Battery pack health state evaluation method and system based on operation mode portrait

Abstract

The invention discloses a battery pack health state evaluation method and system based on an operation mode portrait, and relates to the technical field of battery pack management. Historical operation data of a battery pack is analyzed by adopting an ampere-hour integral method to obtain a plurality of battery pack label capacities, and the operation mode portrait is constructed through the historical operation data; the method comprises the steps of obtaining an operation mode portrait of a battery pack, performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features, performing adaptive redundancy feature screening on each statistical feature through each battery pack label capacity to obtain a plurality of related health indexes, and training a preset battery pack health state evaluation model by adopting each related health index to obtain a battery pack health state evaluation result. And obtaining a target battery pack health state evaluation model. The technical problem that an existing battery pack health state evaluation model is mainly developed based on battery monomer level test data under a fixed working condition, and when obtained time sequence acquisition data is disordered, the battery pack health state cannot be accurately evaluated is solved.

Description

Technical Field

[0001] The present invention relates to the field of battery pack management technology, and in particular to a battery pack health status assessment method and system based on operation mode portrait. Background Art

[0002] The health status is a key indicator for evaluating the service status of a battery pack. It quantifies the degree of aging of the current performance of a battery pack compared to a new battery pack. It cannot be directly measured by sensors and is usually characterized by the capacity decline and internal resistance growth of the battery pack. Accurately evaluating the health status of a battery pack is of great significance for understanding the current endurance and remaining life of the vehicle, responding to maintenance needs in a timely manner, and realizing the rational use of retired power battery packs.

[0003] At present, the existing battery pack health status assessment model is mainly developed based on battery cell level test data under fixed working conditions. When the time series acquisition data obtained is disorganized, it is impossible to accurately assess the health status of the battery pack. Summary of the invention

[0004] The present invention provides a battery pack health status assessment method and system based on operation mode portrait, which solves the technical problem that the existing battery pack health status assessment model is mainly developed based on battery cell level test data under fixed working conditions, and when the time series acquisition data obtained is disorganized, it is impossible to accurately assess the battery pack health status.

[0005] A first aspect of the present invention provides a method for evaluating the health status of a battery pack based on an operation mode portrait, comprising:

[0006] Acquire historical operation data of the battery pack, and perform tag capacity analysis on the historical operation data using an ampere-hour integration method to obtain tag capacities of multiple battery packs;

[0007] constructing an operation mode portrait of the battery pack according to the historical operation data, and performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features;

[0008] Adaptively screening redundant features of each of the statistical features through the capacity of each of the battery pack tags to obtain a plurality of related health indicators;

[0009] Using each of the relevant health indicators and each of the battery pack label capacities to input a preset battery pack health status assessment model for training, to generate a target battery pack health status assessment model;

[0010] When the real-time relevant health indicator of the battery pack is received, the real-time relevant health indicator is input into the target battery pack health status assessment model to obtain a status assessment result corresponding to the battery pack.

[0011] Optionally, the step of performing tag capacity analysis on the historical operation data using the ampere-hour integration method to obtain multiple battery pack tag capacities includes:

[0012] Performing difference calculation on each end state of charge in the historical operation data and the corresponding start state of charge to obtain a plurality of state of charge limit values;

[0013] Determining whether each of the state of charge limit values ​​is less than a preset label threshold value;

[0014] When the state of charge limit is greater than or equal to the tag threshold, the state of charge limit and the historical operation data are input into a preset ampere-hour integration function to obtain the tag capacity;

[0015] A ratio operation is performed on each of the tag capacities and a preset battery pack rated capacity to obtain a plurality of battery pack tag capacities.

[0016] Optionally, the step of constructing an operation mode portrait of the battery pack according to the historical operation data, and performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features includes:

[0017] Constructing an operation mode portrait of the battery pack according to the historical operation data, wherein the operation mode portrait includes a first distribution sequence, a second distribution sequence, a third distribution sequence, a fourth distribution sequence and a fifth distribution sequence;

[0018] Performing statistical analysis on the first distribution sequence, the second distribution sequence, and the third distribution sequence respectively through a variety of data feature operations to obtain a variety of overall aging trend features;

[0019] Performing data feature operations on the fourth distribution sequence and the fifth distribution sequence respectively through a plurality of the data feature operations to obtain a plurality of performance degradation trend features;

[0020] Performing data feature operations on various performance decay trend features through a plurality of data feature operations respectively to obtain a plurality of monomer difference features;

[0021] All of the overall aging trend characteristics and all of the monomer difference characteristics are used as statistical characteristics.

[0022] Optionally, the step of performing adaptive redundant feature screening on each of the statistical features through the capacity of each of the battery pack tags to obtain a plurality of related health indicators includes:

[0023] Calculating the Pearson correlation coefficient between each of the battery pack tag capacities and each of the statistical features respectively by using a preset Pearson correlation coefficient function;

[0024] Determine whether each of the Pearson correlation coefficients is less than a preset correlation threshold;

[0025] When the Pearson correlation coefficient is greater than or equal to the correlation threshold, the Pearson correlation coefficient is determined as the correlation coefficient;

[0026] The maximum value among the correlation coefficients is selected as the benchmark coefficient, and the statistical characteristics associated with the correlation coefficients are screened by the benchmark coefficient to obtain multiple related health indicators.

