CN116879784A

CN116879784A - Method, device, equipment and storage medium for determining battery health state

Info

Publication number: CN116879784A
Application number: CN202310806917.XA
Authority: CN
Inventors: 周国鹏; 魏琼; 严晓; 赵恩海; 赵健; 冯洲武
Original assignee: Shanghai MS Energy Storage Technology Co Ltd
Current assignee: Shanghai MS Energy Storage Technology Co Ltd
Priority date: 2023-07-03
Filing date: 2023-07-03
Publication date: 2023-10-13

Abstract

The invention provides a method, a device, equipment and a storage medium for determining the health state of a battery, wherein the method comprises the following steps: acquiring historical working condition data and health states acquired by a battery to be tested in a plurality of different historical time periods; respectively determining m historical features corresponding to each historical working condition data; taking all the historical features of the same type as a historical feature set, iteratively updating and determining the influence degree of the selected historical feature set on the health state in a mode of iteratively selecting a better historical feature set, and taking the influence degree at the end of iteration as a correlation coefficient between the features and the health state; and determining the health state of the battery to be tested in the current period according to all the current characteristics and the association coefficients. The method, the device, the equipment and the storage medium for determining the battery state of health provided by the embodiment of the invention have the advantages that the association coefficient between the feature and the SOH can be determined more quickly without iteration for a plurality of times, and the processing efficiency is high.

Description

Method, device, equipment and storage medium for determining battery health state

Technical Field

The present invention relates to the field of battery management technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining a battery health status.

Background

In recent years, as the new energy field is receiving more and more attention, the lithium battery has the characteristics of high energy density, long cycle life and the like, and becomes the first choice energy source of the new energy automobile and the energy storage power station. However, as the lithium battery is used for a long time, the performance of the lithium battery is severely restricted by problems such as the service life and capacity degradation of the lithium battery. If the SOH (state of health) of the lithium battery can be accurately calculated, or the future SOH change trend of the battery can be predicted, the method plays an important role in prolonging the service life of the battery and playing the role of the battery.

At present, a special device is needed to test the SOH of a lithium battery, and the SOH of the battery can be accurately analyzed by carrying out complete full charge or full discharge activities on the lithium battery. When the SOH of the lithium battery is low (for example, near or less than 80%), if the lithium battery is fully charged or fully discharged to determine the SOH, the capacity degradation of the lithium battery is aggravated, the service life of the lithium battery is reduced, and the service life of the lithium battery in practical application is obviously reduced.

Some schemes use neural networks and the like to calculate SOH, but the neural networks are complex, so that the calculation process is complex, the calculation efficiency is low, and the hardware occupies large resources.

Disclosure of Invention

In order to solve the existing technical problems, the embodiment of the invention provides a method, a device, equipment and a storage medium for determining the health state of a battery.

In a first aspect, an embodiment of the present invention provides a method for determining a state of health of a battery, including:

acquiring historical working condition data acquired by a battery to be tested in a plurality of different historical time periods, and determining the health state of each historical time period;

respectively extracting features of each piece of history working condition data, and determining m pieces of history features corresponding to each piece of history working condition data, wherein m is more than or equal to 2;

taking the same historical feature extracted from all the historical working condition data as a historical feature set, iteratively updating and determining the influence degree of the selected historical feature set on the health state in a mode of iteratively selecting the better historical feature set according to the health state and the historical feature set, and taking the influence degree determined at the end of iteration as a correlation coefficient between the feature and the health state; wherein the iteration round does not exceed m;

acquiring current working condition data of the battery to be tested in a current period, extracting a plurality of current features related to the association coefficient, and determining the health state of the battery to be tested in the current period according to all the current features and the association coefficient.

In a second aspect, an embodiment of the present invention further provides an apparatus for determining a health status of a battery, including:

the acquisition module is used for acquiring historical working condition data acquired by the battery to be tested in a plurality of different historical time periods and determining the health state of each historical time period;

the extraction module is used for respectively carrying out feature extraction on each piece of history working condition data and determining m pieces of history features corresponding to each piece of history working condition data, wherein m is more than or equal to 2;

the iteration module is used for taking the historical features of the same type extracted from all the historical working condition data as a historical feature set, iteratively selecting the better one of the historical feature sets according to the health state and the historical feature set, iteratively updating and determining the influence degree of the selected historical feature set on the health state, and taking the influence degree determined at the end of iteration as a correlation coefficient between the features and the health state; wherein the iteration round does not exceed m;

the calculation module is used for acquiring current working condition data of the battery to be tested in the current period, extracting a plurality of current features related to the association coefficient, and determining the health state of the battery to be tested in the current period according to all the current features and the association coefficient.

In a third aspect, an embodiment of the present invention provides an apparatus for determining a state of health of a battery, including a processor and a memory, where the memory stores a computer program, and the processor executes the computer program stored in the memory, where the computer program is executed by the processor to implement the method for determining a state of health of a battery according to the first aspect.

In a fourth aspect, an embodiment of the present invention further provides a computer readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the method for determining a state of health of a battery according to the first aspect.

According to the method, the device, the equipment and the storage medium for determining the health state of the battery, the same historical feature is used as one historical feature set, and the influence degree of the selected historical feature set on the health state is determined through iterative updating in a mode of iteratively selecting a better one of the historical feature sets, so that the influence degree of each historical feature set on SOH can be determined under the condition that the number of iterations does not exceed the number m of feature types, namely the correlation coefficient between each feature and SOH can be determined, and the SOH of the battery to be measured can be calculated conveniently in real time. The method does not need to iterate for a plurality of times, can determine the association coefficient between the feature and the SOH more quickly, and has high processing efficiency; and when the SOH of the battery to be measured needs to be determined, the battery to be measured does not need to be fully charged and discharged, the SOH of the battery to be measured can be calculated at any time, the practicability is high, and the practical application of the battery is not affected.

Drawings

In order to more clearly describe the embodiments of the present invention or the technical solutions in the background art, the following description will describe the drawings that are required to be used in the embodiments of the present invention or the background art.

FIG. 1 is a flow chart of a method for determining battery health according to an embodiment of the present invention;

FIG. 2 is a schematic diagram showing the relationship between capacity difference and voltage provided by an embodiment of the present invention;

FIG. 3 is a graph showing a comparison of the results of calculating and testing SOH provided by an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an apparatus for determining a state of health of a battery according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of an apparatus for determining a state of health of a battery according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

Fig. 1 is a flowchart of a method for determining a state of health of a battery according to an embodiment of the present invention. As shown in fig. 1, the method includes:

step 101: and acquiring historical working condition data acquired by the battery to be tested in a plurality of different historical time periods, and determining the health state of each historical time period.

