A method and device for training a battery fault identification model

Abstract

The present application relates to the technical field of battery fault identification, and discloses a method and device for training a battery fault identification model. The method includes respectively performing serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested, obtaining first time series test data and second time series test data; performing dynamic time warping and difference calculation of average differential voltage on each second time series test data and the first time series test data one by one, and finally inputting each dynamic time warping value and each average differential voltage difference as sample feature data into a preset battery fault identification model, and training the battery fault identification model according to each sample feature data. The present application enables the trained battery fault identification model to more accurately identify faults of the target battery to be tested.

Description

Technical Field

[0001] The present application relates to the technical field of battery fault identification, and particularly relates to a method and device for training a battery fault identification model. Background Art

[0002] In the prior art, there are many technical problems in the safety management of battery systems, especially in multi-fault diagnosis. As the core component of electric vehicles, the performance of batteries directly affects the safety and reliability of the whole vehicle. However, common faults in battery systems are characterized by concealment, diversity, and complexity. These faults are often difficult to identify individually. Especially during actual operation, minor faults and multi-faults may occur simultaneously, and existing methods cannot effectively and accurately identify these faults. Therefore, there is a problem of inaccurate fault identification in the prior art. Summary of the Invention

[0003] The present application provides a method and device for training a battery fault identification model to solve one or more technical problems existing in the prior art, and at least provides a beneficial option or creative condition.

[0004] Other features and advantages of the present application will become apparent through the following detailed description, or will be partially learned through the practice of the present application.

[0005] According to one aspect of the embodiments of the present application, a method for training a battery fault identification model is provided. The method includes:

[0006] Performing serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested respectively, to obtain first time series test data of the reference battery and second time series test data of each target battery to be tested;

[0007] Performing dynamic time warping on each of the second time series test data and the first time series test data one by one, to obtain dynamic time warping values of each target battery to be tested and the reference battery;

[0008] Calculating the difference of the average differential voltage between each of the second time series test data and the first time series test data one by one, to obtain the average differential voltage difference between each target battery to be tested and the reference battery;

[0009] Taking each of the dynamic time warping values and each of the average differential voltage differences as sample feature data of each target battery to be tested and inputting them into a preset battery fault identification model, and training the battery fault identification model according to each of the sample feature data;

[0010] The average differential voltage difference between a single target battery to be measured and the reference battery is obtained through the following steps:

[0011] Determine the first open-circuit voltage sequence test data of the reference battery according to the first time-series test data, and determine the second open-circuit voltage sequence test data of the target battery to be measured according to the second time-series test data;

[0012] Determine the first average differential voltage value of the reference battery according to each open-circuit voltage value and the number of each open-circuit voltage value in the first open-circuit voltage sequence test data;

[0013] Determine the second average differential voltage value of the target battery to be measured according to each open-circuit voltage value and the number of each open-circuit voltage value in the second open-circuit voltage sequence test data;

[0014] Determine the average differential voltage difference based on the first average differential voltage value and the second average differential voltage value.

[0015] In an embodiment of the present application, based on the foregoing solution, the dynamic time warping value between a single target battery to be measured and the reference battery is obtained through the following steps:

[0016] Construct a distance matrix according to the first time-series test data of the reference battery and the second time-series test data of the target battery to be measured;

[0017] Calculate the optimal warping distance between any point of the first time-series test data and any point of the second time-series test data in the distance matrix according to the dynamic time warping method, and generate a set of optimal warping distances;

[0018] Search for the target optimal warping distance with the smallest warping distance in the set of optimal warping distances as the dynamic time warping value.

[0019] In an embodiment of the present application, based on the foregoing solution, the battery fault identification model includes a first-layer Gaussian mixture model and a second-layer Gaussian mixture model. Training the battery fault identification model according to each sample feature data includes:

[0020] Obtain the dynamic time warping value and the average differential voltage difference between each target battery to be measured and the reference battery according to each sample feature data;

[0021] Input each dynamic time warping value and each average differential voltage difference into the first-layer Gaussian mixture model to obtain the first fault analysis result of each target battery to be measured;

[0022] Input each of the dynamic time warping values and each of the average differential voltage differences into the second Gaussian mixture model to obtain a second fault analysis result for each of the target batteries to be tested;

[0023] Train the battery fault identification model according to the first fault analysis result and the second fault analysis result.

[0024] In an embodiment of the present application, based on the foregoing solution, the first Gaussian mixture model includes a first probability density function, which is used to determine the first probability distribution points of each of the target batteries to be tested, and based on each of the first probability distribution points, determine whether a short - circuit fault occurs in the target battery to be tested; the second Gaussian mixture model includes a second probability density function, which is used to determine the second probability distribution points of each of the target batteries to be tested, and based on each of the second probability distribution points, determine whether an abnormal degradation fault occurs in the target battery to be tested.

