Battery pack state of health online detection method and device based on multi-feature fusion

Abstract

The application belongs to the technical field of battery pack health state online detection, and provides a battery pack health state online detection method and device based on multi-feature fusion. The method comprises the following steps: measuring and calculating the IC curve of lithium battery aging and the ΔQ(V) curve of lithium battery aging through an aging charge-discharge cycle experiment, extracting corresponding characteristic values, and regressing the relationship between the characteristic values of the IC curve and the ΔQ(V) curve of lithium battery aging and the discharge capacity and generating discharge capacity equal point and its corresponding features; obtaining the charge-discharge capacity-voltage curve of the lithium battery pack during work online, and obtaining the discharge capacity of the current lithium battery pack; clustering the discharge capacity calculated by different features; according to the correlation coefficient corresponding to the discharge capacity calculation value in the optimal clustering, calculating the weight by using the semi-Gaussian distribution, and taking the weighted average value of the discharge capacity calculation value as the final calculation value of the lithium battery pack discharge capacity, and then calculating the health state of the current lithium battery pack.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of battery pack health state online detection, and particularly relates to a battery pack health state online detection method and device based on multi-feature fusion. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] To meet the energy and power requirements, a large number of lithium batteries are connected in series and parallel to form a battery pack. Lithium ion batteries will gradually age during use, resulting in a decrease in the capacity of the battery pack. Accurate estimation of the health status of the battery pack is the key to optimizing battery operation and controlling battery status in the battery management system.

[0004] The series-parallel connection mode of the battery pack and the parameter inconsistency between lithium batteries jointly affect the capacity aging trend of the battery pack during the entire life cycle. Different series-parallel connection modes and battery inconsistencies have different aging characteristics. Although battery aging experiments help to better understand the inconsistency between batteries and the influence of the series-parallel connection mode on the aging of the battery pack, and are one of the important ways to achieve capacity estimation of the battery pack, the battery aging experiment is high in cost and time-consuming, and cannot be widely applied in practice. At present, most of the methods for calculating the health status are based on the battery level. Moreover, the existing methods for calculating the health status of the battery do not consider the series-parallel connection mode of the battery pack and the inconsistency between the batteries, and cannot be directly applied to the calculation of the health status of the battery pack. SUMMARY

[0005] To solve the above technical problems, the present application provides a battery pack health state online detection method and device based on multi-feature fusion, which uses the cycle aging characteristics of a small number of batteries to calculate the discharge capacity of the battery pack under any series-parallel connection mode, considers the inconsistency between the batteries, and online detects the health status of the battery pack, effectively reducing the cost of battery pack capacity calculation, and providing support for improving battery pack management and safety.

[0006] To achieve the above purpose, the present application adopts the following technical solutions:

[0007] The first aspect of the present application provides a battery pack health state online detection method based on multi-feature fusion.

[0008] In one or more embodiments, a battery pack health state online detection method based on multi-feature fusion is provided, comprising:

[0009] IC curve of lithium battery aging is calculated by testing charge-discharge capacity-voltage curves of several lithium batteries through aging charge-discharge cycle experiment; ΔQ(V) curve of lithium battery aging is obtained by subtracting capacity of the first cycle charge-discharge capacity-voltage curve from capacity of the charge-discharge capacity-voltage curve of lithium battery;

[0010] Characteristic values of IC curve and ΔQ(V) curve of lithium battery aging are extracted, and the relationship between characteristic values of IC curve and ΔQ(V) curve of lithium battery aging and discharge capacity is generated by least square support vector regression, and discharge capacity equal point and corresponding characteristics are generated;

[0011] Charge-discharge capacity-voltage curve of lithium battery pack during working is obtained online, corresponding IC curve and ΔQ(V) curve are calculated, characteristic values of current IC curve and ΔQ(V) curve are extracted, and discharge capacity of current lithium battery pack is calculated by linear interpolation calculation according to discharge capacity equal point of lithium battery aging and corresponding characteristics;

[0012] Discharge capacity calculated by different characteristics is clustered and optimal cluster is selected; according to correlation coefficient corresponding to discharge capacity calculation value in the optimal cluster, weight is calculated by using semi-Gaussian distribution, and weighted average value of discharge capacity calculation value is taken as final calculation value of lithium battery pack discharge capacity;

[0013] According to final calculation value of lithium battery pack discharge capacity and rated capacity, health state of current lithium battery pack is calculated.

