOP estimation model (i.e., the first machine learning model) to estimate the second SOP.

[0196] As some embodiments, the second power battery is discharged at (i.e., the first SOP) The moment when the discharge reaches 2.5V is recorded as , and arrive Data As the first SOP correction feature data segment. Here, arrive The time period is the first preset time period.

[0197] For example, you can extract The operating parameters of the second vehicle are obtained and the operating parameters are used as input to a second discharge SOP estimation model (i.e., a second machine learning model) to estimate the second SOP.

[0198] Next, the SOP of charging the second power battery is 5 seconds. Switch to 30-second charging , and then charge for 30 seconds Switch to persistent , and the second state is a charging state and the second preset protection voltage is 3.7V. That is, the second update condition is that the second power battery of the second vehicle is in a charging state and the voltage of the second power battery reaches 3.7V.

[0199] As some embodiments, the second power battery is charged at (i.e., the third SOP) The moment when the charge reaches 3.7V is recorded as , and Data within the last 5 seconds As the modified feature data segment of the third SOP. Here, 5 seconds before The time period is the second preset time period.

[0200] For example, you can extract The operating parameters of the second vehicle are obtained, and the operating parameters are used as input to the first charging SOP estimation model (i.e., the third machine learning model) to estimate the fourth SOP.

[0201] As some embodiments, the second power battery is discharged at (i.e., the third SOP) The moment when the discharge reaches 2.5V is recorded as , and arrive Data As the modified feature data segment of the third SOP. Here, arrive The time period is the second preset time period.

[0202] For example, you can extract The operating parameters of the second vehicle are obtained and the operating parameters are used as input to a second charging SOP estimation model (i.e., the fourth machine learning model) to estimate the second SOP.

[0203] In the above embodiment, regardless of whether the vehicle's power battery is in a discharge state or a charging state during energy feedback, a machine learning model can be trained to capture the relationship between the dynamic changes of operating parameters in a corresponding time period and the SOP of the time period, so that when using the trained machine learning model, the updated SOP of the power battery can be estimated more accurately, thereby accurately updating the SOP of the power battery and protecting the power battery. In addition, estimating the SOP of the power battery using a machine learning model can improve the efficiency of estimating the SOP of the power battery, which helps to update the SOP of the power battery in a timely manner.

[0204] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0205] In some embodiments, the model training device includes a module for executing the model training method of any one of the above embodiments.

[0206] Figure 3 It is a schematic diagram of the structure of a training device for a model according to some embodiments of the present disclosure.

[0207] like Figure 3 As shown, the model training device includes a first acquisition module 301, a second acquisition module 302 and a training module 303.

[0208] The first acquisition module 301 is configured to acquire operating parameters of a plurality of time periods during the driving process of the first vehicle. Here, the plurality of time periods include a first group of time periods, the first group of time periods includes a plurality of first time periods of the same length, and the first power battery of the first vehicle is in a first state during the first group of time periods, the first state being one of a discharge state and a charge state during energy feedback.

[0209] The second acquisition module 302 is configured to acquire the SOP of the first power battery in each first time period. Here, the SOP of the first power battery in each first time period makes the first power battery fail to reach the first preset protection voltage in each first time period.

[0210] The training module 303 is configured to train the first machine learning model by taking the operating parameters of the first vehicle in each first time period as input and the SOP of the first power battery in each first time period as output.

[0211] In some embodiments, the model training device may further include other modules to execute the model training method of any of the above embodiments.

[0212] In some embodiments, the apparatus for updating the power status of a battery includes a module for executing the method for updating the power status of a battery of any one of the above embodiments.

[0213] Figure 4 4 is a schematic diagram of the structure of a device for updating the power status of a battery according to some embodiments of the present disclosure.

[0214] like Figure 4 As shown, the device for updating the power state of a battery includes an acquisition module 401 and an estimation module 402 .