In the embodiment of the invention, a battery that needs to determine a health state (may be simply referred to as SOH) is referred to as a battery to be tested. In the running process of the battery to be tested, the SOH of the battery to be tested is gradually reduced along with the increase of charge and discharge times; in order to calculate the SOH of the battery to be measured at any time conveniently, working condition data, namely historical working condition data, acquired by the battery to be measured in a previous historical period are firstly acquired, and the SOH corresponding to the corresponding historical working condition data, namely the SOH of the historical period, needs to be determined.

It can be understood that the history period is a period of time, which may specifically be a period of time during which the battery to be measured is charged once, discharged once, or charged and discharged once; correspondingly, the historical working condition data are data acquired in the time period.

Where the SOH for each history period can be determined using existing maturation techniques. Because the battery to be tested is in the normal operation process, the condition of full charge or full discharge exists; in addition, the battery under partial scenes has a capacity calibration experiment, which can calibrate the capacity under certain charge and discharge times, and can be fully charged and discharged at the moment; for example, the battery to be tested is fully charged and discharged to 100 circles, 200 circles, 300 circles and the like. Since it is easier to determine the SOH of the battery under test when it is fully charged, the historical period may be the period of time before the battery under test is fully charged or fully discharged.

Step 102: and respectively extracting features of each piece of history working condition data, and determining m pieces of history features corresponding to each piece of history working condition data, wherein m is more than or equal to 2.

In the embodiment of the invention, after a plurality of pieces of history working condition data of the battery to be tested are determined, feature extraction can be carried out on each piece of history working condition data, and corresponding features, namely history features, are extracted. The history feature may be any selected feature, or may be a feature related to SOH determined based on human experience, which is not limited in the embodiment of the present invention. In addition, in order to accurately calculate the SOH of the battery to be measured, various characteristics, namely the number m of the types of the history characteristics, need to be collected, and the number m is more than or equal to 2.

Specifically, for each history working condition data, m different kinds of history features are extracted, namely, each history working condition data corresponds to a set of history features, and each set comprises the m history features; if n pieces of history working condition data of the battery to be tested are obtained in advance, n sets, m kinds of characteristics and n multiplied by m history characteristics can be extracted in total.

For example, in x _i(j) Representing the jth historical feature (i=1, 2, …, n, j=1, 2, …, m) extracted from the ith historical operating condition data, the set corresponding to the ith historical operating condition data may be represented as: [ x ] _i(1) …x _i(m) ]. For example, the historical features of all historical operating condition data may be represented by matrix X, and:wherein each row of the matrix X represents m historical features extracted from one historical operating condition data, each column representing n values extracted over n different historical periods for a certain historical feature.

Step 103: taking the historical features of the same kind extracted from all the historical working condition data as a historical feature set, iteratively updating and determining the influence degree of the selected historical feature set on the health state in a mode of iteratively selecting a better one of the historical feature sets according to the health state and the historical feature set, and taking the influence degree determined at the end of iteration as a correlation coefficient between the features and the health state; wherein the iteration turns do not exceed m.

In the embodiment of the invention, the relation between the characteristic and the health state is expressed by the relation coefficient between the characteristic and the health state, and after the relation coefficient is determined, the corresponding health state, namely SOH, can be determined based on the known characteristic. Each feature has a corresponding association coefficient, that is, m association coefficients can be determined altogether; let the m correlation coefficients be in turn: k (k) ₁ 、k ₂ 、…、k _m SOH of the i-th history period is y _i The method comprises the steps of carrying out a first treatment on the surface of the And as described above, the set of history features extracted from the history condition data of the i-th history period is: [ x ] _i(1) x _i(m) ]The relationship between SOH and history features for the history period is: y is _i ＝k ₁ x _i(1) +k ₂ x _i(2) +…+k _m x _i(m) +c; c represents a constant.

It will be appreciated that the correlation coefficient k _i Is an unknown quantity to be determined, and in the case that the matrices X and Y are known, the matrix K can be determined by linear fitting or the like, so as to determine all the correlation coefficients. However, the modes of linear fitting and the like need to go through more rounds of iterative processes, generally require hundreds of iterations to converge, and have lower efficiency. In the embodiment of the invention, the same type of characteristics (such as a column of characteristics X, n characteristics in each type) are taken as a group, m groups are taken, the association coefficient corresponding to each group of characteristics is respectively determined, and the iteration process needs to be iterated for m times at most, namely, the iteration round is m at most, so that the iteration round can be greatly reduced, and the processing efficiency is improved. For example, if 10 features need to be extracted from the operating condition data, that is, m=10, the iteration number is at most 10 (the actual iteration number may be less than 10), and the iteration number is less.

For convenience of description, the same historical feature extracted from all the historical working condition data is taken as a historical feature set, i.e. the historical feature set contains all the historical features of the same kind. As described above, if there are n history times in total, the history feature set includes n history features of the same type; and, the category number of the history feature is m, and m history feature sets can be determined in total.

For example, each column of the matrix X corresponds to a set of historical features; that is, the history feature set corresponding to the i-th history feature can be expressed as:where i=1, 2, …, m.

In the embodiment of the invention, because the history feature set comprises similar history features at a plurality of history moments, after the history feature set is selected, the influence degree of the history feature set on SOH can be determined based on the data contained in the history feature set; and with the progress of iteration, the influence degree of each selected historical feature set can be gradually updated, so that the finally determined influence degree can more accurately represent the corresponding relation between the features and the SOH.

Specifically, in each iteration process, selecting the better one of the historical feature sets, namely selecting the historical feature set corresponding to the feature most relevant to the SOH, and determining the influence degree of each historical feature set through multiple iterations. For example, embodiments of the present invention relate to 10 features altogether, i.e., m=10, a total of 10 historical feature sets can be determined; in the first iteration, an optimal historical feature set, e.g. the 1 st historical feature set X, is selected ₍₁₎ And determines the 1 st historical feature set X ₍₁₎ Is a degree of influence of (a); at the second iteration, if the 3 rd historical feature set X ₍₃₎ Optimally, then select 3 rd historical feature set X ₍₃₎ And determining a historical feature set X ₍₁₎ And X ₍₃₎ The degree of influence (X) ₍₁₎ The degree of influence of (c) is updated, i.e. changed); at the third iteration, if the 4 th historical feature set X ₍₃₎ Optimally, then select the 4 th historical feature set X ₍₄₎ And determining a historical feature set X ₍₁₎ 、X ₍₃₎ 、X ₍₄₎ And so on. It can be seen that the iteration runs are at most m times, if each iteration is optionalTaking more than one preferred set of historical features, or after a sufficient number (less than m) of sets of historical features have been selected that can represent SOH, the iteration may also be stopped, i.e., the number of iterations is less than m.