[0027] Optionally, the step of screening the statistical features associated with each of the correlation coefficients by the benchmark coefficient to obtain a plurality of related health indicators includes:

[0028] Performing mean operations on the reference coefficient and each of the correlation coefficients respectively to obtain first means corresponding to each of the correlation coefficients;

[0029] Calculating alternative correlation coefficients between the statistical features associated with the reference coefficient and the statistical features associated with each of the correlation coefficients respectively by using the Pearson correlation coefficient function;

[0030] Determining whether each of the candidate correlation coefficients is greater than the corresponding first mean;

[0031] When the candidate correlation coefficient is greater than the corresponding first mean value, the candidate correlation coefficient is determined as the target coefficient;

[0032] Selecting the maximum value among the target coefficients as a new reference coefficient, and jumping to the step of performing mean operations on the reference coefficient and the correlation coefficients to obtain first means corresponding to the correlation coefficients, until any of the candidate correlation coefficients is less than or equal to the corresponding first means;

[0033] When any of the candidate correlation coefficients is less than or equal to the corresponding first mean value, the statistical characteristics associated with each of the benchmark coefficients are determined as relevant health indicators.

[0034] Optionally, the step of using each of the relevant health indicators and each of the battery pack label capacities to input a preset battery pack health status assessment model for training to generate a target battery pack health status assessment model includes:

[0035] Performing data preprocessing on each of the relevant health indicators and each of the battery pack tag capacities to obtain a state feature set;

[0036] Using the state feature set to input a preset battery pack health state assessment model for training to obtain training battery pack health state estimation data;

[0037] Calculating the training loss function value of the state feature set according to the training battery pack health state estimation data;

[0038] When the training loss function value is greater than or equal to a preset standard loss function value, a preset parameter adjustment method is used to adjust the model parameters of the battery pack health status assessment model until the training loss function value is less than the standard loss function value;

[0039] When the training loss function value is less than the standard loss function value, a target battery pack health status assessment model is generated.

[0040] Optionally, the data feature operation includes at least maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness.

[0041] A second aspect of the present invention provides a battery pack health status assessment system based on operation mode portrait, characterized by comprising:

[0042] A collection module, used to obtain historical operation data of the battery pack, and perform tag capacity analysis on the historical operation data using an ampere-hour integration method to obtain tag capacities of multiple battery packs;

[0043] A statistical analysis module, used for constructing an operation mode portrait of the battery pack according to the historical operation data, and performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features;

[0044] A feature screening module, used to perform adaptive redundant feature screening on each of the statistical features according to the capacity of each battery pack tag to obtain a plurality of related health indicators;

[0045] A training module, used to train a preset battery health status assessment model using each of the relevant health indicators and each of the battery pack label capacities to generate a target battery health status assessment model;

[0046] The evaluation module is used to input the real-time relevant health indicators of the battery group into the target battery group health status evaluation model when receiving the real-time relevant health indicators of the battery group, so as to obtain the status evaluation result corresponding to the battery group.

[0047] A third aspect of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the battery pack health status assessment method based on operating mode portrait as described in any one of the above items.

[0048] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the method for evaluating the health status of a battery pack based on an operating mode portrait as described in any one of the above items is implemented.

[0049] It can be seen from the above technical solutions that the present invention has the following advantages:

[0050] The present invention constructs an operating mode portrait of a battery pack according to historical operating data, and statistically analyzes the operating mode portrait to obtain multiple statistical features, and then adaptively screens redundant features of each statistical feature according to the capacity of each battery pack tag to obtain multiple relevant health indicators, and uses each relevant health indicator to train a preset battery pack health status assessment model, thereby obtaining a target battery pack health status assessment model suitable for different battery types and any dynamic working conditions. The existing battery pack health status assessment model is mainly developed based on battery cell level test data under fixed working conditions, and the technical problem that the battery pack health status cannot be accurately assessed when the time series acquisition data obtained in the face is chaotic. Compared with the traditional battery pack status assessment method, the present invention adaptively screens redundant features of the operating mode portrait of the battery pack to obtain the most relevant and independent relevant health indicators, and then uses the relevant health indicators to train the preset battery pack health status assessment model to obtain the target battery pack health status assessment model. The target battery pack health status assessment model has no specific requirements for the battery pack charging strategy and usage behavior, is suitable for different battery types and any dynamic working conditions, and can accurately assess the battery pack health status. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0052] Figure 1 A flowchart of a method for evaluating the health status of a battery pack based on an operation mode portrait provided in Embodiment 1 of the present invention;

[0053] Figure 2 A flowchart of a method for evaluating the health status of a battery pack based on an operation mode portrait provided in Embodiment 2 of the present invention;

[0054] Figure 3 A schematic diagram of battery tag capacity trend in an actual application scenario provided by the second embodiment of the present invention;

[0055] Figure 4A schematic diagram of a battery pack operating mode provided in the second embodiment of the present invention;

[0056] Figure 5 A schematic diagram of a battery pack health status estimation result provided in the second embodiment of the present invention;

[0057] Figure 6 A structural block diagram of a battery pack health status assessment system based on operation mode profiling provided in Embodiment 3 of the present invention;

[0058] Figure 7 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0059] The embodiments of the present invention provide a method and system for evaluating the health status of a battery pack based on an operation mode portrait, which is used to solve the technical problem that the existing battery pack health status evaluation model is mainly developed based on battery cell-level test data under fixed working conditions, and when the time series acquisition data obtained is disorganized, it is impossible to accurately evaluate the health status of the battery pack.