[0215] The acquisition module 401 is configured to: in response to the second vehicle satisfying the first update condition during driving, acquire the operating parameters of the first preset time period of the second vehicle. Here, the first update condition is: the second power battery of the second vehicle is in the first state, and the voltage of the second power battery reaches the first preset protection voltage, the first state is one of the discharge state and the charging state during energy feedback, the current SOP of the second power battery is the first SOP, and the duration of the first preset time period is less than or equal to the duration of the current time period corresponding to the first SOP.

[0216] The estimation module 402 is configured to input the operating parameters of the first preset time period into the machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and update the first SOP to the second SOP.

[0217] In some embodiments, the apparatus for updating the power state of a battery may further include other modules to execute the method for updating the power state of a battery in any of the above embodiments.

[0218] Figure 5 is a schematic diagram of the structure of an electronic device according to some embodiments of the present disclosure.

[0219] like Figure 5 As shown, the electronic device 500 includes a memory 501 and a processor 502 coupled to the memory 501, and the processor 502 is configured to execute the method of any of the aforementioned embodiments based on the instructions stored in the memory 501. As some implementation methods, the processor 502 is configured to execute the model training method or the battery power state update method of any of the aforementioned embodiments based on the instructions stored in the memory 501.

[0220] The memory 501 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.

[0221] In some embodiments, the electronic device 500 may further include an input / output interface 503, a network interface 504, a storage interface 505, etc. The input / output interface 503, the network interface 504, the storage interface 505, and the memory 501 and the processor 502 may be connected, for example, via a bus 506. The input / output interface 503 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 504 provides a connection interface for various networked devices. The storage interface 505 provides a connection interface for external storage devices such as an SD card and a USB flash drive.

[0222] The embodiment of the present disclosure further provides a computer-readable storage medium, including computer program instructions, which implement the steps of the method of any one of the above embodiments when executed by a processor.

[0223] The embodiments of the present disclosure further provide a computer program product, including a computer program, which implements the steps of the method of any one of the above embodiments when executed by a processor.

[0224] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0225] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0226] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present disclosure. It should be understood that the functions specified in one or more processes in the flowchart and / or one or more blocks in the block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the functions specified in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0227] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0228] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0229] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A model training method, comprising: Acquiring operating parameters of a first vehicle in a plurality of time periods during driving, the plurality of time periods comprising a first group of time periods, the first group of time periods comprising a plurality of first time periods of the same duration, wherein a first power battery of the first vehicle is in a first state in the first group of time periods, the first state being one of a discharge state and a charge state during energy feedback; Acquire a power state SOP of the first power battery in each first time period, wherein the SOP of the first power battery in each first time period makes the first power battery fail to reach a first preset protection voltage in each first time period; and The first machine learning model is trained by taking the operating parameters of the first vehicle in each first time period as input and the SOP of the first power battery in each first time period as output.

2. The method according to claim 1, wherein: The first group of time periods further includes a plurality of second time periods of the same duration, wherein the durations of the plurality of second time periods are different from the durations of the plurality of first time periods, and the method further includes: acquiring the SOP of the first power battery in each second time period, wherein the SOP of the first power battery in each second time period makes the first power battery fail to reach the first preset protection voltage in each second time period; and The second machine learning model is trained by taking the operating parameters of the first vehicle in each second time period as input and the SOP of the first power battery in each second time period as output.

3. The method according to claim 1 or 2, wherein: The multiple time periods further include a second group of time periods, the second group of time periods includes a plurality of third time periods, wherein the first power battery is in a second state in the second group of time periods, the durations of the plurality of third time periods are the same as the durations of the plurality of first time periods, and the second state is another of the discharge state and the charge state, and the method further includes: acquiring the SOP of the first power battery in each third time period, wherein the SOP of the first power battery in each third time period makes the first power battery fail to reach the second preset protection voltage in each third time period; and The third machine learning model is trained by taking the operating parameters of the first vehicle in each third time period as input and the SOP of the first power battery in each third time period as output.