After the influence degree corresponding to each historical feature set is finally determined, the influence degree can be used as the association coefficient k between each feature and SOH _i . That is, in the case where the number of iterations does not exceed m, the correlation coefficient k between each feature and SOH can be determined _i The treatment efficiency is high. In practical application, whenever SOH of the battery to be measured can be determined in other manners, for example, when the battery to be measured is fully charged and discharged, the working condition data collected at the moment can be used as new historical working condition data, and the correlation coefficient k is calculated _i And fast updating is carried out, so that higher precision is kept all the time when the SOH is calculated subsequently.

Step 104: acquiring current working condition data of the battery to be tested in a current period, extracting a plurality of current features related to the association coefficient, and determining the health state of the battery to be tested in the current period according to all the current features and the association coefficient.

In the embodiment of the invention, the association coefficient k corresponding to all the features is determined _i And then, calculating the SOH of the battery to be tested under the condition that the battery to be tested is not fully charged and discharged. Specifically, in a current period in which SOH needs to be calculated, working condition data of the current period, that is, current working condition data, may be collected, and the current working condition data may not be data when full charge or full discharge is performed; extracting corresponding features, namely current features, from the current working condition data in the same mode of extracting historical features; for example, m current features may be extracted altogether. After extracting the current features, the SOH corresponding to the current features, i.e. the health status of the current period, can be calculated based on all the association coefficients that have been determined.

It should be noted that if m features can affect the SOH of the battery, each feature corresponds to the correlation coefficient k _i Is non-zero; conversely, if a certain correlation coefficient k _i If the value is 0, the corresponding feature is irrelevant to SOH, and the feature must be extracted when the feature is extracted from the current working condition data; i.e. Only a plurality of current features related to the non-zero association coefficient may be extracted.

According to the method for determining the state of health of the battery, the same historical feature is used as one historical feature set, and the influence degree of the selected historical feature set on the state of health is determined through iterative updating in a mode of iteratively selecting a better historical feature set, so that the influence degree of each historical feature set on SOH can be determined under the condition that the iterative times do not exceed the feature type number m, namely the correlation coefficient between each feature and SOH can be determined, and the SOH of the battery to be measured can be calculated conveniently in real time. The method does not need to iterate for a plurality of times, can determine the association coefficient between the feature and the SOH more quickly, and has high processing efficiency; and when the SOH of the battery to be measured needs to be determined, the battery to be measured does not need to be fully charged and discharged, the SOH of the battery to be measured can be calculated at any time, the practicability is high, and the practical application of the battery is not affected.

Optionally, the step 103 "determining the influence degree of the selected history feature set on the health status by iteratively selecting the better one of the history feature sets" may specifically include the following steps A1 to A4. Wherein, because of the need of multiple iterations, any iteration process in any round can be processed according to steps A1 to A4. The iterative process of the g-th round will be mainly described below, and it will be appreciated that g does not exceed m.

Step A1: in the iteration process of the g round, determining a first influence value of each residual history feature set on the residual matrix determined in the previous round; the remaining historical feature set is a historical feature set which is not selected in the previous iteration process; the initial value of the residual matrix is a health state matrix containing all health states.

Step A2: taking the remaining historical feature set with the largest absolute value of the first influence value as a selected historical feature set, and adding the selected historical feature set into a selection set matrix; the selection set matrix includes the historical feature sets selected in all previous iterations.

In the embodiment of the invention, in each iteration process, a better historical feature set is selected, and accordingly, unselected historical feature sets exist; for convenience of description, the history feature set that is not selected is referred to as "remaining history feature set". In each iteration process, calculating a first influence value of the residual historical feature set on the residual matrix determined in the previous iteration process, and selecting a better historical feature set based on the first influence value. It will be appreciated that in round 1 of the iterative process, each historical feature set has not been selected, and is therefore the remaining historical feature set. The residual error matrix represents the rest part after the influence of the selected history feature set on SOH is removed from the health state matrix; accordingly, in the 1 st round of iteration process, the residual matrix is the health state matrix, that is, the initial value of the residual matrix is the health state matrix. Each iteration process may determine a corresponding residual matrix, which is described in detail later.

The health state matrix comprises health states corresponding to all historical moments. For example, SOH of the ith history period is y _i The matrix containing all the history periods SOH is represented by a matrix Y, and the health state matrix Y can be represented as:

in the embodiment of the invention, the first influence value can represent the influence degree of the residual history feature set on the SOH, and the larger the absolute value of the first influence value is, the larger the influence of the residual history feature set on the SOH is, and the more the residual history feature set is optimal when the SOH is calculated. Therefore, the remaining historical feature set with the largest absolute value of the first influence value is used as the optimal historical feature set of the g round of iteration, and the remaining historical feature set can be used as the selected historical feature set.

And all the selected history feature sets are added into the selection set matrix, namely the selection set matrix not only comprises the history feature sets selected by the g th round, but also comprises the history feature sets selected by the previous iterative process (namely the g-1 th round, the g-2 nd round and the like).

For example, in X _new Representing the pick set matrix, which is initially empty. In the 1 st iteration, if the a-th historical feature set X _(a) For being selectedHistorical feature set, then the pick set matrixIn the 2 nd iteration, if the b-th historical feature set X _(b) For the selected historical feature set, it is added to the selection matrix, in which case the selection matrix +.>Which is an n x 2 matrix, each column corresponding to a selected set of historical features.

Optionally, the first influence value is calculated as a dot product. Specifically, the first influence value satisfies:

w _i (1)＝X' _(i) ·R _g-1 (1)

wherein X 'is' _(i) Representing the i-th historical feature set after standardization, wherein the i-th historical feature set is currently the remaining historical feature set; r is R _g-1 Representing the residual matrix determined in round g-1; w (w) _i (1) Representing the i-th set of historical features X _(i) Residual matrix R for round g-1 _g-1 Is a first influence value of (1); is a dot product inner product operation.

In the embodiment of the invention, because each characteristic in the battery has larger difference and generally different orders of magnitude, when the first influence value of the residual historical characteristic set is calculated, the residual historical characteristic set is standardized so as to unify the historical characteristic sets corresponding to different characteristics under the same order of magnitude. For example, the normalization may be to normalize the set of historical features corresponding to different features to the same range, e.g., 0-1; alternatively, the normalization may be that the sum of all the history features in the history feature set is normalized to 1, and the embodiment of the present invention does not limit the normalization manner.