[0014] As an embodiment, several new lithium batteries are subjected to aging cycle test until the battery capacity is lower than 70% of the rated capacity; the charging condition is similar to the actual working condition of the calculated lithium battery pack; the discharge condition is set according to the standard lithium battery capacity test condition; the charging capacity, voltage and discharge capacity of lithium battery at each cycle are recorded.

[0015] As an embodiment, the charge-discharge capacity-voltage curves of several lithium batteries tested through aging charge-discharge cycle experiment are smoothed, and the derivative of voltage of the smoothed lithium battery charge-discharge capacity-voltage curve is calculated to obtain the IC curve of lithium battery aging.

[0016] As an embodiment, the correlation coefficient corresponding to the discharge capacity calculation value is:

[0017]

[0018] Wherein, F CellCycle_i is the characteristic of IC or ΔQ(V) curve of the i-th cycle of lithium battery; C disCycle_i is the discharge capacity of the i-th cycle of lithium battery; μ F_CellCycle , σ F_CellCycl are all the cycles FCellCycle_i the mean and standard deviation of all cycles C Cell_disC , σ Cell_disC , respectively. disCycle_i

[0019] As an implementation, for F Pack_Cell greater than or equal to the minimum value of F CellCycle_i and less than or equal to the maximum value of F CellCycle_i , or F Pack_Cell less than the minimum value of F CellCycle_i and greater than the maximum value of F CellCycle_i but the minimum value of F CellCycle_i is located at the beginning and end of F CellCycle_i data, the formula of linear interpolation is:

[0020]

[0021] where F Pack_Cell is a battery pack feature converted to a feature on the battery level, C Pack_Cell is the discharge capacity on the corresponding battery level of the lithium battery pack calculated by linear interpolation of F Pack_Cell , F CellCycle_1 , F CellCycle_2 are the two closest lithium battery features to F Pack_Cell , C CellCycle_1 , C CellCycle_2 are the discharge capacities of the lithium battery corresponding to F CellCycle_1 , F CellCycle_2 ; F CellCycle_i is the feature of the IC or ΔQ(V) curve of the i-th cycle of the lithium battery.

[0022] As an implementation, if the minimum value of F CellCycle_i is located in the middle of its data sequence, C Pack_Cell is calculated using the following formula:

[0023]

[0024] where F Pack_Cell is a battery pack feature converted to a feature on the battery level, C Pack_Cell is the discharge capacity on the corresponding battery level of the lithium battery pack calculated by linear interpolation of F Pack_Cell , F CellCycle_1 , F CellCycle_2 are the two closest lithium battery features to F Pack_Cell , C CellCycle_1 , C CellCycle_2 are the discharge capacities of the lithium battery corresponding to F CellCycle_1 , F CellCycle_2 ; F CellCycle_i ​IC or ΔQ(V) curve of the i-th cycle of the lithium battery; TempC Pack_Cell_1 and TempC Pack_Cell_2 are intermediate parameters.

[0025] As an implementation, the calculation formula of the weight is:

[0026] W i = N(|p Cell_F_C |1, s G )

[0027] wherein N() represents a Gaussian probability density function. |p Cell_F_C | is the absolute value of the correlation coefficient of the feature corresponding to the calculation result in the optimal clustering. The mean of the Gaussian probability density function is 1, and s G is the standard deviation of the Gaussian probability density function.

[0028] The second aspect of the application provides a battery pack state of health online detection device based on multi-feature fusion.