4. The method according to claim 3, wherein: The second group of time periods further includes a plurality of fourth time periods of the same duration, wherein the durations of the plurality of fourth time periods are different from the durations of the plurality of third time periods, and the method further includes: acquiring the SOP of the first power battery in each fourth time period, wherein the SOP of the first power battery in each fourth time period makes the first power battery not reach the second preset protection voltage in each fourth time period; and The fourth machine learning model is trained by taking the operating parameters of the first vehicle in each fourth time period as input and the SOP of the first power battery in each fourth time period as output.

5. The method according to claim 1, wherein: The operating parameters of each time period include at least one of the following: The state of charge (SOC) of the first power battery at the beginning of each time period; The rate of change of the SOC of the first power battery in each time period; The health state SOH of the first power battery in each time period; and The mileage of the first vehicle in each time period.

6. The method according to claim 5, wherein: The operating parameters of each time period include the mileage of the first vehicle in each time period and at least one of the following: The SOC of the first power battery at the beginning of each time period; a change rate of the SOC of the first power battery in each time period; and The SOH of the first power battery in each time period.

7. The method according to claim 5, wherein: The operating parameters of each time period also include at least one of the following: The temperature of the first power battery at the beginning of each time period; and The rate of change of the temperature of the first power battery in each time period.

8. The method according to any one of claims 5 to 7, wherein: In the case where the first state is the charging state, the operating parameter of each time period further includes at least one of a charging voltage of the first power battery at the beginning of each time period and a rate of change of the charging voltage of the first power battery in each time period; When the first state is the discharge state, the operating parameter of each time period further includes at least one of a discharge voltage of the first power battery at the beginning of each time period and a change rate of the discharge voltage of the first power battery in each time period.

9. The method according to claim 1 or 2, wherein: The start and end points of each time period in the first group of time periods are the moments when the SOP of the first vehicle changes, wherein: In a case where the first state is the discharging state, each time period in the first group of time periods is between a time when the first vehicle starts to a time when the first vehicle starts to brake; and In a case where the first state is the charging state, each time period in the first group of time periods is between a time when the first vehicle starts braking and a time when the first vehicle ends braking.

10. A method for updating a power state of a battery, comprising: In response to the second vehicle satisfying a first update condition during driving, obtaining operating parameters of a first preset time period of the second vehicle, wherein the first update condition is: a second power battery of the second vehicle is in a first state, and a voltage of the second power battery reaches a first preset protection voltage, wherein the first state is one of a discharge state and a charging state during energy feedback, a current power state SOP of the second power battery is a first SOP, and a duration of the first preset time period is less than or equal to a duration of a current time period corresponding to the first SOP; and The operating parameters of the first preset time period are input into the machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and the first SOP is updated to the second SOP.

11. The method according to claim 10, wherein: The step of inputting the operating parameters of the first preset time period into a machine learning model to estimate the second SOP of the second power battery corresponding to the current time period corresponding to the first SOP, and updating the first SOP to the second SOP includes: In a case where the current time period corresponding to the first SOP is the first time period, the operating parameters of the second vehicle in the first preset time period are input into a first machine learning model, wherein the first machine learning model is trained based on the operating parameters of the first vehicle in a plurality of the first time periods during driving and the SOP of the first power battery of the first vehicle in a plurality of the first time periods, and the first power battery is in the first state in the first time period; When the current time period corresponding to the first SOP is the second time period, the operating parameters of the second vehicle in the first preset time period are input into the second machine learning model, wherein the second machine learning model is trained based on the operating parameters of the first vehicle in multiple second time periods during driving and the SOP of the first power battery in multiple second time periods, the length of the second time period is different from the length of the first time period, and the first power battery is in the first state during the second time period.

12. The method according to claim 10 or 11, wherein: In the case where the moment when the voltage of the second power battery reaches the first preset protection voltage is the end point of the current time period corresponding to the first SOP, the first preset time period is the current time period corresponding to the first SOP; When the moment when the voltage of the second power battery reaches the first preset protection voltage is before the end of the current time period corresponding to the first SOP, the first preset time period is from the starting point of the current time period corresponding to the first SOP to the moment when the voltage of the second power battery reaches the first preset protection voltage.