Specifically, if the i-th historical feature set X _(i) For the remaining historical feature set, for the ith historical feature set X _(i) After normalization, the i-th historical feature set X 'after normalization can be obtained' _(i) 。

It will be appreciated that the above formula (1) can also be usedMatrix multiplication. Specifically, if X' _(i) And R is _g-1 All are n×1 matrices, then w _i (1)＝X' _(i) ^T ×R _g-1 。

Step A3: setting a second influence value for each historical feature set in the selected set matrix, and updating the second influence value by minimizing the difference between the health state matrix and the selected set matrix multiplied by the second influence value; the updated second influence value is the influence degree of the corresponding historical feature set of the selection set matrix on the health state.

In the embodiment of the invention, after the selected historical feature sets are determined, each historical feature set is determined, and a second influence value is determined; and determining the second impact value in a minimized form. Specifically, each of the history feature sets in the selection set matrix is a selected history feature set, and an initial second influence value is set for the history feature sets, for example, a first influence value of the history feature sets or a second influence value determined by a previous round is used as the second influence value initially set by the current round (i.e., the g-th round).

The difference between the state of health matrix and the selected set matrix multiplied by the second impact value can represent the difference between the SOH determined with the second impact value as the impact level and the SOH represented by the state of health matrix. Specifically, let the health state matrix be Y and the selected set matrix be X _new The matrix determined by all the second influence values is W ₂ The difference can be expressed as Y-X _new ×W ₂ By minimizing the difference, the matrix W can be determined ₂ I.e. each second influence value. And, the updated second influence value is the influence degree of the corresponding history feature set, and let the matrix of the influence degree be W, w=w ₂ 。

For example, the second impact value may be determined by a least squares update. Specifically, the formula min Y-X is based _new ×W ₂ The matrix W can be determined by the least square method ₂ 。

Wherein, because the least square method is used for solving the problem with long time, the embodiment of the invention can be used for solving in other ways. Specifically, the step A3 "updating the second influence value by minimizing the difference between the health state matrix and the selected set matrix multiplied by the second influence value" may include:

step A31: performing matrix decomposition on the intermediate matrix to determine an inverse matrix of the intermediate matrix; the intermediate matrix is the product of the transpose of the pick set matrix and the pick set matrix.

Step A32: and multiplying the inverse matrix of the intermediate matrix by the transpose of the selection set matrix and multiplying the result of the health state matrix to obtain an updated second influence value.

In the embodiment of the invention, the method is used for controlling the min Y-X _new ×W ₂ Solving of I corresponds to matrix W ₂ The method comprises the following steps: w (W) ₂ ＝(X _new ^T X _new ) ^- ¹ X _new ^T X Y. Transpose X of the set matrix is to be selected _new ^T And a pick set matrix X _new Product of (i.e. X _new ^T X _new ) As the intermediate matrix, an inverse matrix of the intermediate matrix is required. The matrix decomposition, such as LU decomposition, QR decomposition, etc., is performed on the intermediate matrix, so that the inverse matrix of the intermediate matrix can be quickly solved, without determining the inverse matrix by solving the accompanying matrix, etc., and the solving efficiency is high. Thereafter, based on W ₂ ＝(X _new ^T X _new ) ^-1 X _new ^T The matrix W can be determined by X and Y ₂ A second impact value for each selected set of historical features may be determined.

Step A4: determining a residual matrix of the g-th round, and meeting the following conditions: r is R _g ＝Y-X _new X W; ending the iteration under the condition that the residual matrix converges or the current iteration number g is equal to m; wherein R is _g Representing the residual matrix of the g-th round, Y representing the health state matrix, X _new Representing the pick set matrix, W represents a matrix containing all the degrees of influence currently determined.

In the embodiment of the invention, after all influence degrees are determined, a matrix W can be formed; as described above, w=w ₂ . Based on the determined pick set matrix X _new The residual matrix of the current round, i.e. the residual matrix of the g-th round, can be determined: r is R _g ＝Y-X _new X W. It will be appreciated that as the iteration proceeds, a set matrix X is selected _new And the matrix W of the degree of influence will vary. Since the degree of influence (i.e. the updated second influence value) is determined in a minimized form, the residual matrix gradually converges. If the residual matrix is converged in the current round, the iteration is not required to be continued, and the iteration can be ended; if the residual matrix of the current round is not converged, the next round of iteration is carried out, namely, the iteration process of the (g+1) th round is carried out. Furthermore, it will be appreciated that the current round is the last round, e.g. g=m, and the iteration is also ended.

If the difference between the residual matrix of the current round and the residual matrix of the previous round is small, the residual matrix may be considered to be converged. For example, 2-norm R of the residual matrix _g || ₂ Relative to the last iteration _g-1 || ₂ When less than the expected value (e.g., 0.05), the iteration may end.

In the embodiment of the present invention, when the iteration is finished, if the residual matrix determined by the last round is basically zero, only the finally determined association coefficient (i.e. the influence degree determined by the last round) may be used to represent the relationship between the feature and the SOH. For example, if all the correlation coefficients are represented by matrix K, the relationship between matrix X including all the history features and health state matrix Y may be represented as y=xk. If, at the end of the iteration, the residual matrix determined in the last round is not zero, the corresponding intercept value c is determined based on the residual matrix, and the relationship between the feature and SOH may be expressed as y=xk+c; wherein the two norms of the residual matrix may be taken as the intercept value c.

Optionally, the step 103 "determining the influence degree of the selected historical feature set on the health status by iteratively selecting the better historical feature set" further includes a step A5.

Step A5: determining an invalid residual historical feature set according to a preset first influence value standard, and eliminating the invalid residual historical feature set; wherein the absolute value of the ratio between the first influence value of the ineffective residual history feature set and the first influence value with the largest absolute value is smaller than the first influence value reference; the first impact value reference is between 0 and 0.5.