[0029] In one or more embodiments, a battery pack state of health online detection device based on multi-feature fusion comprises:

[0030] An aging experiment test module is configured to test the charge-discharge capacity-voltage curves of a plurality of lithium batteries through an aging charge-discharge cycle experiment, calculate the IC curve of the lithium battery aging, and obtain the ΔQ(V) curve of the lithium battery aging by subtracting the capacity of the first cycle charge-discharge capacity-voltage curve from the capacity of the charge-discharge capacity-voltage curve of the lithium battery;

[0031] An aging test feature extraction module is configured to extract the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging, utilize the least squares support vector to regress the relationship between the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging and the discharge capacity, and generate the discharge capacity equally divided points and the corresponding features;

[0032] A linear interpolation calculation module is configured to obtain the charge-discharge capacity-voltage curve of the lithium battery pack in operation online, calculate the corresponding IC curve and ΔQ(V) curve, extract the characteristic values of the current IC curve and ΔQ(V) curve, perform linear interpolation calculation according to the discharge capacity equally divided points of the lithium battery aging and the corresponding features, and obtain the discharge capacity of the current lithium battery pack.

[0033] A discharge capacity calculation module is configured to cluster and select the optimal clustering of the discharge capacity calculated by different features, utilize the semi-Gaussian distribution to calculate the weight according to the correlation coefficient corresponding to the discharge capacity calculation value in the optimal clustering, and take the weighted average value of the discharge capacity calculation value as the final calculation value of the discharge capacity of the lithium battery pack.

[0034] a health status calculation module configured to calculate a health status of the lithium battery pack according to the final calculated value of the discharge capacity of the lithium battery pack and the rated capacity.

[0035] A third aspect of the present application provides a computer readable storage medium.

[0036] A computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the battery pack health status online detection method based on multi-feature fusion as described above.

[0037] A fourth aspect of the present application provides an electronic device.

[0038] An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the battery pack health status online detection method based on multi-feature fusion as described above when executing the program.

[0039] Compared with the prior art, the present application has the following beneficial effects:

[0040] (1) The present application calculates the health status of a lithium battery pack composed of a large number of lithium batteries using a small amount of aging data of test lithium batteries, and performs multi-feature fusion through a variety of algorithms such as least squares support vector regression and density-based noisy space clustering, so that the health status of the battery pack can be detected online without training in any series-parallel mode, and the inconsistency between the batteries is considered.

[0041] (2) The present application does not require historical operation data of the battery pack, but only requires current charge curve data of the battery pack, so that the health status of the current battery pack can be calculated, the speed of online detection is accelerated, and the influence of long-term static aging of the battery pack is avoided.

[0042] (3) The present application has high accuracy, wide universality and strong stability in calculating the health status of the lithium battery pack, does not require aging experiments on the lithium battery pack, only requires aging experiments on a small amount of lithium batteries, can greatly reduce the health status calculation cost of the lithium battery pack, and provides strong support for optimizing battery operation and controlling battery status in the battery management system of the electric vehicle and large-scale energy storage system. BRIEF DESCRIPTION OF DRAWINGS

[0043] The accompanying drawings, which form a part of the present application, are used to provide further understanding of the present application, and the schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute improper limitations on the present application.

[0044] Figure 1 is a flowchart of the battery pack health status online detection method based on multi-feature fusion of the embodiments of the present application;

[0045] Figure 2 is a structure schematic diagram of a battery pack health state online detection device based on multi-feature fusion according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] The present application will be further described below in conjunction with the accompanying drawings and embodiments.

[0047] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0048] It should be noted that the terms used herein are only intended to describe specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of the features, steps, operations, devices, components and / or combinations thereof.

[0049] Figure 1 is a flowchart of a battery pack health state online detection method based on multi-feature fusion according to an embodiment of the present application, as shown in Figure 1 The battery pack health state online detection method based on multi-feature fusion according to the present embodiment can include:

[0050] S101, the charge-discharge capacity-voltage curves of a plurality of lithium batteries are tested by an aging charge-discharge cycle experiment, and the IC curve of the lithium battery aging is calculated; the capacity of the charge-discharge capacity-voltage curve of the lithium battery is subtracted from the capacity of the charge-discharge capacity-voltage curve of the first cycle to obtain the ΔQ(V) curve of the lithium battery aging;

[0051] S102, the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging are extracted, and the relationship between the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging and the discharge capacity is used to generate the discharge capacity equally divided points and their corresponding characteristics by least square support vector regression;