13. The method according to claim 10 or 11, further comprising: In response to the second vehicle satisfying a second update condition during driving, obtaining operating parameters of a second preset time period of the second vehicle, wherein the second update condition is: a second power battery of the second vehicle is in a second state, and a voltage of the second power battery reaches a second preset protection voltage, wherein the second state is the other of the discharge state and the charge state, the current power state SOP of the second power battery is a third SOP, and the duration of the second preset time period is less than or equal to the duration of a current time period corresponding to the third SOP; and The operating parameters of the second preset time period are input into the machine learning model to estimate the fourth SOP of the second power battery corresponding to the current time period corresponding to the third SOP, and the third SOP is updated to the fourth SOP.

14. The method according to claim 13, wherein: The step of inputting the operating parameters of the second preset time period into a machine learning model to estimate the fourth SOP of the second power battery corresponding to the current time period corresponding to the third SOP, and updating the third SOP to the fourth SOP includes: In a case where the current time period corresponding to the third SOP is the third time period, the operating parameters of the second vehicle in the second preset time period are input into a third machine learning model, wherein the third machine learning model is trained based on the operating parameters of the first vehicle in a plurality of the third time periods during driving and the SOP of the first power battery of the first vehicle in a plurality of the third time periods, and the first power battery is in the second state in the third time period; When the current time period corresponding to the third SOP is the fourth time period, the operating parameters of the second preset time period are input into the fourth machine learning model, wherein the fourth machine learning model is trained based on the operating parameters of multiple fourth time periods of the first vehicle during driving and the SOP of the first power battery in multiple fourth time periods, the length of the fourth time period is different from the length of the third time period, and the first power battery is in the second state during the fourth time period.

15. The method of claim 13, wherein: In the case where the moment when the voltage of the second power battery reaches the second preset protection voltage is the end point of the current time period corresponding to the third SOP, the second preset time period is the current time period corresponding to the third SOP; When the moment when the voltage of the second power battery reaches the second preset protection voltage is before the end of the current time period corresponding to the third SOP, the second preset time period is from the starting point of the current time period corresponding to the third SOP to the moment when the voltage of the second power battery reaches the second preset protection voltage.

16. The method according to claim 10, wherein: The operating parameters of the first preset time period include at least one of the following: The state of charge (SOC) of the second power battery at the beginning of the first preset time period; a change rate of the SOC of the second power battery in the first preset time period; The health state SOH of the second power battery within the first preset time period; as well as The mileage of the second vehicle within the first preset time period.

17. The method according to claim 16, wherein: The operating parameters of the first preset time period include the mileage of the second vehicle in the first preset time period and at least one of the following: The SOC of the second power battery at the beginning of the first preset time period; a change rate of the SOC of the second power battery in the first preset time period; and The SOH of the second power battery in the first preset time period.

18. The method according to claim 16, wherein: The operating parameters of the first preset time period also include at least one of the following: the temperature of the second power battery at the beginning of the first preset time period; and The rate of change of the temperature of the second power battery in the first preset time period.

19. The method according to any one of claims 16 to 18, wherein: When the first state is the charging state, the operating parameter of the first preset time period further includes at least one of a charging voltage of the second power battery at the beginning of the first preset time period and a rate of change of the charging voltage of the second power battery in the first preset time period; When the first state is the discharge state, the operating parameters of the first preset time period also include at least one of a discharge voltage of the second power battery at the beginning of the first preset time period and a rate of change of the discharge voltage of the second power battery in the first preset time period.

20. A model training device, comprising: A module configured to execute the method according to any one of claims 1 to 9.

21. A device for updating the power status of a battery, comprising: A module configured to perform the method according to any one of claims 10 to 19.

22. An electronic device, comprising: Memory; as well as A processor coupled to the memory, configured to execute the method of any one of claims 1-19 based on instructions stored in the memory.

23. A computer-readable storage medium comprising a computer program, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 19 are implemented.

24. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 19 are implemented.

Images