[0060] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] See also Figure 1 , Figure 1 A flowchart of the steps of a battery pack health status assessment method based on operation mode profiling is provided in Embodiment 1 of the present invention.

[0062] The present invention provides a battery pack health status assessment method based on operation mode portrait, comprising:

[0063] Step 101, obtaining historical operation data of a battery pack, and performing tag capacity analysis on the historical operation data using an ampere-hour integration method to obtain tag capacities of multiple battery packs;

[0064] In an embodiment of the present invention, historical operating data of a battery pack is obtained, and a difference operation is performed between each termination state of charge in the historical operating data and the corresponding starting state of charge to obtain a plurality of state of charge limit values, and it is determined whether each state of charge limit value is less than a preset label threshold value. When the state of charge limit value is greater than or equal to the label threshold value, the state of charge limit value and the historical operating data are input into a preset ampere-hour integral function to obtain a label capacity, and a ratio operation is performed between each label capacity and a preset battery pack rated capacity to obtain a plurality of battery pack label capacities.

[0065] It should be noted that the historical operating data includes but is not limited to the battery pack's current, voltage, SOC (state of charge), temperature, operating time, current interval, voltage interval, state of charge interval, temperature interval, and the start and end time for calculating the tag capacity, as well as the battery pack's current, voltage, SOC (state of charge), temperature, operating time, current interval, voltage interval, state of charge interval, temperature interval, and the start and end time for calculating the tag capacity, etc.

[0066] Step 102: construct an operation mode portrait of the battery pack according to the historical operation data, and perform statistical analysis on the operation mode portrait to obtain multiple statistical features;

[0067] In an embodiment of the present invention, an operating mode portrait of a battery pack is constructed based on historical operating data, and a statistical analysis is performed on the operating mode portrait to obtain multiple statistical features. The operating mode portrait includes a first distribution sequence, a second distribution sequence, a third distribution sequence, a fourth distribution sequence and a fifth distribution sequence.

[0068] Step 103, adaptively screening redundant features of various statistical features through the capacity of each battery pack tag to obtain multiple related health indicators;

[0069] In the embodiment of the present invention, adaptive redundant feature screening is performed on each statistical feature according to the capacity of each battery pack tag to obtain a plurality of related health indicators.

[0070] Step 104: Use various relevant health indicators and label capacities of various battery packs to input a preset battery pack health status assessment model for training, and generate a target battery pack health status assessment model;

[0071] In an embodiment of the present invention, a training feature set is constructed using various relevant health indicators and various battery pack label capacities, and a preset battery pack health status assessment model is trained using the training feature set to generate a target battery pack health status assessment model.

[0072] It should be noted that the battery pack health status assessment model can be a neural network model or a random forest regression model, etc.

[0073] Step 105 : When the real-time relevant health indicators of the battery pack are received, the real-time relevant health indicators are input into the target battery pack health status assessment model to obtain the status assessment result corresponding to the battery pack.

[0074] Real-time relevant health indicators refer to relevant health indicators of the battery pack obtained in real time.

[0075] In an embodiment of the present invention, real-time relevant health indicators of a battery pack are obtained, and a state assessment of the real-time relevant health indicators is performed using a target battery pack health state assessment model to obtain a state assessment result corresponding to the battery pack.

[0076] In an embodiment of the present invention, a battery pack operation mode portrait is constructed according to historical operation data, and the operation mode portrait is statistically analyzed to obtain multiple statistical features, and then each statistical feature is adaptively redundantly screened according to the capacity of each battery pack tag to obtain multiple relevant health indicators, and each relevant health indicator is used to train the preset battery pack health status assessment model, thereby obtaining a target battery pack health status assessment model suitable for different battery types and any dynamic working conditions. The existing battery pack health status assessment model is mainly developed based on battery cell level test data under fixed working conditions, and the technical problem that the battery pack health status cannot be accurately assessed when the time series acquisition data obtained in the face is chaotic. Compared with the traditional battery pack status assessment method, the present invention adaptively redundantly screens the operation mode portrait of the battery pack to obtain the most relevant and independent relevant health indicators, and then uses the relevant health indicators to train the preset battery pack health status assessment model to obtain the target battery pack health status assessment model. The target battery pack health status assessment model has no specific requirements for the battery pack charging strategy and usage behavior, is suitable for different battery types and any dynamic working conditions, and can accurately assess the battery pack health status.

[0077] See also Figure 2 , Figure 2 A flowchart of the steps of a battery pack health status assessment method based on operation mode portrait is provided in Embodiment 2 of the present invention.

[0078] The present invention provides a battery pack health status assessment method based on operation mode portrait, comprising:

[0079] Step 201, obtaining historical operation data of the battery pack, performing difference calculation on each termination state of charge in the historical operation data and the corresponding initial state of charge, to obtain a plurality of state of charge limit values;

[0080] In the embodiment of the present invention, historical operation data of the battery pack is obtained, and each termination state of charge (i.e. ) and the corresponding initial state of charge (i.e. ) to perform difference calculation to obtain multiple state of charge limit values.

[0081] Step 202: Determine whether each state of charge limit is less than a preset tag threshold value;

[0082] In the embodiment of the present invention, it is determined whether each state of charge limit value is less than 40%.