[0052] S103, the charge-discharge capacity-voltage curve of the lithium battery pack during operation is obtained online, the corresponding IC curve and ΔQ(V) curve are calculated, the characteristic values of the current IC curve and ΔQ(V) curve are extracted, and the discharge capacity of the current lithium battery pack is calculated by linear interpolation according to the discharge capacity equally divided points of the lithium battery aging and their corresponding characteristics;

[0053] S104, clustering the discharge capacity calculated for different characteristics and selecting the optimal cluster; according to the correlation coefficient corresponding to the discharge capacity calculation value in the optimal cluster, calculating the weight using the half-Gaussian distribution, and taking the weighted average value of the discharge capacity calculation value as the final calculation value of the lithium battery pack discharge capacity;

[0054] S105, calculating the health state of the current lithium battery pack according to the final calculation value of the lithium battery pack discharge capacity and the rated capacity.

[0055] In step S101, a number of new lithium batteries are subjected to aging cycle tests until the battery capacity is lower than 70% of the rated capacity; the charging condition is similar to the actual working condition of the calculated lithium battery pack; the discharge condition is set according to the standard lithium battery capacity test condition; the charging capacity, voltage and discharge capacity of the lithium battery at each cycle are recorded.

[0056] The charge-discharge capacity-voltage curves of the lithium batteries tested by the aging charge-discharge cycle experiment are smoothed, and the derivative of the voltage of the smoothed lithium battery charge-discharge capacity-voltage curve is obtained to obtain the IC curve of the lithium battery aging.

[0057] In step S102, before extracting the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging, the IC curve and the ΔQ(V) curve of the lithium battery aging are also smoothed to remove noise.

[0058] The extracted characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging include but are not limited to mean, standard deviation, peak value, voltage at peak value, maximum value, minimum value, etc.

[0059] Wherein, the correlation coefficient between the characteristics of the IC curve and the ΔQ(V) curve and the discharge capacity:

[0060]

[0061] Wherein, F CellCycle_i is the characteristic of the IC or ΔQ(V) curve of the i-th cycle of the lithium battery; C disCycle_i is the discharge capacity of the i-th cycle of the lithium battery; μ F_CellCycle , σ F_CellCycl are the mean and standard deviation of all cycles F CellCycle_i ; μ Cell_disC , σ Cell_disC are the mean and standard deviation of all cycles C disCycle_i .

[0062] The least squares support vector regression is used to determine the relationship between the lithium battery discharge capacity and the corresponding IC and AQ(V) curve characteristics with the lithium battery discharge capacity as the abscissa, the purpose being to ignore points with large deviations and to perform regression fitting on the main concentrated points. Then, the discharge capacity equidivided points and the corresponding characteristics are generated according to the least squares support vector regression results.

[0063] In step S103, the lithium battery pack working time one charge capacity-voltage curve is acquired online, and the capacity-voltage curve is smoothed to remove noise; the derivative of the voltage of the smoothed lithium battery pack charge and discharge capacity-voltage curve is calculated to obtain the IC curve of the lithium battery pack; the capacity of the capacity-voltage curve is subtracted from the capacity of the first cycle capacity-voltage curve to obtain the AQ(V) curve of the lithium battery pack;

[0064] The IC curve and the AQ(V) curve of the lithium battery pack are smoothed to remove noise; multiple characteristics of the IC and the AQ(V) curve of the lithium battery pack are extracted, including but not limited to mean, variance, peak value, voltage at the peak value, maximum value, minimum value, etc.