In the embodiment of the invention, when the historical feature set is selected based on the first influence value, the historical feature set with little influence on SOH can be removed based on the first influence value. Specifically, in the iteration process of the g-th round, if the a-th historical feature set has the first influence value w with the largest absolute value _a (1) The b-th historical feature set is the remaining historical feature set, and the first influence value is w _b (1) The method comprises the steps of carrying out a first treatment on the surface of the If the absolute value of the ratio of the two is |w _b (1)/w _a (1) If i is greater than the first impact value reference, then the b-th set of historical features is retained, conversely, if i w _b (1)/w _a (1) And if the I is smaller than the first influence value standard, the b-th historical feature set can be considered as an invalid residual historical feature set, and can be eliminated. The removed historical feature set does not need to be considered in the subsequent iteration process, which is equivalent to the fact that the kicked historical feature set does not exist all the time, so that the iteration times can be reduced, the processing capacity of each iteration process can be reduced, and the iteration progress is quickened.

Optionally, to avoid false positives, only the set of historical features that are considered invalid for successive rounds is culled. Specifically, the step A5 "eliminates the invalid remaining history feature set", including: and when the invalid residual history feature set is determined to be invalid in the continuous multi-round iteration process, eliminating the invalid residual history feature set.

For example, only the remaining set of historical features that are invalid for two consecutive rounds are culled. For example, a certain historical feature set A is considered invalid for the first time in the current round, and the historical feature set A is not rejected; if the historical feature set A is again considered invalid in the next round, the historical feature set A is rejected again.

Optionally, the step 103 "determining the influence degree of the selected historical feature set on the health status by iteratively selecting the better historical feature set" further includes a step A6.

Step A6: in addition to taking the remaining history feature set with the largest absolute value of the first influence value as the selected history feature set, taking the remaining history feature set with the absolute value of the ratio between the first influence value and the first influence value with the largest absolute value larger than the second influence value standard as the selected history feature set; the second impact value reference is between 0.7 and 1.

In the embodiment of the invention, only one residual history feature set, namely the residual history feature set with the largest absolute value of the first influence value, can be selected in each iteration process; alternatively, a plurality of remaining history feature sets may be selected, and a remaining history feature set having an absolute value of the first influence value that is not greatly different from an absolute value of the first influence value having the largest absolute value may be used as the selected history feature set.

For example, if the a-th historical feature set has the first influence value w with the largest absolute value _a (1) The b-th historical feature set is the remaining historical feature set, and the first influence value is w _b (1) The method comprises the steps of carrying out a first treatment on the surface of the If |w _b (1)/w _a (1) And if the I is larger than the second influence value standard, taking the b-th historical characteristic set as the historical characteristic set selected by the current round.

It will be appreciated that the first and second impact value references are two different references and that the first impact value reference is less than the second impact value reference. In an embodiment of the present invention, the first influence value is between 0 and 0.5, for example, 0.382; the second influence value is between 0.7 and 1, for example 0.8, and the value of the second influence value and the second influence value can be determined based on actual conditions.

Optionally, since two similar historical feature sets have similar first influence values, which affect the accuracy of the finally determined association coefficient, in the embodiment of the present invention, the historical feature set with higher similarity is removed in advance, and only one of the two historical feature sets is kept. Specifically, the step 103 "taking the same historical feature extracted from all the historical operating condition data as a historical feature set" may include:

Step B1: and taking the historical characteristics of the same kind extracted from all the historical working condition data as a to-be-determined characteristic set.

Step B2: a similarity between the set of features to be determined and any other set of features to be determined is determined.

Step B3: and taking the feature set to be determined as a historical feature set under the condition that the similarity is smaller than a preset threshold value.

Step B4: and under the condition that the similarity is larger than a preset threshold value, selecting one of the to-be-determined feature set and other to-be-determined feature sets with the similarity larger than the preset threshold value as one historical feature set.

In the embodiment of the invention, if M historical features are extracted from the historical working condition data, M feature sets to be determined can be determined; if the similarity of two (or more) to-be-determined feature sets is higher, only one to be used as the history feature set is selected, and M is less than or equal to M finally obtained. The similarity may be cosine similarity, a jaccard similarity coefficient, and the like.

The flow of the method is described in detail below by way of one embodiment.

In the embodiment of the invention, a certain ternary power battery is used as a battery to be tested, and when the battery works and runs to 100 th, 200 th, 300 th, 400 th, 500 th, 600 th, 700 th, 800 th and 1000 th circles (corresponding to 10 historical moments, namely n=10 th), a capacity calibration test experiment is carried out to obtain SOH when the battery is fully charged and discharged, so that a health state matrix Y can be formed; and extracting corresponding working condition data, namely historical working condition data, extracting corresponding historical characteristics from the historical working condition data, and forming a matrix X of the historical characteristics. These historical features are independent of whether full-fill is put, i.e., features that are also extractable in the case of non-full-fill are taken as historical features.

For example, for each historical operating condition data, determining a capacity corresponding to each voltage, and further determining a corresponding capacity difference, the capacity difference representing a capacity difference between two adjacent voltages; the relationship between capacity difference and voltage can be seen in fig. 2. As shown in fig. 2, there are a plurality of peaks (typically including two distinct peaks) in the capacity difference, and the peaks are generated during charge and discharge, and are not at end positions, i.e., are not features at full charge or full discharge, so some features can be extracted from the peaks as history features. In the embodiment of the invention, besides determining the volume difference, the corresponding voltage of each voltage can be determinedIn turn, 9 features (i.e., m=9) were extracted as historical features: d, d ₁ Peak heightPeak position->Rising to peak point d ₁ Slope of +.>Peak point d ₁ Slope of descent +.>d ₂ Peak height->Peak position->Peak d ₂ To peak d ₁ Total change in volume betweend ₁ Peak corresponding temperature>d ₂ Peak corresponding temperature>

Based on these 9 history features, 9 history feature sets can be generated.

In the 1 st iteration process, the 9 history feature sets are all the remaining history feature sets, and the residual matrix is a health state matrix.

And (3) determining a first influence value corresponding to each historical feature set based on the formula (1), and selecting the historical feature set with the largest absolute value of the first influence value. Suppose the a-th historical feature set X _(a) The absolute value of the first influence value of (2) is the largest, it is addedAdd to the initially empty pick set matrix X _new In, i.e. X _new ＝X _(a) . And calculate the a-th historical feature set X _(a) Is a second influence value of (a). Wherein, can be based on the formula w _a (2)＝X _(a) Y calculates its second impact value w _a (2) Wherein X is _(a) For historical feature set X _(a) In itself, no standardization is required. At this time, the residual matrix R of the first round ₁ Is R ₁ ＝Y-X _(a) ×w _a (2). At this time, the selection set matrix has only one non-empty element X _(a) The method such as least square method is not needed to carry out the minimization treatment.