[0083] Step 203: When the state of charge limit is greater than or equal to the tag threshold, the state of charge limit and the historical operation data are input into a preset ampere-hour integration function to obtain the tag capacity;

[0084] In the embodiment of the present invention, when the state of charge limit value is greater than or equal to 40%, the state of charge limit value (ie ) and historical operation data are input into the preset ampere-hour integration function to obtain the tag capacity.

[0085] It should be noted that the final tag capacity trend is as follows Figure 2 shown.

[0086] It should be noted that the ampere-hour integral function is specifically:

[0087]

[0088] in, is the label capacity, is the battery current, is the starting time, The end time.

[0089] Step 204 , respectively perform ratio calculations on each tag capacity and a preset battery pack rated capacity to obtain a plurality of battery pack tag capacities.

[0090] In the embodiment of the present invention, the ratio between each tag capacity and 155Ah is calculated respectively to obtain multiple battery pack tag capacities.

[0091] Step 205: construct an operation mode portrait of the battery pack according to the historical operation data, and perform statistical analysis on the operation mode portrait to obtain multiple statistical features;

[0092] Further, step 205 includes the following sub-steps:

[0093] S11, constructing an operation mode portrait of the battery pack according to the historical operation data, wherein the operation mode portrait includes a first distribution sequence, a second distribution sequence, a third distribution sequence, a fourth distribution sequence and a fifth distribution sequence;

[0094] The first distribution sequence refers to the distribution sequence of the cumulative operating time under the battery pack current stress.

[0095] The second distribution sequence refers to the distribution sequence of the cumulative operating time under the voltage stress of the battery pack.

[0096] The third distribution sequence refers to the distribution sequence of the cumulative operating time under the charge state stress of the battery pack.

[0097] The fourth distribution sequence refers to the distribution sequence of the cumulative operating time under the voltage stress of the battery pack.

[0098] The fifth sequence refers to the distribution sequence of the cumulative operating time under the temperature stress of the battery pack cells.

[0099] In the embodiment of the present invention, refer to Figure 4 As shown, based on the preset first operating mode portrait function, the upper and lower limits and intervals of the current stress, voltage stress, SOC stress (i.e., state of charge stress), and temperature stress of the battery pack in the historical operating data are statistically analyzed; based on the preset second operating model portrait function, the first distribution sequence, second distribution sequence, third distribution sequence, fourth distribution sequence and fifth distribution sequence of the battery pack and the cells in the battery pack at different stress intervals in the historical operating data are statistically analyzed.

[0100] It should be noted that, in the first operating mode portrait function, the maximum current is 160A, the minimum current is -120A, the current interval is 20A, the maximum battery pack voltage is 420V, the minimum battery pack voltage is 330V, the battery pack voltage interval is 2V, the maximum SOC is 100%, the minimum SOC is 0%, the SOC interval is 5%, the maximum single cell voltage is 4.3V, the minimum single cell voltage is 3.3V, the single cell voltage interval is 0.01V, the maximum single cell temperature is 40℃, the minimum single cell temperature is 0℃, and the single cell temperature interval is 1℃.

[0101] It should be noted that the portrait function of the first operation mode is specifically:

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] in, is the current stress, is the voltage stress, is the state-of-charge stress, is the single cell voltage stress, is the temperature stress, is the minimum current, is the maximum current, is the minimum voltage, is the maximum voltage, is the minimum state of charge, is the maximum state of charge, is the minimum cell voltage, is the maximum single cell voltage, is the minimum temperature, is the maximum temperature, is the current interval, is the voltage interval, is the state of charge interval, is the single cell voltage interval, is the temperature interval.

[0108] It should be noted that the second running model portrait function is specifically:

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] in, is the first distribution sequence, is the second distribution sequence, is the third distribution sequence, is the fourth distribution sequence, is the fifth sequence, M is the number of cells in the battery pack, N is the number of temperature sampling points in the battery pack, n is the number of intervals, is the cumulative operating time under the first interval of current stress, is the cumulative operating time under the second interval of current stress, is the cumulative operating time under the nth interval of current stress, is the cumulative operating time under the first interval of voltage stress, is the cumulative operating time under the second interval of voltage stress, is the cumulative operating time under the nth interval of voltage stress, is the cumulative running time in the first interval of the state of charge stress, is the cumulative running time under the second interval of the state of charge stress, is the cumulative running time under the nth interval of the state of charge stress, is the cumulative operating time of the first interval of the i-th cell voltage stress, is the cumulative operating time of the second interval of the voltage stress of the i-th cell, is the cumulative operating time of the i-th cell voltage stress under the n-th interval, is the cumulative running time in the first interval of the stress at the i-th temperature collection point, is the cumulative running time of the stress at the i-th temperature collection point in the second interval, It is the cumulative running time of the stress at the i-th temperature collection point in the n-th interval.

[0115] S12, performing statistical analysis on the first distribution sequence, the second distribution sequence, and the third distribution sequence respectively through a variety of data feature operations to obtain a variety of overall aging trend features;

[0116] It should be noted that data feature operations include at least maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness.

[0117] The overall aging trend characteristic refers to the overall aging trend of the battery pack.

[0118] In an embodiment of the present invention, the maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness of the first distribution sequence, the second distribution sequence and the third distribution sequence are calculated respectively, and the maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness of the first distribution sequence, the second distribution sequence and the third distribution sequence are used as the overall aging trend characteristics.