[0065] The discharge capacity of the corresponding lithium battery pack is calculated by linear interpolation calculation using the multiple characteristics of the battery pack IC and AQ(V) curve and combining the corresponding characteristics of the tested lithium battery. Before the linear interpolation calculation, the IC and AQ(V) curve characteristics of the parallel battery pack need to be converted to the lithium battery level, and the conversion formula is:

[0066] IC cell_peakVol = IC p_peakVol ;

[0067]

[0068] Wherein, m is the number of parallel batteries of the parallel battery pack, IC p_mean , IC cell_mean is the mean of the IC curve of the parallel battery pack and the mean of the IC curve on the corresponding battery level, IC p_peak , IC cell_peak is the peak value of the IC curve of the parallel battery pack and the peak value of the IC curve on the corresponding battery level, IC p_peakVol , IC cell_peakVol is the voltage at the peak value of the IC curve of the parallel battery pack and the voltage at the peak value of the IC curve on the corresponding battery level, IC p_var , IC cell_var is the variance of the IC curve of the parallel battery pack and the variance of the IC curve on the corresponding battery level, IC p_max , IC cell_max is the maximum value of the IC curve of the parallel battery pack and the maximum value of the IC curve on the corresponding battery level, ICp_min , IC cell_min is the minimum value of the parallel battery pack IC curve and the minimum value of the IC curve at its corresponding battery level, IC p_var , IC cell_var is the variance of the parallel battery pack IC curve and the variance of the IC curve at its corresponding battery level, AQ p_mean (V), AQ cell_mean (V) is the mean of the parallel battery pack AQ(V) curve and the mean of the AQ(V) curve at its corresponding battery level, AQ p (V1), AQ cell (V1) is the value of the parallel battery pack AQ(V) curve at V1 and the value of the AQ(V) curve at V1 at its corresponding battery level, V1 is an arbitrarily chosen battery terminal voltage value, AQ p_var (V), AQ cell_var (V) is the variance of the parallel battery pack AQ(V) curve and the variance of the AQ(V) curve at its corresponding battery level, AQ p_max (V), AQ cell_max (V) is the maximum value of the parallel battery pack AQ(V) curve and the maximum value of the AQ(V) curve at its corresponding battery level, AQ p_min (V), AQ cell_min (V) is the minimum value of the parallel battery pack AQ(V) curve and the minimum value of the AQ(V) curve at its corresponding battery level.

[0069] For series battery packs, the battery management system can directly collect battery level data, and can directly calculate the features of the battery level IC and AQ(V) curves in the series battery pack, without the above conversion.

[0070] A linear interpolation calculation is performed using multiple features of the battery pack IC and AQ(V) curves in combination with the corresponding features of the tested lithium battery.

[0071] For F Pack_Cell , the minimum value of F CellCycle_i and the maximum value of F CellCycle_i , or F Pack_Cell , the minimum value of F CellCycle_i and the maximum value of F CellCycle_i , but the maximum value of F CellCycle_i is located at the beginning and end of the F CellCycle_i data, the formula for linear interpolation is:

[0072]

[0073] where F Pack_Cell is a battery pack feature converted to a battery level feature, C Pack_Cell is the value of FPack_Cell F CellCycle_1 , CellCycle_2 F Pack_Cell are two lithium battery characteristics, C CellCycle_1 , CellCycle_2 F CellCycle_1 , CellCycle_2 are the corresponding lithium battery discharge capacities; F CellCycle_i is a characteristic of the IC or ΔQ(V) curve of the i-th cycle of the lithium battery.

[0074] If the extreme value of F CellCycle_i is located in the middle of its data sequence, then C Pack_Cell is calculated using the following formula:

[0075]

[0076] where F Pack_Cell is a battery pack characteristic converted to a characteristic at the battery level, C Pack_Cell is the discharge capacity at the battery level calculated by linear interpolation of F Pack_Cell , F CellCycle_1 , CellCycle_2 F Pack_Cell are two lithium battery characteristics, C CellCycle_1 , CellCycle_2 F CellCycle_1 , CellCycle_2 are the corresponding lithium battery discharge capacities; F CellCycle_i is a characteristic of the IC or ΔQ(V) curve of the i-th cycle of the lithium battery; TempC Pack_Cell_1 and TempC Pack_Cell_2 are both intermediate parameters.

[0077] In step S104, the discharge capacities calculated by different characteristics are clustered using density-based noisy space clustering; the optimal cluster is selected, and the criterion for selecting the optimal cluster is that the discharge capacity calculation value C Pack_Cell corresponding to the characteristic F Pack_Cell in the cluster is the most.

