Lithium battery charging voltage balance control method and system

Abstract

The invention provides a lithium battery charging voltage equalization control method and system, and the method comprises the steps: determining a first charging mode of each single lithium battery at a current stage and a second charging mode of each single lithium battery at a next stage based on the monitoring of the charging characteristic parameters of the single lithium batteries in a lithium battery pack in a charging process; when the difference value between the first time when the single lithium battery is switched from the first charging mode to the second charging mode and the current time is smaller than or equal to a preset time difference threshold value, generating a first time data sequence of the charging characteristic parameters, predicting a second time when each single lithium battery is switched from the first charging mode to a second charging mode based on the first time data sequence; and adjusting the charging voltage of the single lithium batteries in the lithium battery pack to align the second time when each single lithium battery in the lithium battery pack is switched from the first charging mode to the second charging mode, so that the problem of overcharging caused by inconsistent capacities of the single batteries in the lithium battery pack can be effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a lithium battery charging voltage balancing control method and system. Background Art

[0002] Lithium batteries are widely used in power batteries for new energy vehicles due to their many advantages such as no memory effect, high energy density and low self-discharge rate. Since single lithium batteries are not enough to provide sufficient output power, lithium batteries are usually used to form power batteries in the form of battery packs. However, due to the manufacturing process of single cells, individual differences in the production and processing of lithium batteries, and the unique chemical properties of lithium batteries themselves, the power of single lithium batteries in lithium battery packs will be inconsistent. The inconsistency of single cells in the battery pack will increase with the increase of usage time. Among them, batteries with smaller capacity will frequently be overcharged and over-discharged. Due to their large internal resistance, frequent overcharge and over-discharge will increase the risk of thermal runaway of the battery. Therefore, people have studied a large number of technologies for realizing battery balancing control, such as passive balancing control technology such as connecting a controllable shunt resistor in parallel at both ends of each single cell to reduce the charging speed of the battery, or active balancing control technology that uses capacitors and inductors as energy transfer media to transfer the power in single cells with excessively high voltage to single cells with lower voltage through energy transfer media. However, the above-mentioned passive balancing control technology and active balancing control technology can only alleviate the impact of the inconsistency of single cells to a certain extent, but the inconsistency problem of single cells still exists, and the risk of thermal runaway caused by the inconsistency of single cells cannot be completely avoided. When charging the battery pack, the inconsistency of the single cell capacity will still cause the problem of overcharging of some low-capacity single cells. Summary of the invention

[0003] Based on the above problems, the present invention proposes a lithium battery charging voltage balancing control method and system, which can effectively solve the overcharging problem caused by inconsistent capacity of single cells in a lithium battery pack.

[0004] In view of this, a first aspect of the present invention provides a lithium battery charging voltage balancing control method, comprising: During the charging process, the charging characteristic parameter of each single lithium battery in the lithium battery pack is monitored, wherein the charging characteristic parameter is a characteristic parameter or a combination of multiple characteristic parameters reflecting the charging state of the single lithium battery; Determine a first charging mode of the current stage and a second charging mode of the next stage for each single lithium battery based on the charging characteristic parameters; estimating, based on empirical data, a first time for each single lithium battery to switch from the first charging mode to the second charging mode; When the difference between the first time and the current time is less than or equal to a preset time difference threshold, a first time data sequence of the charging characteristic parameter is generated, wherein the first time data sequence is a data sequence obtained by arranging discrete values ​​of the charging characteristic parameter of the single lithium battery in time sequence during the charging process; Predicting a second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to align the second time when each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode.

[0005] Furthermore, the step of estimating the first time for each single lithium battery to switch from the first charging mode to the second charging mode according to empirical data specifically includes: Using historical charging data of a single lithium battery to fit a first charging characteristic function of its charging characteristic parameters relative to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Determine a first target charging characteristic parameter corresponding to a first parameter threshold for switching from the first charging mode to the second charging mode; Obtaining a current value of a first target charging characteristic parameter of each single lithium battery; Determine a first function time of a first parameter threshold of the first target charging characteristic parameter in the first charging characteristic function and a second function time of a current value of the first target charging characteristic parameter in the first charging characteristic function; Calculate the time difference between the first function time and the second function time; A time obtained by summing the current time and the time difference is determined as the first time.