[0025] In an embodiment of the present application, based on the foregoing solution, the abscissa and ordinate of the first probability distribution point and the second probability distribution point are respectively the dynamic time warping value and the average differential voltage difference.

[0026] In an embodiment of the present application, based on the foregoing solution, whether a short - circuit fault occurs in the target battery to be tested can be determined through the following steps:

[0027] Obtain the abscissa value and ordinate value of the first probability distribution point of the target battery to be tested. The abscissa value is the dynamic time warping value between the target battery to be tested and the reference battery, and the ordinate value is the average differential voltage difference between the target battery to be tested and the reference battery;

[0028] If the abscissa value is greater than a first preset threshold and / or the ordinate value is greater than a second preset threshold, it is determined that the target battery to be tested has a short - circuit fault.

[0029] In an embodiment of the present application, based on the foregoing solution, whether an abnormal degradation fault occurs in the target battery to be tested can be determined through the following steps:

[0030] Obtain the abscissa value and ordinate value of the second probability distribution point of the target battery to be tested. The abscissa value is the dynamic time warping value between the target battery to be tested and the reference battery, and the ordinate value is the average differential voltage difference between the target battery to be tested and the reference battery;

[0031] If the abscissa value is greater than a third preset threshold or the ordinate value is greater than a fourth preset threshold, it is determined that the target battery to be tested has an abnormal degradation fault.

[0032] In one embodiment of the present application, based on the foregoing solution, the serialization test related to battery characteristics is to test the voltage characteristics of the battery in sequence according to the time sequence in a preset time period.

[0033] According to one aspect of the embodiments of the present application, a device for training a battery fault recognition model is provided. The device includes:

[0034] A test unit configured to perform serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested respectively, so as to obtain first time series test data of the reference battery and second time series test data of each target battery to be tested;

[0035] A first calculation unit configured to perform dynamic time warping on each of the second time series test data and the first time series test data one by one, so as to obtain dynamic time warping values of each target battery to be tested and the reference battery;

[0036] A second calculation unit configured to calculate the difference of the average differential voltage between each of the second time series test data and the first time series test data one by one, so as to obtain the average differential voltage difference between each target battery to be tested and the reference battery;

[0037] A training unit configured to input each of the dynamic time warping values and each of the average differential voltage differences as sample feature data of each target battery to be tested into a preset battery fault recognition model, and train the battery fault recognition model according to each of the sample feature data;

[0038] The average differential voltage difference between a single target battery to be tested and the reference battery is obtained through the following steps:

[0039] Determine first open-circuit voltage sequence test data of the reference battery according to the first time series test data, and determine second open-circuit voltage sequence test data of the target battery to be tested according to the second time series test data;

[0040] Determine a first average differential voltage value of the reference battery according to each open-circuit voltage value and the number of each open-circuit voltage value in the first open-circuit voltage sequence test data;

[0041] Determine a second average differential voltage value of the target battery to be tested according to each open-circuit voltage value and the number of each open-circuit voltage value in the second open-circuit voltage sequence test data;

[0042] Determine the average differential voltage difference based on the first average differential voltage value and the second average differential voltage value.

[0043] The beneficial effects of the present application are as follows:

[0044] By separately performing serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested, the first time-series test data of the reference battery and the second time-series test data of each of the target batteries to be tested can be used to characterize the battery characteristics of the reference battery and the target batteries to be tested over time.

[0045] Furthermore, the dynamic time warping values between each of the target batteries to be tested and the reference battery can be used to reflect the differences between the target batteries to be tested and the reference battery (normal battery), and the average differential voltage differences between each of the target batteries to be tested and the reference battery can reflect the inconsistencies between the target batteries to be tested and the reference battery (normal battery). The differences and inconsistencies obtained through analysis can be used as sample feature data of the target batteries to be tested and input into the battery fault identification model to train the model, enabling the trained battery fault identification model to accurately identify whether there are excessive differences or inconsistencies between the target batteries to be tested and the normal reference battery, and then analyze and identify whether the target batteries to be tested have failed, improving the accuracy of fault identification of the model. Description of the Drawings

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly describe the drawings required for the description of the embodiments. Obviously, the described drawings are only a part of the embodiments of the present application, rather than all the embodiments. Those skilled in the art can also obtain other design solutions and drawings based on these drawings without creative efforts.

[0047] Figure 1 FIG. is a flowchart of a method for training a battery fault identification model according to an embodiment of the present application;

[0048] Figure 2 FIG. is a circuit structure diagram of an equivalent circuit model according to an embodiment of the present application;

[0049] Figure 3 FIG. is a block diagram of a device for training a battery fault identification model according to an embodiment of the present application;

[0050] Figure 4 FIG. is a probability density distribution diagram for fault identification using a double Gaussian mixture model according to an embodiment of the present application. Detailed Embodiments

[0051] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art.