[0078] The calculation formula of the weight is:

[0079] W i = N(|p Cell_F_C ||1, s G )

[0080] where N() represents a Gaussian probability density function. |p Cell_F_C | is the absolute value of the correlation coefficient of the characteristic corresponding to the calculation result in the optimal cluster. The mean of the Gaussian probability density function is 1, and sG is the standard deviation of the Gaussian probability density function.

[0081] is the standard deviation of the Gaussian probability density function. G The following formula can be used:

[0082]

[0083] where σ Cell_ρ is the standard deviation of |p Cell_F_C |, μ Cell_ρ is the mean of |p Cell_F_C |, and |p Cell_F_C | is the absolute value of all correlation coefficients of the extracted features. Then, W i is normalized as follows:

[0084]

[0085] Then, the weighted average of the calculated values is obtained as follows:

[0086]

[0087] where C Pack_Cel_i only considers the discharge capacity calculated value contained in the optimal cluster.

[0088] The above is only the calculation result of the lithium battery pack at the battery level discharge capacity using the aging data of a test lithium battery. Multiple test lithium batteries are used to calculate the average value as the optimal calculation result:

[0089]

[0090] Then, the final calculation value of the discharge capacity of the lithium battery pack is calculated. For parallel battery packs, the calculation result of the lithium battery pack at the battery level discharge capacity needs to be converted to the battery pack level:

[0091]

[0092] For series battery packs, the minimum discharge capacity of the cells in the series battery pack is taken as the discharge capacity of the series battery pack:

[0093]

[0094] In step S105, the calculation formula of the health status of the current lithium battery pack is as follows:

[0095]

[0096] where C rated is the rated capacity of the battery pack.

[0097] Figure 2 is a structure diagram of a battery pack health state online detection device based on multi-feature fusion in an embodiment of the present application, and the embodiment corresponds to the battery pack health state online detection method based on multi-feature fusion in the embodiment of the present application, as shown in Figure 1 , the battery pack health state online detection device based on multi-feature fusion in the embodiment can include: Figure 2

[0098] The aging experiment test module 201 is configured to test the charge-discharge capacity-voltage curves of a plurality of lithium batteries through an aging charge-discharge cycle experiment, calculate the IC curve of the lithium battery aging, and obtain the ΔQ(V) curve of the lithium battery aging by subtracting the capacity of the first cycle charge-discharge capacity-voltage curve from the capacity of the charge-discharge capacity-voltage curve of the lithium battery;

[0099] The aging test feature extraction module 202 is configured to extract the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging, utilize the least squares support vector to regress the relationship between the characteristic values of the IC curve and the ΔQ(V) curve of the lithium battery aging and the discharge capacity, and generate the discharge capacity equant point and the corresponding feature;

[0100] The linear interpolation calculation module 203 is configured to obtain the charge-discharge capacity-voltage curve of the lithium battery pack in operation online, calculate the corresponding IC curve and ΔQ(V) curve, extract the characteristic values of the current IC curve and ΔQ(V) curve, perform linear interpolation calculation according to the discharge capacity equant point of the lithium battery aging and the corresponding feature, and obtain the discharge capacity of the current lithium battery pack;

[0101] The discharge capacity calculation module 204 is configured to cluster and select the optimal cluster of the discharge capacity calculated by different features, utilize the semi-Gaussian distribution to calculate the weight according to the correlation coefficient corresponding to the discharge capacity calculation value in the optimal cluster, and take the weighted average value of the discharge capacity calculation value as the final calculation value of the lithium battery pack discharge capacity;

[0102] The health state calculation module 205 is configured to calculate the health state of the current lithium battery pack according to the final calculation value of the discharge capacity of the lithium battery pack and the rated capacity.