[0006] Furthermore, the step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter specifically includes: Converting the first time data series into input data of a data series prediction model, wherein the data series prediction model is a recurrent neural network model trained based on historical charging data of single lithium batteries; Inputting the input data into the data sequence prediction model to obtain output data; Converting the output data into a second time data series; The second time when the single lithium battery switches from the first charging mode to the second charging mode is calculated based on the second time data series.

[0007] Furthermore, the step of converting the first time data series into input data of a data series prediction model specifically includes: Determine the largest data element in the first time data series: , and the minimum data element: , in 1 to A positive integer between is the first time data series data elements, is the number of data elements in the first time data series; According to the maximum data element and the minimum data element Calculate the normalization function of the first time data series: ; Normalizing the first time data series based on a normalization function to obtain a normalized data series; Add at the end of the normalized data sequence The input data is obtained by filling the number of The length of the pre-configured input data.

[0008] Furthermore, the step of converting the output data into a second time data series specifically includes: Get each data element in the output data ,in 1 to A positive integer between is the number of data elements in the output data; Based on the output data data element calculates the first Data elements: ; Calculate the starting time of the second time data series and end time ; Based on starting time and end time Generate a corresponding time for each data element in the second time data series.

[0009] Furthermore, the step of calculating the second time for the single lithium battery to switch from the first charging mode to the second charging mode based on the second time data series specifically includes: Fitting a second charging characteristic function of the data elements in the second time data series with respect to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Calculating the charging characteristic value as the target time corresponding to the first parameter threshold by the second charging characteristic function; The target time is determined as the second time.

[0010] Furthermore, before the step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter, the method further includes: Determine a minimum input sequence length and a minimum output sequence length ; Each historical charging data of a single lithium battery is input with the minimum input sequence length The minimum output sequence length is used as a reference for segmentation and conversion into input data samples and output data samples; Using the input data samples and the output data samples to construct a training data set and a validation data set for the data sequence prediction model; The data sequence prediction model is trained using the training data set and the validation data set.

[0011] Furthermore, the steps of dividing each historical charging data of the single lithium battery based on the minimum input sequence length and the maximum input sequence length and converting them into input data samples and output data samples specifically include: Get the input data length of the pre-configured data sequence prediction model And output data sequence ; The historical charging data is divided into A temporary sample dataset; Configure a count range from 1 to Count variable , so that each temporary sample data set has a corresponding value as its dataset number, and for the number Temporary sample data set, where the length of each historical charging data is satisfy: ; For the number The following processing steps are performed for each historical charging data in the temporary sample data set: Sequentially number each data element in the historical charging data; Extract data elements whose numbers are less than or equal to the historical charging data. Generates a sequence of input data elements from the data elements of Extract data elements with numbers greater than The data elements generate an output data element sequence; Add at the end of the input data element sequence The input data sample is obtained by filling the number of paddings; Add at the end of the output data element sequence The output data samples are obtained by filling the number of samples.

[0012] Furthermore, the step of adjusting the charging voltage of the single lithium battery in the lithium battery pack to align the second time at which each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode specifically includes: Get the current charging mode; Determine a third time corresponding to the first charging mode based on the second time of each single lithium battery in the lithium battery pack; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to adjust the switching time of each single lithium battery from the first charging mode to the second charging mode to the third time.

[0013] The second aspect of the present invention proposes a lithium battery charging voltage balancing control system, comprising a charging characteristic parameter monitoring unit for obtaining the charging characteristic parameters of each single lithium battery in a lithium battery pack, a charging voltage regulating unit for adjusting the charging voltage of each single lithium battery in the lithium battery pack, and a control unit connected to the charging characteristic parameter monitoring unit and the charging voltage regulating unit, wherein the control unit is configured to implement the lithium battery charging voltage balancing control method described in any one of the first aspect of the present invention.