In the following iteration 2, if the b-th historical feature set X _(b) The first influence value of the first matrix is the largest, it is added to the pick set matrix X _new In, i.e. X _new ＝[X _(a) X _(b) ]. By minimising a determined matrix W ₂ The method comprises the following steps:which contains the second influence value w of two sets of history features _a (2)、w _b (2). It will be appreciated that the second impact value w determined for this round _a (2) W determined from the previous round _a (2) May be different. And, the residual matrix R determined in round 2 ₂ The method comprises the following steps:

where n=9.

The subsequent iteration is similar to the iteration of round 2 and will not be described in detail here.

It can be understood that the selected set matrix may also be an n×m matrix, where initial values are all 0; when the selected historical feature set is added to the selection set matrix, the elements of the corresponding columns of the selection set matrix are modified into the selected historical feature set, and the rest elements are still 0. For example, in round 1 iteration, if the a-th historical feature set X _(a) For the selected historical feature set, then the selected set matrixHistorical feature set X _(a) Located in a pick set matrix X _new Is shown in column a. And, matrix W corresponding to the second influence value ₂ The initial steps are: n×1 matrix with all elements of 0; after determining a certain second influence value, matrix W ₂ The element of the corresponding row in (c) is modified to the second impact value. For example, the a-th historical feature set X _(a) The second influence value of w _a (2) Matrix W at this time ₂ Is->I.e. the element of row a is modified to the second impact value w _a (2). The residual matrix still satisfies at this time: r is R _g ＝Y-X _new X W; wherein matrix W is equal to matrix W ₂ . In the subsequent iteration process of the 2 nd round, the 3 rd round, etc., the processing manner is similar to that described above, and the description is omitted here.

The embodiment of the invention iterates 3 times to determine three characteristics (slope rising to peak pointPeak point falling slopePeak d ₂ To peak d ₁ Total change in volume between->) The correlation coefficient of the other features is directly 0, no additional calculation is needed, the overall matrix K is as follows, and the intercept value determined based on the residual matrix of the 3 rd round is 78.4364496.

Accordingly, the SOH is calculated as follows:

after that, SOH at the current time can be calculated based on the above equation (2). Specifically, after that, calculation is performed once every 200 turns, and the calculation result SOH is determined _pred . In addition, in order to verify the accuracy of the method provided by the embodiment of the invention, under the state corresponding to all calculation moments, the battery is fully charged and discharged as required to test the real SOH, and the real SOH, namely the SOH, is determined based on the working condition data of the fully charged and discharged at the moment _test . The results obtained are shown in Table 1 and FIG. 3.

TABLE 1

It can be seen that the SOH error of 200 rounds after prediction is 0.08%, and the SOH error of 1000 rounds after prediction is 0.03%, indicating that the method is reliable and has high stability.

The method for determining the battery state of health provided by the embodiment of the invention is described in detail above, and the method can also be realized by a corresponding device, and the device for determining the battery state of health provided by the embodiment of the invention is described in detail below.

Fig. 4 is a schematic structural diagram of an apparatus for determining a state of health of a battery according to an embodiment of the present invention. As shown in fig. 4, the apparatus for determining a state of health of a battery includes:

the acquisition module 41 is configured to acquire historical operating condition data acquired by the battery to be tested in a plurality of different historical periods, and determine a health status of each historical period;

the extracting module 42 is configured to perform feature extraction on each piece of history working condition data, and determine m historical features corresponding to each piece of history working condition data, where m is greater than or equal to 2;

The iteration module 43 is configured to take the same historical feature extracted from all the historical operating mode data as a historical feature set, iteratively select a better one of the historical feature sets according to the health state and the historical feature set, iteratively update and determine the influence degree of the selected historical feature set on the health state, and take the influence degree determined at the end of iteration as a correlation coefficient between the feature and the health state; wherein the iteration round does not exceed m;

the calculating module 44 is configured to obtain current working condition data of the battery to be tested in a current period, extract a plurality of current features related to the association coefficient, and determine a health state of the battery to be tested in the current period according to all the current features and the association coefficient.

In one possible implementation, the iterative module 43 iteratively updates the selected historical feature set to determine the extent of the influence of the selected historical feature set on the health state in a manner that iteratively selects a better one of the historical feature set, including:

in the iteration process of the g round, determining a first influence value of each residual history feature set on the residual matrix determined in the previous round; the remaining historical feature set is a historical feature set which is not selected in the previous iteration process; the initial value of the residual error matrix is a health state matrix containing all the health states;

Taking the remaining historical feature set with the maximum absolute value of the first influence value as a selected historical feature set, and adding the selected historical feature set into a selection set matrix; the selected set matrix comprises the historical feature sets selected in all previous iteration processes;

setting a second influence value for each history feature set in the selected set matrix, and updating the second influence value by minimizing the difference between the health state matrix and the selected set matrix multiplied by the second influence value; the updated second influence value is the influence degree of the corresponding historical feature set of the selection set matrix on the health state;

determining a residual matrix of the g-th round, and meeting the following conditions: r is R _g ＝Y-X _new X W; ending iteration under the condition that the residual matrix converges or the current iteration number g is equal to m;

wherein R is _g Representing the residual matrix of the g th round, Y representing the health state matrix, X _new Representing the pick set matrix, W representing a matrix containing all of the presently determined degrees of influence.

In one possible implementation, the first impact value satisfies:

w _i (1)＝X' _(i) ·R _g-1 ；

wherein X 'is' _(i) Representing a normalized i-th historical feature set, wherein the i-th historical feature set is currently a residual historical feature set; r is R _g-1 Representing the residual matrix determined in round g-1; w (w) _i (1) Representing the i-th set of historical features X _(i) Residual matrix R for round g-1 _g-1 Is a first influence value of (1); is a dot product inner product operation.

In a possible implementation manner, the iteration module 43 iteratively updates and determines the influence degree of the selected historical feature set on the health state in a manner of iteratively selecting the better one of the historical feature sets, and further includes:

determining an invalid residual historical feature set according to a preset first influence value standard, and eliminating the invalid residual historical feature set; wherein the absolute value of the ratio between the first influence value of the ineffective residual history feature set and the first influence value with the largest absolute value is smaller than the first influence value reference; the first impact value reference is between 0 and 0.5.

In one possible implementation, the iterating module 43 eliminates the invalid remaining historical feature set, including:

and rejecting the invalid residual history feature set when the invalid residual history feature set is determined to be invalid in the continuous multi-round iteration process.

In addition to taking the remaining history feature set with the largest absolute value of the first influence value as the selected history feature set, taking the remaining history feature set with the ratio of the absolute value of the first influence value to the first influence value with the largest absolute value larger than the second influence value standard as the selected history feature set; the second influence value reference is between 0.7 and 1.