[0119] S13, performing data feature operations on the fourth distribution sequence and the fifth distribution sequence respectively through multiple data feature operations to obtain multiple performance degradation trend features;

[0120] The performance degradation trend characteristic refers to the performance degradation trend of the cells in the battery pack.

[0121] In an embodiment of the present invention, the maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness of the fourth distribution sequence and the fifth distribution sequence are calculated respectively, and the maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness of the fourth distribution sequence and the fifth distribution sequence are used as performance degradation trend characteristics.

[0122] It should be noted that, since there are M battery cells and N temperature sampling points in the battery pack, M fourth distribution sequences and N fifth distribution sequences will be obtained.

[0123] S14, performing data feature operations on various performance decay trend features through multiple data feature operations to obtain multiple monomer difference features;

[0124] In the embodiment of the present invention, data feature operations are performed on various performance degradation trend features through various data feature operations to obtain multiple individual difference features. For example, there are M fourth distribution sequence maxima, and the maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness between the M fourth distribution sequence maxima are calculated respectively.

[0125] S15. Use all overall aging trend characteristics and all monomer difference characteristics as statistical features.

[0126] In the embodiment of the present invention, all the overall aging trend features and all the individual difference features are used as statistical features.

[0127] Step 206, adaptively screening redundant features of various statistical features by using the tag capacity of each battery pack to obtain multiple related health indicators;

[0128] Further, step 206 includes the following sub-steps:

[0129] S21, respectively calculating the Pearson correlation coefficient between each battery pack tag capacity and each statistical feature using a preset Pearson correlation coefficient function;

[0130] In the embodiment of the present invention, the capacity of each battery pack tag and each statistical feature are input into a preset Pearson correlation coefficient function to obtain a Pearson correlation coefficient.

[0131] It should be noted that the Pearson correlation coefficient function is specifically:

[0132]

[0133] in, is the i-th health indicator, Calculate the label for the ith battery pack, is the mean value of health indicators, Calculate the label mean for the battery group.

[0134] S22, determining whether each Pearson correlation coefficient is less than a preset correlation threshold;

[0135] In the embodiment of the present invention, it is determined whether each Pearson correlation coefficient is less than 0.2.

[0136] S23. When the Pearson correlation coefficient is greater than or equal to the correlation threshold, the Pearson correlation coefficient is determined as the correlation coefficient;

[0137] In the embodiment of the present invention, when the Pearson correlation coefficient is greater than or equal to 0.2, the Pearson correlation coefficient is determined as the correlation coefficient.

[0138] S24. Select the maximum value among the correlation coefficients as the benchmark coefficient, and screen the statistical characteristics associated with the correlation coefficients through the benchmark coefficient to obtain multiple related health indicators.

[0139] Further, S24 includes the following sub-steps:

[0140] S241, respectively performing mean operations on the benchmark coefficient and each correlation coefficient to obtain a first mean value corresponding to each correlation coefficient;

[0141] In the embodiment of the present invention, the correlation coefficients are sorted according to their sizes, the first-ranked correlation coefficient is used as the benchmark coefficient, and the benchmark coefficient and each correlation coefficient are respectively averaged to obtain the first mean corresponding to each correlation coefficient.

[0142] S242, respectively calculating the candidate correlation coefficients between the statistical features associated with the benchmark coefficient and the statistical features associated with each correlation coefficient by using the Pearson correlation coefficient function;

[0143] In the embodiment of the present invention, the statistical features associated with the benchmark coefficient and the statistical features associated with each correlation coefficient are input into the Pearson correlation coefficient function to obtain a plurality of candidate correlation coefficients.

[0144] S243, judging whether each candidate correlation coefficient is greater than the corresponding first mean;

[0145] In the embodiment of the present invention, it is determined whether each candidate correlation coefficient is greater than the corresponding first mean value.

[0146] S244, when the candidate correlation coefficient is greater than the corresponding first mean value, determining the candidate correlation coefficient as the target coefficient;

[0147] In the embodiment of the present invention, when the candidate correlation coefficient is greater than the corresponding first mean value, the candidate correlation coefficient is used as the target coefficient.

[0148] S245, selecting the maximum value among the target coefficients as a new reference coefficient, and jumping to the step of performing mean operations on the reference coefficient and the correlation coefficients to obtain first means corresponding to the correlation coefficients, until any candidate correlation coefficient is less than or equal to the corresponding first mean;

[0149] In the embodiment of the present invention, the maximum value among the target coefficients is selected as the new reference coefficient, and the process jumps to S241 until any candidate correlation coefficient is less than or equal to the corresponding first mean value.

[0150] S246. When any candidate correlation coefficient is less than or equal to the corresponding first mean, the statistical characteristics associated with each benchmark coefficient are determined as relevant health indicators.

[0151] In the embodiment of the present invention, when each candidate correlation coefficient is less than or equal to the corresponding first mean value, the statistical characteristics associated with each benchmark coefficient are determined as relevant health indicators.

[0152] Step 207: Use each relevant health indicator and each battery pack label capacity to input a preset battery pack health status assessment model for training to generate a target battery pack health status assessment model;

[0153] Further, step 207 includes the following sub-steps:

[0154] S31, performing data preprocessing on each relevant health indicator and each battery pack tag capacity to obtain a state feature set;

[0155] In an embodiment of the present invention, data preprocessing is performed on each relevant health indicator and each battery pack tag capacity to obtain a state feature set, wherein the state feature set uses each battery pack tag capacity as a real charge state calculation tag of the state feature set.