[0014] The present invention proposes a lithium battery charging voltage balancing control method and system, which determines the first charging mode of each single lithium battery in the current stage and the second charging mode of the next stage based on monitoring the charging characteristic parameters of the single lithium batteries in the lithium battery group during the charging process; when the difference between the first time when the single lithium battery switches from the first charging mode to the second charging mode and the current time is less than or equal to a preset time difference threshold, generates a first time data sequence of the charging characteristic parameters, and predicts the second time when each single lithium battery switches from the first charging mode to the second charging mode based on the first time data sequence; adjusts the charging voltage of the single lithium batteries in the lithium battery group to align the second time when each single lithium battery in the lithium battery group switches from the first charging mode to the second charging mode, which can effectively solve the overcharging problem caused by inconsistent capacity of the single lithium batteries in the lithium battery group. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a lithium battery charging voltage equalization control method provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a lithium battery charging voltage equalization control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0017] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0018] In the description of the present invention, the term "multiple" refers to two or more. Unless otherwise clearly defined, the orientation or positional relationship indicated by the terms "upper", "lower", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation of the present invention. The terms "connection", "installation", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second", etc. can explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "multiple" is two or more.

[0019] In the description of this specification, the description of the terms "one embodiment", "some implementations", "specific embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0020] A lithium battery charging voltage equalization control method and system according to some embodiments of the present invention will be described below with reference to the accompanying drawings.

[0021] like Figure 1 As shown, the first aspect of the present invention provides a lithium battery charging voltage balancing control method, comprising: During the charging process, the charging characteristic parameter of each single lithium battery in the lithium battery pack is monitored, wherein the charging characteristic parameter is a characteristic parameter or a combination of multiple characteristic parameters reflecting the charging state of the single lithium battery; Determine a first charging mode of the current stage and a second charging mode of the next stage for each single lithium battery based on the charging characteristic parameters; estimating, based on empirical data, a first time for each single lithium battery to switch from the first charging mode to the second charging mode; When the difference between the first time and the current time is less than or equal to a preset time difference threshold, a first time data sequence of the charging characteristic parameter is generated, wherein the first time data sequence is a data sequence obtained by arranging discrete values ​​of the charging characteristic parameter of the single lithium battery in time sequence during the charging process; Predicting a second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to align the second time when each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode.

[0022] Specifically, the charging characteristic parameters include but are not limited to the real-time battery voltage, charging voltage, charging current, charging power, etc. of the single lithium battery at each moment during the charging process. In the step of monitoring the charging characteristic parameters of each single lithium battery, the discrete values ​​of the charging characteristic parameters are collected and recorded at a preset sampling frequency, and stored in association with the collection time. The first time data sequence includes each discrete value of the charging characteristic parameter and the collection time corresponding to each discrete value. When there are multiple charging characteristic parameters, a first time data sequence is generated for each charging characteristic parameter.

[0023] Furthermore, the charging characteristic parameter also includes the real-time temperature of the single lithium battery, and the real-time temperature is used to dynamically configure the threshold value of the charging characteristic parameter of the single lithium battery for switching between different charging modes.

[0024] The charging modes include trickle pre-charging mode, constant current charging mode, constant voltage charging mode and trickle floating charging mode of the lithium battery, and the corresponding charging stages are pre-charging stage, constant current stage, constant voltage stage and floating charging stage respectively. The lithium battery switches sequentially in these charging stages according to different charging characteristic parameters. The current stage and the next stage are two consecutive charging stages among the above-mentioned charging stages. Correspondingly, the first charging mode and the second charging mode are two consecutive charging modes among the above-mentioned charging modes.

[0025] Furthermore, before the step of estimating the first time for each single lithium battery to switch from the first charging mode to the second charging mode according to empirical data, the method further includes: A time monitoring period and a time difference threshold for performing charging voltage equalization control; After the step of estimating the first time for each single lithium battery to switch from the first charging mode to the second charging mode according to empirical data, the following steps are performed with the time monitoring period as the execution period: Get the current system time as the current time; Calculating the difference between the first time and the current time; Determine whether the difference between the first time and the current time is greater than a preset time difference threshold.

[0026] Further, after the step of determining whether the difference between the first time and the current time is greater than a preset time difference threshold, the method further includes: When the difference between the first time and the current time is greater than a preset time difference threshold, wait for the next time monitoring cycle to cyclically execute the above steps.