In one possible implementation, the iteration module 43 updates the second influence value by minimizing a difference between the health state matrix and the selected set matrix multiplied by the second influence value, including:

performing matrix decomposition on the intermediate matrix, and determining an inverse matrix of the intermediate matrix; wherein the intermediate matrix is the product of the transpose of the pick matrix and the pick matrix;

multiplying the inverse matrix of the intermediate matrix by the transpose of the selected set matrix and then multiplying the result of the health state matrix as an updated second influence value.

In one possible implementation, the iteration module 43 uses, as a set of historical features, the same historical feature extracted from all the historical operating condition data, including:

Taking the same historical characteristic extracted from all the historical working condition data as a to-be-determined characteristic set;

determining a similarity between the set of pending features and any other set of pending features;

taking the feature set to be determined as a historical feature set under the condition that the similarity is smaller than a preset threshold value;

and under the condition that the similarity is larger than a preset threshold value, selecting one of the to-be-determined feature set and other to-be-determined feature sets with the similarity larger than the preset threshold value as one historical feature set.

It should be noted that, when the device for determining the battery health status provided in the above embodiment implements the corresponding function, only the division of the above functional modules is used for illustration, and in practical application, the above functional allocation may be implemented by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to implement all or part of the functions described above. In addition, the device for determining the state of health of the battery provided in the above embodiment belongs to the same concept as the method embodiment for determining the state of health of the battery, and the detailed implementation process of the device is referred to in the method embodiment, which is not repeated here.

According to one aspect of the application, the embodiment of the application also provides a computer program product comprising a computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through a communication section. The method for determining the state of health of a battery provided by the embodiment of the application is performed when the computer program is executed by a processor.

In addition, the embodiment of the application also provides a device for determining the health state of a battery, which comprises a processor and a memory, wherein the memory stores a computer program, the processor can execute the computer program stored in the memory, and the method for determining the health state of the battery provided by any embodiment can be realized when the computer program is executed by the processor.

For example, fig. 5 illustrates an apparatus for determining battery health status provided by an embodiment of the present application, the apparatus comprising a bus 1110, a processor 1120, a transceiver 1130, a bus interface 1140, a memory 1150, and a user interface 1160.

In an embodiment of the present application, the apparatus further includes: computer programs stored on the memory 1150 and executable on the processor 1120, which when executed by the processor 1120, implement the various processes of the method embodiments described above for determining battery health.

A transceiver 1130 for receiving and transmitting data under the control of the processor 1120.

In an embodiment of the invention, represented by bus 1110, bus 1110 may include any number of interconnected buses and bridges, with bus 1110 connecting various circuits, including one or more processors, represented by processor 1120, and memory, represented by memory 1150.

Bus 1110 represents one or more of any of several types of bus structures, including a memory bus and a memory controller, a peripheral bus, an accelerated graphics port (Accelerate Graphical Port, AGP), a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such an architecture includes: industry standard architecture (Industry Standard Architecture, ISA) bus, micro channel architecture (Micro Channel Architecture, MCA) bus, enhanced ISA (EISA) bus, video electronics standards association (Video Electronics Standards Association, VESA) bus, peripheral component interconnect (Peripheral Component Interconnect, PCI) bus.

Processor 1120 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments may be implemented by instructions in the form of integrated logic circuits in hardware or software in a processor. The processor includes: general purpose processors, central processing units (Central Processing Unit, CPU), network processors (Network Processor, NP), digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field Programmable Gate Array, FPGA), complex programmable logic devices (Complex Programmable Logic Device, CPLD), programmable logic arrays (Programmable Logic Array, PLA), micro control units (Microcontroller Unit, MCU) or other programmable logic devices, discrete gates, transistor logic devices, discrete hardware components. The methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. For example, the processor may be a single-core processor or a multi-core processor, and the processor may be integrated on a single chip or located on multiple different chips.

The processor 1120 may be a microprocessor or any conventional processor. The steps of the method disclosed in connection with the embodiments of the present invention may be performed directly by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software modules may be located in a random access Memory (Random Access Memory, RAM), flash Memory (Flash Memory), read-Only Memory (ROM), programmable ROM (PROM), erasable Programmable ROM (EPROM), registers, and so forth, as are known in the art. The readable storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.

Bus 1110 may also connect together various other circuits such as peripheral devices, voltage regulators, or power management circuits, bus interface 1140 providing an interface between bus 1110 and transceiver 1130, all of which are well known in the art. Accordingly, the embodiments of the present invention will not be further described.

The transceiver 1130 may be one element or a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. For example: the transceiver 1130 receives external data from other devices, and the transceiver 1130 is configured to transmit the data processed by the processor 1120 to the other devices. Depending on the nature of the computer system, a user interface 1160 may also be provided, for example: touch screen, physical keyboard, display, mouse, speaker, microphone, trackball, joystick, stylus.

It should be appreciated that in embodiments of the present invention, the memory 1150 may further comprise memory located remotely from the processor 1120, such remotely located memory being connectable to a server through a network. One or more portions of the above-described networks may be an ad hoc network (ad hoc network), an intranet, an extranet (extranet), a Virtual Private Network (VPN), a Local Area Network (LAN), a Wireless Local Area Network (WLAN), a Wide Area Network (WAN), a Wireless Wide Area Network (WWAN), a Metropolitan Area Network (MAN), the Internet (Internet), a Public Switched Telephone Network (PSTN), a plain old telephone service network (POTS), a cellular telephone network, a wireless fidelity (Wi-Fi) network, and a combination of two or more of the above-described networks. For example, the cellular telephone network and wireless network may be a global system for mobile communications (GSM) system, a Code Division Multiple Access (CDMA) system, a Worldwide Interoperability for Microwave Access (WiMAX) system, a General Packet Radio Service (GPRS) system, a Wideband Code Division Multiple Access (WCDMA) system, a Long Term Evolution (LTE) system, an LTE Frequency Division Duplex (FDD) system, an LTE Time Division Duplex (TDD) system, a long term evolution-advanced (LTE-a) system, a Universal Mobile Telecommunications (UMTS) system, an enhanced mobile broadband (Enhance Mobile Broadband, embbb) system, a mass machine type communication (massive Machine Type of Communication, mctc) system, an ultra reliable low latency communication (Ultra Reliable Low Latency Communications, uirllc) system, and the like.