[0156] S32, using the state feature set to input a preset battery pack health status assessment model for training, to obtain training battery pack health status estimation data;

[0157] In an embodiment of the present invention, each relevant health indicator of the state feature set is input into a preset battery pack health state assessment model for training to obtain training battery pack health state estimation data.

[0158] S33, calculating the training loss function value of the state feature set according to the training battery pack health state estimation data;

[0159] In the embodiment of the present invention, a real state of charge calculation label associated with the training battery health state estimation data is selected from the state feature set, and the deviation value between the training battery health state estimation data and the real state of charge calculation label is calculated. The deviation value is used as the training loss function value of the state feature set.

[0160] S34, when the training loss function value is greater than or equal to the preset standard loss function value, adjusting the model parameters of the battery pack health status assessment model using a preset parameter adjustment method until the training loss function value is less than the standard loss function value;

[0161] In an embodiment of the present invention, when the training loss function value is greater than or equal to a preset standard loss function value, the model parameters of the battery pack health status assessment model are adjusted using a gradient descent method, a grid search method, or a random search method until the training loss function value is less than the standard loss function value.

[0162] S35. When the training loss function value is less than the standard loss function value, a target battery pack health status assessment model is generated.

[0163] In an embodiment of the present invention, if the training loss function value is less than the standard loss function value, a target battery pack health status assessment model is generated.

[0164] It should be noted that the optimal feature subset obtained by screening under the 30% state feature set is used as the model input, and the model effect is evaluated by calculating the mean absolute percentage error and root mean square error. The mathematical expressions of the two are as follows:

[0165]

[0166]

[0167] in, is the estimated value of the battery pack health status output by the model, Calculate the label for the true SOH, is the mean absolute percentage error, is the root mean square error. After calculation, the target battery pack health status assessment model has a MAPE of 0.88% and a RMSE of 1.07%, which achieves a good assessment of the health status of the lithium-ion battery pack throughout its life cycle. Figure 5 Shown are the model estimation results.

[0168] Step 208: When the real-time relevant health indicators of the battery pack are received, the real-time relevant health indicators are input into the target battery pack health status assessment model to obtain the status assessment result corresponding to the battery pack.

[0169] In an embodiment of the present invention, real-time relevant health indicators of the battery pack are acquired in real time, and the real-time relevant health indicators are input into a target battery pack health status assessment model to obtain a status assessment result corresponding to the battery pack.

[0170] In an embodiment of the present invention, a battery pack operation mode portrait is constructed according to historical operation data, and the operation mode portrait is statistically analyzed to obtain multiple statistical features, and then each statistical feature is adaptively redundantly screened according to the capacity of each battery pack tag to obtain multiple relevant health indicators, and each relevant health indicator is used to train the preset battery pack health status assessment model, thereby obtaining a target battery pack health status assessment model suitable for different battery types and any dynamic working conditions. The existing battery pack health status assessment model is mainly developed based on battery cell level test data under fixed working conditions, and the technical problem that the battery pack health status cannot be accurately assessed when the time series acquisition data obtained in the face is chaotic. Compared with the traditional battery pack status assessment method, the present invention adaptively redundantly screens the operation mode portrait of the battery pack to obtain the most relevant and independent relevant health indicators, and then uses the relevant health indicators to train the preset battery pack health status assessment model to obtain the target battery pack health status assessment model. The target battery pack health status assessment model has no specific requirements for the battery pack charging strategy and usage behavior, is suitable for different battery types and any dynamic working conditions, and can accurately assess the battery pack health status.

[0171] See also Figure 6 , Figure 6 This is a structural block diagram of a battery pack health status assessment system based on operation mode profiling provided in Embodiment 3 of the present invention.

[0172] The present invention provides a battery pack health status assessment system based on operation mode portrait, comprising:

[0173] The acquisition module 301 is used to obtain the historical operation data of the battery pack, and perform tag capacity analysis on the historical operation data using the ampere-hour integration method to obtain the tag capacities of multiple battery packs;

[0174] A statistical analysis module 302 is used to construct an operation mode portrait of the battery pack according to historical operation data, and perform statistical analysis on the operation mode portrait to obtain multiple statistical features;

[0175] A feature screening module 303 is used to perform adaptive redundant feature screening on each statistical feature through the capacity of each battery pack tag to obtain multiple related health indicators;

[0176] A training module 304 is used to train a preset battery health status assessment model using various relevant health indicators and various battery label capacities to generate a target battery health status assessment model;

[0177] The evaluation module 305 is used to input the real-time relevant health indicators of the battery group into the target battery group health status evaluation model when receiving the real-time relevant health indicators of the battery group, so as to obtain the status evaluation result corresponding to the battery group.

[0178] Furthermore, the acquisition module 301 includes:

[0179] The difference operation submodule is used to perform difference operation on each end state of charge in the historical operation data and the corresponding start state of charge to obtain multiple state of charge limit values;

[0180] A first analysis submodule, used to determine whether each state of charge limit value is less than a preset label threshold value;

[0181] When the state of charge limit is greater than or equal to the tag threshold, the state of charge limit and historical operation data are input into the preset ampere-hour integration function to obtain the tag capacity;

[0182] The capacity of each tag is calculated by ratio with the preset rated capacity of the battery pack to obtain multiple battery pack tag capacities.