[0027] Furthermore, the step of estimating the first time for each single lithium battery to switch from the first charging mode to the second charging mode according to empirical data specifically includes: Using historical charging data of a single lithium battery to fit a first charging characteristic function of its charging characteristic parameters relative to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Determine a first target charging characteristic parameter corresponding to a first parameter threshold for switching from the first charging mode to the second charging mode; Obtaining a current value of a first target charging characteristic parameter of each single lithium battery; Determine a first function time of a first parameter threshold of the first target charging characteristic parameter in the first charging characteristic function and a second function time of a current value of the first target charging characteristic parameter in the first charging characteristic function; Calculate the time difference between the first function time and the second function time; A time obtained by summing the current time and the time difference is determined as the first time.

[0028] Specifically, the empirical data refers to the historical charging data of the single lithium battery in the lithium battery pack. The historical charging data of the single lithium battery is the discrete value of the charging characteristic parameters of one or more single lithium batteries in the lithium battery pack that are collected at a preset sampling frequency during the historical charging process, and the first charging characteristic function is a curve function obtained by fitting using a function fitting algorithm such as least squares, polynomial fitting or neural network fitting. Similarly, when there are multiple charging characteristic parameters, the first charging characteristic function of each charging characteristic parameter is fitted accordingly, that is, each charging characteristic parameter corresponds to the first charging characteristic function one by one. More specifically, a typical charging data of a single lithium battery in the historical charging data can be selected to fit the first charging characteristic function, or multiple historical charging data of several single lithium batteries in the lithium battery pack can be integrated and processed, including calculating the mean after filtering the abnormal data, and then fitting the function using a fitting algorithm.

[0029] In the technical solution of the above-mentioned embodiment, the first charging characteristic function is fitted with time as the independent variable and the charging characteristic parameter as the dependent variable. The first charging characteristic function uses the time experienced by a complete charging process as the time axis to present the change of the value of the charging characteristic parameter during the charging process. In order to distinguish the aforementioned first time and second time, the present invention refers to the time point on the time axis of the first charging characteristic function as the function time, that is, the first function time and the second function time are the time on the time axis of the first charging characteristic function. For the convenience of calculation, the time axis of the first charging characteristic function starts counting from 0, and the value of any charging characteristic parameter in the first charging characteristic function corresponds to the only time point on the time axis that is greater than or equal to 0. The aforementioned first time and second time are natural time, that is, they are usually presented in the form of year, month, day, hour, minute, and second, or the time format of the first time and second time is converted into an absolute time presented in milliseconds.

[0030] Furthermore, the step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter specifically includes: Converting the first time data series into input data of a data series prediction model, wherein the data series prediction model is a recurrent neural network model trained based on historical charging data of single lithium batteries; Inputting the input data into the data sequence prediction model to obtain output data; Converting the output data into a second time data series; The second time when the single lithium battery switches from the first charging mode to the second charging mode is calculated based on the second time data series.

[0031] Specifically, the recurrent neural network model is a neural network specially used to process sequence data. The core structure of the recurrent neural network model is a recurrent unit, which allows information to be circulated in the neural network. In the first-time data sequence prediction, the input of each iterative step is combined with the hidden state of the previous iterative step, so that the historical information in the sequence can be remembered. Therefore, the recurrent neural network model has a very good performance in predicting the charging characteristic parameter data sequence of lithium batteries.

[0032] In the technical solution of the present invention, the input data of the data sequence prediction model is a data sequence of a specific length, and the sequence length is greater than the first time data sequence.

[0033] In the technical solutions of some other embodiments of the present invention, the step of converting the first time data series into input data of the data series prediction model specifically includes: Preprocessing the first time data series, wherein the preprocessing includes outlier correction or deletion and data standardization; Get the preconfigured input data length and the length of the first time data series ; Calculate the input data length and the length of the first time data series The difference: ; At the end of the first time data series, add The input data of the data sequence prediction model is obtained by filling in the number of paddings.

[0034] The data standardization process refers to converting the data into distribution data with a mean of 0 and a standard deviation of 1.

[0035] Furthermore, the step of converting the first time data series into input data of a data series prediction model specifically includes: Determine the largest data element in the first time data series: , and the minimum data element: , in 1 to A positive integer between is the first time data series data elements, is the number of data elements in the first time data series; According to the maximum data element and the minimum data element Calculate the normalization function of the first time data series: ; Normalizing the first time data series based on a normalization function to obtain a normalized data series; Add at the end of the normalized data sequence The input data is obtained by filling the number of The length of the pre-configured input data.