It should be appreciated that the memory 1150 in embodiments of the present invention may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. Wherein the nonvolatile memory includes: read-Only Memory (ROM), programmable ROM (PROM), erasable Programmable EPROM (EPROM), electrically Erasable EPROM (EEPROM), or Flash Memory (Flash Memory).

The volatile memory includes: random access memory (Random Access Memory, RAM) which acts as an external cache. By way of example, and not limitation, many forms of RAM are available, such as: static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (ddr SDRAM), enhanced SDRAM (Enhanced SDRAM), synchronous DRAM (SLDRAM), and Direct RAM (DRAM). Memory 1150 described in embodiments of the present invention includes, but is not limited to, the above and any other suitable types of memory.

In an embodiment of the invention, memory 1150 stores the following elements of operating system 1151 and application programs 1152: an executable module, a data structure, or a subset thereof, or an extended set thereof.

Specifically, the operating system 1151 includes various system programs, such as: a framework layer, a core library layer, a driving layer and the like, which are used for realizing various basic services and processing tasks based on hardware. The applications 1152 include various applications such as: a Media Player (Media Player), a Browser (Browser) for implementing various application services. A program for implementing the method of the embodiment of the present invention may be included in the application 1152. The application 1152 includes: applets, objects, components, logic, data structures, and other computer system executable instructions that perform particular tasks or implement particular abstract data types.

In addition, the embodiment of the present invention further provides a computer readable storage medium, on which a computer program is stored, where the computer program when executed by a processor implements each process of the above embodiment of the method for determining a battery health status, and the same technical effects can be achieved, and for avoiding repetition, a detailed description is omitted herein.

The computer-readable storage medium includes: persistent and non-persistent, removable and non-removable media are tangible devices that may retain and store instructions for use by an instruction execution device. The computer-readable storage medium includes: electronic storage, magnetic storage, optical storage, electromagnetic storage, semiconductor storage, and any suitable combination of the foregoing. The computer-readable storage medium includes: phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), non-volatile random access memory (NVRAM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassette storage, magnetic tape disk storage or other magnetic storage devices, memory sticks, mechanical coding (e.g., punch cards or bump structures in grooves with instructions recorded thereon), or any other non-transmission medium that may be used to store information that may be accessed by a computing device. In accordance with the definition in the present embodiments, the computer-readable storage medium does not include a transitory signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a pulse of light passing through a fiber optic cable), or an electrical signal transmitted through a wire.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus, devices and methods may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. In addition, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices, or elements, or may be an electrical, mechanical, or other form of connection.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one position, or may be distributed over a plurality of network units. Some or all of the units can be selected according to actual needs to solve the problem to be solved by the scheme of the embodiment of the application.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the embodiments of the present invention is essentially or partly contributing to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (including: a personal computer, a server, a data center or other network device) to perform all or part of the steps of the method according to the embodiments of the present invention. And the storage medium includes various media as exemplified above that can store program codes.

In the description of the embodiments of the present invention, those skilled in the art should appreciate that the embodiments of the present invention may be implemented as a method, an apparatus, a device, and a storage medium. Thus, embodiments of the present invention may be embodied in the following forms: complete hardware, complete software (including firmware, resident software, micro-code, etc.), a combination of hardware and software. Furthermore, in some embodiments, embodiments of the invention may also be implemented in the form of a computer program product in one or more computer-readable storage media having computer program code embodied therein.

Any combination of one or more computer-readable storage media may be employed by the computer-readable storage media described above. The computer-readable storage medium includes: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of the computer readable storage medium include the following: portable computer diskette, hard disk, random Access Memory (RAM), read-only Memory (ROM), erasable programmable read-only Memory (EPROM), flash Memory (Flash Memory), optical fiber, compact disc read-only Memory (CD-ROM), optical storage device, magnetic storage device, or any combination thereof. In embodiments of the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, device.

The computer program code embodied in the computer readable storage medium may be transmitted using any appropriate medium, including: wireless, wire, fiber optic cable, radio Frequency (RF), or any suitable combination thereof.

Computer program code for carrying out operations of embodiments of the present invention may be written in assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or in one or more programming languages, including an object oriented programming language such as: java, smalltalk, C ++, also include conventional procedural programming languages, such as: c language or similar programming language. The computer program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of remote computers, the remote computers may be connected via any sort of network, including: a Local Area Network (LAN) or a Wide Area Network (WAN), which may be connected to the user's computer or to an external computer.

The embodiments of the present invention describe the provided methods, apparatuses, devices through flowcharts and/or block diagrams.

It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions. These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.

These computer readable program instructions may also be stored in a computer readable storage medium that can cause a computer or other programmable data processing apparatus to function in a particular manner. Thus, instructions stored in a computer-readable storage medium produce an instruction means which implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.

The foregoing is merely a specific implementation of the embodiment of the present invention, but the protection scope of the embodiment of the present invention is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the embodiment of the present invention, and the changes or substitutions are covered by the protection scope of the embodiment of the present invention. Therefore, the protection scope of the embodiments of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method of determining a state of health of a battery, comprising:

2. The method of claim 1, wherein iteratively updating the determination of the degree of influence of the selected set of historical features on the health state in a manner that iteratively selects a better one of the set of historical features comprises:

3. The method of claim 2, wherein the first impact value satisfies:

wherein X 'is' _(i) Representing a normalized i-th set of history features, and the i-th set of history features is currently the remaining history featuresA collection; r is R _g-1 Representing the residual matrix determined in round g-1; w (w) _i (1) Representing the i-th set of historical features X _(i) Residual matrix R for round g-1 _g-1 Is a first influence value of (1); is a dot product inner product operation.

4. The method of claim 2, wherein iteratively updating determines the degree of influence of the selected historical feature set on the health state in a manner that iteratively selects a better one of the historical feature sets, further comprising:

5. The method of claim 4, wherein said culling said invalid remaining set of historical features comprises:

6. The method of claim 2, wherein iteratively updating determines the degree of influence of the selected historical feature set on the health state in a manner that iteratively selects a better one of the historical feature sets, further comprising:

7. The method of claim 2, wherein updating the second influence value by minimizing a difference between the health state matrix and a pick set matrix multiplied by the second influence value comprises:

8. The method of claim 1, wherein said taking as a set of historical features the same historical feature extracted from all of said historical operating condition data comprises:

9. An apparatus for determining the state of health of a battery, comprising:

10. An apparatus for determining battery state of health comprising a processor and a memory, the memory storing a computer program, wherein the processor executes the computer program stored in the memory to implement the method of determining battery state of health of any one of claims 1 to 8.

11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of determining the state of health of a battery according to any of claims 1 to 8.