[0183] Furthermore, the statistical analysis module 302 includes:

[0184] A construction submodule, used to construct an operation mode portrait of the battery pack according to the historical operation data, wherein the operation mode portrait includes a first distribution sequence, a second distribution sequence, a third distribution sequence, a fourth distribution sequence and a fifth distribution sequence;

[0185] A first statistical analysis submodule is used to perform statistical analysis on the first distribution sequence, the second distribution sequence and the third distribution sequence through various data feature operations to obtain various overall aging trend features;

[0186] A second statistical analysis submodule is used to perform data feature operations on the fourth distribution sequence and the fifth distribution sequence through multiple data feature operations to obtain multiple performance degradation trend features;

[0187] The third statistical analysis submodule is used to perform data feature operations on various performance decay trend features through multiple data feature operations to obtain multiple monomer difference features;

[0188] All overall aging trend characteristics and all monomer difference characteristics were used as statistical features.

[0189] Furthermore, the feature screening module 303 includes:

[0190] A second analysis submodule, used to calculate the Pearson correlation coefficient between each battery pack tag capacity and each statistical feature respectively through a preset Pearson correlation coefficient function;

[0191] The third analysis submodule is used to determine whether each Pearson correlation coefficient is less than a preset correlation threshold;

[0192] When the Pearson correlation coefficient is greater than or equal to the correlation threshold, the Pearson correlation coefficient is determined as the correlation coefficient;

[0193] The benchmark feature screening submodule is used to select the maximum value among various correlation coefficients as the benchmark coefficient, and screen the statistical features associated with each correlation coefficient through the benchmark coefficient to obtain multiple related health indicators.

[0194] Furthermore, the benchmark feature screening submodule includes:

[0195] A first analysis unit is used to perform mean operations on the reference coefficient and each correlation coefficient respectively to obtain a first mean value corresponding to each correlation coefficient;

[0196] A second analysis unit is used to calculate the candidate correlation coefficients between the statistical features associated with the benchmark coefficient and the statistical features associated with each correlation coefficient respectively through the Pearson correlation coefficient function;

[0197] A third analysis unit, used to determine whether each candidate correlation coefficient is greater than a corresponding first mean;

[0198] When the candidate correlation coefficient is greater than the corresponding first mean value, the candidate correlation coefficient is determined as the target coefficient;

[0199] A jump unit, used for selecting the maximum value among the target coefficients as a new reference coefficient, and jumping to execute the step of performing mean operations on the reference coefficient and the correlation coefficients respectively to obtain the first mean values ​​corresponding to the correlation coefficients, until any candidate correlation coefficient is less than or equal to the corresponding first mean value;

[0200] When any candidate correlation coefficient is less than or equal to the corresponding first mean value, the statistical characteristics associated with each benchmark coefficient are determined as relevant health indicators.

[0201] Furthermore, the training module 304 includes:

[0202] A preprocessing submodule is used to perform data preprocessing on each relevant health indicator and each battery pack tag capacity to obtain a state feature set;

[0203] A training submodule, used to train a preset battery pack health status assessment model using a state feature set as input, to obtain training battery pack health status estimation data;

[0204] A fourth analysis submodule, used to calculate a training loss function value of a state feature set based on training battery pack health state estimation data;

[0205] When the training loss function value is greater than or equal to the preset standard loss function value, a preset parameter adjustment method is used to adjust the model parameters of the battery pack health status assessment model until the training loss function value is less than the standard loss function value;

[0206] When the training loss function value is less than the standard loss function value, a target battery pack health status assessment model is generated.

[0207] Furthermore, the data feature operations include at least maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness.

[0208] See also Figure 7 , Figure 7 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention.

[0209] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402, wherein the memory 402 stores a computer program; when the computer program is executed by the processor 402, the processor 402 executes a battery pack health status assessment method based on an operating mode portrait as in any of the above embodiments.

[0210] The memory 401 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk or a ROM. The memory 401 has a storage space 403 for a program code 413 for executing any method step in the above method. For example, the storage space 403 for the program code may include individual program codes 413 for implementing the various steps in the above method, respectively. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disk (CD), a memory card or a floppy disk. The program code may be compressed, for example, in an appropriate form. When these codes are run by a computing and processing device, the computing and processing device is caused to execute the various steps in the above-described method.

[0211] Embodiment 5 of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for evaluating the health status of a battery pack based on an operating mode portrait as described in any of the above embodiments is implemented.

[0212] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0213] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0214] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0215] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0216] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0217] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating the health status of a battery pack based on an operation mode portrait, characterized in that: include: Acquire historical operation data of the battery pack, and perform tag capacity analysis on the historical operation data using an ampere-hour integration method to obtain tag capacities of multiple battery packs; constructing an operation mode portrait of the battery pack according to the historical operation data, and performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features; Adaptively screening redundant features of each of the statistical features through the capacity of each of the battery pack tags to obtain a plurality of related health indicators; Using each of the relevant health indicators and each of the battery pack label capacities to input a preset battery pack health status assessment model for training, to generate a target battery pack health status assessment model; When the real-time relevant health indicator of the battery group is received, the real-time relevant health indicator is input into the target battery group health status assessment model to obtain the status assessment result corresponding to the battery group.

2. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 1, characterized in that: The step of using the ampere-hour integration method to perform label capacity analysis on the historical operation data to obtain multiple battery pack label capacities includes: Performing difference calculation on each end state of charge in the historical operation data and the corresponding start state of charge to obtain a plurality of state of charge limit values; Determining whether each of the state of charge limit values ​​is less than a preset label threshold value; When the state of charge limit is greater than or equal to the tag threshold, the state of charge limit and the historical operation data are input into a preset ampere-hour integration function to obtain the tag capacity; A ratio operation is performed on each of the tag capacities and a preset battery pack rated capacity to obtain a plurality of battery pack tag capacities.

3. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 1, characterized in that: The step of constructing an operating mode portrait of the battery pack according to the historical operating data, and performing statistical analysis on the operating mode portrait to obtain a plurality of statistical features includes: Constructing an operation mode portrait of the battery pack according to the historical operation data, wherein the operation mode portrait includes a first distribution sequence, a second distribution sequence, a third distribution sequence, a fourth distribution sequence and a fifth distribution sequence; Performing statistical analysis on the first distribution sequence, the second distribution sequence, and the third distribution sequence respectively through a variety of data feature operations to obtain a variety of overall aging trend features; Performing data feature operations on the fourth distribution sequence and the fifth distribution sequence respectively through a plurality of the data feature operations to obtain a plurality of performance degradation trend features; Performing data feature operations on various performance decay trend features through a plurality of data feature operations respectively to obtain a plurality of monomer difference features; All of the overall aging trend characteristics and all of the monomer difference characteristics are used as statistical characteristics.

4. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 1, characterized in that: The step of adaptively screening the statistical features for redundant features through the capacity of each battery pack tag to obtain a plurality of related health indicators includes: Calculating the Pearson correlation coefficient between each of the battery pack tag capacities and each of the statistical features respectively by using a preset Pearson correlation coefficient function; Determine whether each of the Pearson correlation coefficients is less than a preset correlation threshold; When the Pearson correlation coefficient is greater than or equal to the correlation threshold, the Pearson correlation coefficient is determined as the correlation coefficient; The maximum value among the correlation coefficients is selected as the benchmark coefficient, and the statistical characteristics associated with the correlation coefficients are screened by the benchmark coefficient to obtain multiple related health indicators.

5. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 4, characterized in that: The step of screening the statistical features associated with each of the correlation coefficients by the benchmark coefficient to obtain a plurality of related health indicators includes: Performing mean operations on the reference coefficient and each of the correlation coefficients respectively to obtain first means corresponding to each of the correlation coefficients; Calculating alternative correlation coefficients between the statistical features associated with the reference coefficient and the statistical features associated with each of the correlation coefficients respectively by using the Pearson correlation coefficient function; Determining whether each of the candidate correlation coefficients is greater than the corresponding first mean; When the candidate correlation coefficient is greater than the corresponding first mean value, the candidate correlation coefficient is determined as the target coefficient; Selecting the maximum value among the target coefficients as a new reference coefficient, and jumping to the step of performing mean operations on the reference coefficient and the correlation coefficients to obtain first means corresponding to the correlation coefficients, until any of the candidate correlation coefficients is less than or equal to the corresponding first means; When any of the candidate correlation coefficients is less than or equal to the corresponding first mean value, the statistical characteristics associated with each of the benchmark coefficients are determined as relevant health indicators.

6. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 1, characterized in that: The step of using each of the relevant health indicators and each of the battery pack label capacities to input a preset battery pack health status assessment model for training to generate a target battery pack health status assessment model includes: Performing data preprocessing on each of the relevant health indicators and each of the battery pack tag capacities to obtain a state feature set; Using the state feature set to input a preset battery pack health state assessment model for training to obtain training battery pack health state estimation data; Calculating the training loss function value of the state feature set according to the training battery pack health state estimation data; When the training loss function value is greater than or equal to a preset standard loss function value, a preset parameter adjustment method is used to adjust the model parameters of the battery pack health status assessment model until the training loss function value is less than the standard loss function value; When the training loss function value is less than the standard loss function value, a target battery pack health status assessment model is generated.

7. The method for evaluating the health status of a battery pack based on an operation mode portrait according to claim 3, characterized in that: The data feature operations include at least maximum value, minimum value, range, mean value, variance, standard deviation, kurtosis and skewness.

8. A battery pack health status assessment system based on operation mode portrait, characterized in that: include: A collection module, used to obtain historical operation data of the battery pack, and perform tag capacity analysis on the historical operation data using an ampere-hour integration method to obtain tag capacities of multiple battery packs; A statistical analysis module, used for constructing an operation mode portrait of the battery pack according to the historical operation data, and performing statistical analysis on the operation mode portrait to obtain a plurality of statistical features; A feature screening module, used to perform adaptive redundant feature screening on each of the statistical features according to the capacity of each battery pack tag to obtain a plurality of related health indicators; A training module, used to train a preset battery health status assessment model using each of the relevant health indicators and each of the battery pack label capacities to generate a target battery health status assessment model; The evaluation module is used to input the real-time relevant health indicators of the battery group into the target battery group health status evaluation model when receiving the real-time relevant health indicators of the battery group, so as to obtain the status evaluation result corresponding to the battery group.

9. An electronic device, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the battery pack health status assessment method based on the operating mode portrait as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the method for evaluating the health status of a battery pack based on operating mode portrait as described in any one of claims 1 to 7 is implemented.

Images