[0036] In the technical solution of the above embodiment, the data normalization function is used to map each data element in the first time data series to the interval [0,1]. It should be known that the input data length and the length of the first time data series The fill number refers to the number of data elements in the input data and the first time data sequence, respectively. The fill number may be a pre-configured specific value, and the number 0 is usually used as the fill number.

[0037] Furthermore, the step of converting the output data into a second time data series specifically includes: Get each data element in the output data ,in 1 to A positive integer between is the number of data elements in the output data; Based on the output data data element calculates the first Data elements: ; Calculate the starting time of the second time data series and end time ; Based on starting time and end time Generate a corresponding time for each data element in the second time data series.

[0038] Specifically, the numerical elements in the output data are all values ​​in the interval [0,1]. The length of the second time data series is the same as that of the output data, both of which are .

[0039] Furthermore, the step of calculating the start time and the end time of the second time data series specifically includes: Get the end time of the first time data series and the collection time interval between two adjacent data elements in the first time data sequence ,in For 1 to Any value between satisfy: , in is the first time data series Data elements The collection time; Calculate the starting time of the second time data series: ; Calculate the end time of the second time data series: .

[0040] Furthermore, the step of calculating the second time for the single lithium battery to switch from the first charging mode to the second charging mode based on the second time data series specifically includes: Fitting a second charging characteristic function of the data elements in the second time data series with respect to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Calculating the charging characteristic value as the target time corresponding to the first parameter threshold by the second charging characteristic function; The target time is determined as the second time.

[0041] Similarly, the second charging characteristic function is a curve function obtained by fitting using a function fitting algorithm such as least squares method, polynomial fitting or neural network fitting, with time as the independent variable and the charging characteristic parameter as the dependent variable. In the step of calculating the charging characteristic value as the target time corresponding to the first parameter threshold value through the second charging characteristic function, an inverse function corresponding to the second charging characteristic function is first generated, and then the first parameter threshold value is input into the inverse function to calculate the corresponding target time.

[0042] Furthermore, before the step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter, the method further includes: Determine a minimum input sequence length and a minimum output sequence length ; Each historical charging data of a single lithium battery is input with the minimum input sequence length The minimum output sequence length is used as a reference for segmentation and conversion into input data samples and output data samples; Using the input data samples and the output data samples to construct a training data set and a validation data set for the data sequence prediction model; The data sequence prediction model is trained using the training data set and the validation data set.

[0043] Specifically, since the duration of multiple charges of the same single lithium battery or different single lithium batteries is not exactly the same, the historical charging data collected and recorded using the same sampling frequency often have different sequence lengths, that is, the number of data elements in the historical charging data is not a fixed value. At the same time, since the "current time" for generating the first time data series may be within different time ranges in any charging stage, the length of the first time data series and the length of the second time data series to be predicted are not fixed values. The lengths of the input data and output data of the recurrent neural network model are usually fixed. In this case, it is necessary to construct input data samples and output data samples corresponding to first time data sequences of different lengths to train the data sequence prediction model so that the data sequence prediction model has the ability to perform data sequence prediction on the input data converted from first time data sequences of different lengths.

[0044] Furthermore, the steps of dividing each historical charging data of the single lithium battery based on the minimum input sequence length and the maximum input sequence length and converting them into input data samples and output data samples specifically include: Get the input data length of the pre-configured data sequence prediction model And output data sequence ; The historical charging data is divided into A temporary sample dataset; Configure a count range from 1 to Count variable , so that each temporary sample data set has a corresponding value as its dataset number, and for the number Temporary sample data set, where the length of each historical charging data is satisfy: ; For the number The following processing steps are performed for each historical charging data in the temporary sample data set: Sequentially number each data element in the historical charging data; Extract data elements whose numbers are less than or equal to the historical charging data. Generates a sequence of input data elements from the data elements of Extract data elements with numbers greater than The data elements generate an output data element sequence; Add at the end of the input data element sequence The input data sample is obtained by filling the number of paddings; Add at the end of the output data element sequence The output data samples are obtained by filling the number of samples.

[0045] In the technical solution of the above implementation, the length of the input data element sequence is The length of the output data element sequence is .