[0103] It should be noted that, Figure 2 the modules in the battery pack health state online detection device based on multi-feature fusion in the embodiment correspond to the steps in the battery pack health state online detection method based on multi-feature fusion in the embodiment, and the specific implementation process is the same, which will not be repeated here. Figure 1

[0104] ​​In one or more embodiments, the electronic device includes a central processing unit (CPU) which can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) or a program loaded into a random access memory (RAM) from a storage section. In the RAM, various programs and data required for system operation are also stored. The central processing unit, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0105] Connected to the I / O interface are an input section including a keyboard, a mouse, etc.; an output section including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section including a hard disk, etc.; and a communication section including a network interface card such as a local area network (LAN) card, a modem, etc. The communication section performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as necessary. A removable media such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive as necessary, so that a computer program read out therefrom is installed into the storage section as necessary.

[0106] The central processing unit in the electronic device of the present embodiment, when executing the program, realizes the steps in the method for online detection of state of health of a battery pack based on multi-feature fusion as shown in Figure 1

[0107] In particular, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the method as shown in Figure 1 In such embodiments, the computer program can be downloaded and installed from a network via the communication section, and / or installed from a removable media. When the computer program is executed by the central processing unit, various functions defined in the apparatus of the present application are performed.

[0108] The computer program instructions corresponding to the method as shown in Figure 1 may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction apparatus, which realizes the functions specified in the flow Figure 1 one flow or multiple flows and / or the functions specified in one block or multiple blocks. Figure 1 one flow or multiple flows and / or the functions specified in one block or multiple blocks.

[0109] ​Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, the program can include the processes of the above-mentioned embodiment methods. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM) or the like.

[0110] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement and the like within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for online detection of battery health status based on multi-feature fusion, characterized in that: include: The charge and discharge capacity-voltage curves of several lithium batteries were tested through aging charge and discharge cycle experiments, and the IC curves of lithium battery aging were calculated; The ΔQ(V) curve of lithium battery aging is obtained by subtracting the capacity of the charge-discharge capacity-voltage curve of the first cycle from the capacity of the charge-discharge capacity-voltage curve of the lithium battery; Extract the characteristic values ​​of the IC curve and ΔQ(V) curve of the lithium battery aging, use the least squares support vector regression to reconstruct the relationship between the characteristic values ​​of the IC curve and ΔQ(V) curve of the lithium battery aging and the discharge capacity, and generate the discharge capacity equal points and their corresponding characteristics; Online acquisition of the charge and discharge capacity-voltage curves of the lithium battery pack during operation, calculation of the corresponding IC curve and ΔQ(V) curve, extraction of the characteristic values ​​of the current IC curve and ΔQ(V) curve, and linear interpolation calculation based on the discharge capacity equalization points of the lithium battery aging and their corresponding characteristics to obtain the current discharge capacity of the lithium battery pack; Clustering the discharge capacities calculated based on different features and selecting the optimal cluster; calculating weights using a semi-Gaussian distribution based on the correlation coefficients corresponding to the discharge capacity calculation values ​​in the optimal cluster, and taking the weighted average of the discharge capacity calculation values ​​as the final calculated discharge capacity of the lithium battery pack; Calculate the current health status of the lithium battery pack based on the final calculated value of the discharge capacity and the rated capacity of the lithium battery pack.

2. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: Perform aging cycle tests on several new lithium batteries until the battery capacity is less than 70% of the rated capacity; the charging conditions are similar to the actual conditions of the calculated lithium battery pack; the discharge conditions are set according to the standard lithium battery capacity test conditions; Record the charge capacity, voltage and discharge capacity of each cycle of the lithium battery.

3. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: The charge and discharge capacity-voltage curves of several lithium batteries tested through aging charge and discharge cycle experiments are smoothed, and the voltage derivative of the smoothed lithium battery charge and discharge capacity-voltage curves is calculated to obtain the IC curve of lithium battery aging.

4. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: Correlation coefficient between the characteristics of IC curve and ΔQ(V) curve and discharge capacity: ； in, F CellCycle_i For lithium batteries i IC or Δ of the next cycle Q ( V ) characteristics of the curve; C disCycle_i For the i The discharge capacity of lithium battery in cycles; μ F_CellCycle 、 σ F_CellCycl All loops are F CellCycle_i The mean and standard deviation of μ Cell_disC 、 σ Cell_disC All loops are C disCycle_i The mean and standard deviation of .

5. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: for F Pack_Cell Greater than or equal to F CellCycle_i The minimum value of and less than or equal to F CellCycle_i The maximum value of F Pack_Cell Less than F CellCycle_i The minimum value of F CellCycle_i The maximum value of F CellCycle_i The maximum value is at F CellCycle_i At the beginning and end of the data, the linear interpolation formula is: in F Pack_Cell Convert a battery pack feature into a battery level feature. C Pack_Cell for the reason F Pack_Cell The discharge capacity of the lithium battery pack at the corresponding battery level calculated by linear interpolation, F CellCycle_1 、 F CellCycle_2 For the closest F Pack_Cell Two lithium battery characteristics, C CellCycle_1 、 C CellCycle_2 for F CellCycle_1 、 F CellCycle_2 Corresponding lithium battery discharge capacity; F CellCycle_i For lithium batteries i IC or Δ of the next cycle Q ( V ) characteristics of the curve.

6. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: if F CellCycle_i The maximum value of is in the middle of its data sequence, then use the following formula to calculate C Pack_Cell : in F Pack_Cell Convert a battery pack feature into a battery level feature. C Pack_Cell for the reason F Pack_Cell The discharge capacity of the lithium battery pack at the corresponding battery level calculated by linear interpolation, F CellCycle_1 、 F CellCycle_2 、 F CellCycle_3 、 F CellCycle_4 For the closest F Pack_Cell Four lithium battery characteristics, C CellCycle_1 、 C CellCycle_2 、 C CellCycle_3 、 C CellCycle_4 for F CellCycle_1 、 F CellCycle_2 、 F CellCycle_3 、 F CellCycle_4 Corresponding lithium battery discharge capacity; F CellCycle_i For lithium batteries i IC or Δ of the next cycle Q ( V ) characteristics of the curve; and All are intermediate parameters.

7. The method for online detection of battery health status based on multi-feature fusion according to claim 1, characterized in that: The weight calculation formula is: in, represents the Gaussian probability density function, | ρ Cell_F_C | is the absolute value of the correlation coefficient of the feature corresponding to the calculation result in the optimal clustering, and the mean of the Gaussian probability density function is 1. σ G is the standard deviation of the Gaussian probability density function.

8. An online detection device for battery pack health status based on multi-feature fusion, characterized in that: include: An aging test module is used to test the charge-discharge capacity-voltage curves of several lithium batteries through an aging charge-discharge cycle experiment, and calculate the IC curve of the lithium battery aging; the ΔQ(V) curve of the lithium battery aging is obtained by subtracting the capacity of the charge-discharge capacity-voltage curve of the first cycle from the capacity of the charge-discharge capacity-voltage curve of the lithium battery; An aging test feature extraction module is used to extract the characteristic values ​​of the IC curve and ΔQ(V) curve of the lithium battery aging, and use the least squares support vector regression to reconstruct the relationship between the characteristic values ​​of the IC curve and ΔQ(V) curve of the lithium battery aging and the discharge capacity, and generate the discharge capacity equal points and their corresponding features; A linear interpolation calculation module is used to online obtain the charge and discharge capacity-voltage curves of the lithium battery pack during operation, calculate the corresponding IC curve and ΔQ(V) curve, extract the characteristic values ​​of the current IC curve and ΔQ(V) curve, and perform linear interpolation calculations based on the discharge capacity equalization points of the lithium battery aging and their corresponding characteristics to obtain the current discharge capacity of the lithium battery pack; A discharge capacity calculation module is used to cluster the discharge capacities calculated based on different features and select the optimal cluster; weights are calculated using a semi-Gaussian distribution based on the correlation coefficients corresponding to the discharge capacity calculated values ​​in the optimal cluster, and the weighted average of the discharge capacity calculated values ​​is used as the final calculated value of the lithium battery pack discharge capacity; The health status calculation module is used to calculate the current health status of the lithium battery pack based on the final calculated value of the discharge capacity and the rated capacity of the lithium battery pack.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method for online detection of battery pack health status based on multi-feature fusion as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps in the method for online detection of battery pack health status based on multi-feature fusion as described in any one of claims 1 to 7 are implemented.

Images