[0046] Further, when obtaining the input data length of the pre-configured data sequence prediction model And output data sequence Before the steps, also include: Get the maximum charging data length in the historical charging data , the maximum charging data length The number of data elements in the charging data sequence with the longest single charging time of the single lithium battery in the historical charging data; The maximum charging data length With the minimum output sequence length The input data length is determined by the difference: ; The maximum charging data length With the minimum input sequence length The difference is determined as the output data length: .

[0047] Furthermore, the step of adjusting the charging voltage of the single lithium battery in the lithium battery pack to align the second time at which each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode specifically includes: Get the current charging mode; Determine a third time corresponding to the first charging mode based on the second time of each single lithium battery in the lithium battery pack; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to adjust the switching time of each single lithium battery from the first charging mode to the second charging mode to the third time.

[0048] Furthermore, the step of determining a third time corresponding to the first charging mode based on the second time of each single lithium battery in the lithium battery pack specifically includes: determining one or more second parameter thresholds corresponding to the first charging mode, and a second target charging characteristic parameter corresponding to the second parameter thresholds; Aggregating the second time of all single lithium batteries in the lithium battery pack into a second time set; A third time is determined in the second time set so that when the switching time of each single lithium battery in the lithium battery group from the first charging mode to the second charging mode is adjusted to the third time, the value of the second target charging characteristic parameter of each single lithium battery in the lithium battery group is less than or equal to the second parameter threshold.

[0049] Preferably, in the step of determining a third time in the second time set, the value of the second target charging characteristic parameter of at least one single lithium battery in the lithium battery pack is equal to the second parameter threshold.

[0050] like Figure 2 As shown, the second aspect of the present invention proposes a lithium battery charging voltage balancing control system, including a charging characteristic parameter monitoring unit for obtaining the charging characteristic parameters of each single lithium battery in the lithium battery pack, a charging voltage regulating unit for adjusting the charging voltage of each single lithium battery in the lithium battery pack, and a control unit connected to the charging characteristic parameter monitoring unit and the charging voltage regulating unit, wherein the control unit is configured to implement the lithium battery charging voltage balancing control method described in any one of the first aspect of the present invention.

[0051] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0052] According to the embodiments of the present invention as described above, these embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and changes can be made based on the above description. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can make good use of the present invention and the modified use based on the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A lithium battery charging voltage equalization control method, characterized in that: include: During the charging process, the charging characteristic parameter of each single lithium battery in the lithium battery pack is monitored, wherein the charging characteristic parameter is a characteristic parameter or a combination of multiple characteristic parameters reflecting the charging state of the single lithium battery; Determine a first charging mode of the current stage and a second charging mode of the next stage for each single lithium battery based on the charging characteristic parameters; estimating, based on empirical data, a first time for each single lithium battery to switch from the first charging mode to the second charging mode; When the difference between the first time and the current time is less than or equal to a preset time difference threshold, a first time data sequence of the charging characteristic parameter is generated, wherein the first time data sequence is a data sequence obtained by arranging discrete values ​​of the charging characteristic parameter of the single lithium battery in time sequence during the charging process; Predicting a second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to align the second time when each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode.

2. The lithium battery charging voltage equalization control method according to claim 1, characterized in that: The step of estimating the first time for each single lithium battery to switch from the first charging mode to the second charging mode according to empirical data specifically includes: Using historical charging data of a single lithium battery to fit a first charging characteristic function of its charging characteristic parameters relative to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Determine a first target charging characteristic parameter corresponding to a first parameter threshold for switching from the first charging mode to the second charging mode; Obtaining a current value of a first target charging characteristic parameter of each single lithium battery; Determine a first function time of a first parameter threshold of the first target charging characteristic parameter in the first charging characteristic function and a second function time of a current value of the first target charging characteristic parameter in the first charging characteristic function; Calculate the time difference between the first function time and the second function time; A time obtained by summing the current time and the time difference is determined as the first time.

3. The lithium battery charging voltage equalization control method according to claim 1, characterized in that: The step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter specifically includes: Converting the first time data series into input data of a data series prediction model, wherein the data series prediction model is a recurrent neural network model trained based on historical charging data of single lithium batteries; Inputting the input data into the data sequence prediction model to obtain output data; Converting the output data into a second time data series; The second time when the single lithium battery switches from the first charging mode to the second charging mode is calculated based on the second time data series.

4. The lithium battery charging voltage equalization control method according to claim 3, characterized in that: The step of converting the first time data series into input data of a data series prediction model specifically includes: Determine the largest data element in the first time data series: , and the minimum data element: ， in 1 to A positive integer between is the first time data series data elements, is the number of data elements in the first time data series; According to the maximum data element and the minimum data element Calculate the normalization function of the first time data series: ； Normalizing the first time data series based on a normalization function to obtain a normalized data series; Add at the end of the normalized data sequence The input data is obtained by filling the number of The length of the pre-configured input data.

5. The lithium battery charging voltage equalization control method according to claim 4, characterized in that: The step of converting the output data into a second time data series specifically includes: Get each data element in the output data ,in 1 to A positive integer between is the number of data elements in the output data; Based on the output data data element calculates the first Data elements: ； Calculate the starting time of the second time data series and end time ; Based on starting time and end time Generate a corresponding time for each data element in the second time data series.

6. The lithium battery charging voltage equalization control method according to claim 5, characterized in that: The step of calculating the second time for the single lithium battery to switch from the first charging mode to the second charging mode based on the second time data series specifically includes: Fitting a second charging characteristic function of the data elements in the second time data series with respect to time; Obtaining a preconfigured first parameter threshold for switching from the first charging mode to the second charging mode; Calculating the charging characteristic value as the target time corresponding to the first parameter threshold by the second charging characteristic function; The target time is determined as the second time.

7. The lithium battery charging voltage equalization control method according to claim 3, characterized in that: Before the step of predicting the second time for each single lithium battery to switch from the first charging mode to the second charging mode based on the first time data sequence of the charging characteristic parameter, the method further includes: Determine a minimum input sequence length and a minimum output sequence length ; Each historical charging data of a single lithium battery is input with the minimum input sequence length The minimum output sequence length is used as a reference for segmentation and conversion into input data samples and output data samples; Using the input data samples and the output data samples to construct a training data set and a validation data set for the data sequence prediction model; The data sequence prediction model is trained using the training data set and the validation data set.

8. The lithium battery charging voltage equalization control method according to claim 7, characterized in that: The steps of dividing each historical charging data of a single lithium battery based on the minimum input sequence length and the maximum input sequence length and converting them into input data samples and output data samples specifically include: Get the input data length of the pre-configured data sequence prediction model And output data sequence ; The historical charging data is divided into A temporary sample dataset; Configure a count range from 1 to Count variable , so that each temporary sample data set has a corresponding value as its dataset number, and for the number Temporary sample data set, where the length of each historical charging data is satisfy: ； For the number The following processing steps are performed for each historical charging data in the temporary sample data set: Sequentially number each data element in the historical charging data; Extract data elements whose numbers are less than or equal to the historical charging data. Generates a sequence of input data elements from the data elements of Extract data elements with numbers greater than The data elements generate an output data element sequence; Add at the end of the input data element sequence The input data sample is obtained by filling the number of paddings; Add at the end of the output data element sequence The output data samples are obtained by filling the number of samples.

9. The lithium battery charging voltage equalization control method according to claim 7, characterized in that: The step of adjusting the charging voltage of the single lithium battery in the lithium battery pack to align the second time when each single lithium battery in the lithium battery pack switches from the first charging mode to the second charging mode specifically includes: Get the current charging mode; Determine a third time corresponding to the first charging mode based on the second time of each single lithium battery in the lithium battery pack; The charging voltage of the single lithium battery in the lithium battery pack is adjusted to adjust the switching time of each single lithium battery from the first charging mode to the second charging mode to the third time.

10. A lithium battery charging voltage equalization control system, characterized in that: It includes a charging characteristic parameter monitoring unit for obtaining the charging characteristic parameters of each single lithium battery in a lithium battery pack, a charging voltage adjustment unit for adjusting the charging voltage of each single lithium battery in the lithium battery pack, and a control unit connected to the charging characteristic parameter monitoring unit and the charging voltage adjustment unit, wherein the control unit is configured to implement the lithium battery charging voltage balancing control method as described in any one of claims 1 to 9.

Images