[0052] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of this application. However, those skilled in the art will realize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be used. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of this application.

[0053] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontrol node devices.

[0054] The flowcharts shown in the drawings are merely illustrative and do not necessarily include all the content and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0055] It should be noted that: "a plurality" as mentioned in this document refers to two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0056] The implementation details of the technical solutions of the embodiments of this application are elaborated in detail below:

[0057] According to one aspect of the embodiments of this application, a method for training a battery fault identification model is provided. Figure 1 The flowchart of the battery thermal runaway state warning method shown according to the embodiments of this application, this method at least includes steps 110 to 140, which are introduced in detail as follows:

[0058] In step 110, serialization tests related to battery characteristics are respectively performed on a preset reference battery and each target battery to be tested, and first time-series test data of the reference battery and second time-series test data of each target battery to be tested are obtained.

[0059] Specifically, the input data of the target battery to be tested or the reference battery cannot directly form sequence test data to provide a data basis for subsequent fault analysis. It is necessary to transform the input data input to the target battery to be tested or the reference battery through an equivalent circuit model to form time-series test data. Therefore, serialization tests related to battery characteristics are respectively performed on the preset reference battery and each target battery to be tested, that is, sequence data with spatio-temporal characteristics is formed according to the change of the internal battery parameters of the target battery to be tested or the reference battery in chronological order through the equivalent circuit model, that is, the first time-series test data of the reference battery and the second time-series test data of each target battery to be tested.

[0060] In an embodiment of the present application, the serialization test related to battery characteristics is to test the voltage characteristics of the battery in chronological order in a preset time period.

[0061] Specifically, the preset time period can be a time period selected according to actual needs, such as a certain time period during the charging process of the battery, or a certain time period during the discharging process of the battery. Data analysis of the characteristics of the battery, especially the voltage characteristics, can be performed in the selected preset time period, and the data related to the voltage characteristics actually collected can be converted into the time-series test data in the embodiment of the present application through the equivalent circuit model.

[0062] In an embodiment of the present application, the equivalent circuit model can be specifically as Figure 2 shown. The equivalent circuit model consists of an ideal voltage source, a series resistor, and two resistor-capacitor (RC) circuits. R0 is the ohmic resistance, R1 and C1 respectively represent the activation polarization resistance and capacitance, and R2 and C2 respectively represent the concentration polarization resistance and capacitance. According to Kirchhoff's law, the state equations of the second-order RC model can be described by equations (1) and (2):

[0063] (1)

[0064] (2)

[0065] In the formula, U t and U OCV respectively represent the terminal voltage and open-circuit voltage of the battery; I refers to the current; U 1 and U 2 are the polarization voltages of the two RC circuits.

[0066] In addition, when this method is actually applied to time-domain sequences, discretization processing is required and is described in (3) and (4). Next, the present invention applies RLS to online parameter identification. The specific process is as follows:

[0067] (3)

[0068] (4)

[0069] (5)

[0070] where k is the sampling time, K k is the gain matrix; P k-1 represents the error covariance matrix; is the data matrix; is the parameter matrix; E is the identity matrix; y k is the system observation; μ is the forgetting factor, which is taken as 1 here.

[0071] The data matrix is expanded as:

[0072] (6)

[0073] Therefore, the above formulas (1)-(6) can be used to form time series test data according to the actual parameter changes inside the equivalent simulated battery.

[0074] In step 120, each of the second time series test data is dynamically time warped with the first time series test data to obtain the dynamic time warping values of each of the target batteries to be tested and the reference battery.

[0075] Specifically, a battery pack is usually composed of multiple single cells connected in series or in parallel. When the battery is operating normally, each battery exhibits similar characteristics. Therefore, the differences between series-connected batteries are regarded as fault characteristics for diagnosing battery faults. The Dynamic Time Warping (DTW) algorithm is a method for measuring the similarity of two time series. In this application, DTW is used to quantify the degree of fault, that is, the fault difference between the target battery to be tested and the normal reference battery, which is calculated through the first time series test data and the second time series test data. In addition, the smaller the DTW value, the higher the similarity between the curves.

[0076] DTW is used to accurately calculate the difference value between each independent series battery. Specifically, the error is accurately calculated by comparing and analyzing the time series of the batteries. The operating data of the batteries (P and Q), where P and Q represent the first time series test data and the second time series test data respectively, and the difference is calculated based on the Euclidean distance of the time series (such as the difference between voltage points). The rule of DTW is to dynamically adjust the time series and find the optimal matching path to minimize their differences

[0077] In an embodiment of the present application, the dynamic time warping value of a single target battery to be tested and the reference battery is obtained through the following steps:

[0078] Construct a distance matrix based on the first time series test data of the reference battery and the second time series test data of the target battery to be tested;

[0079] Calculate the best warping distance between any point of the first time series test data and any point of the second time series test data in the distance matrix according to the dynamic time warping method, and generate a set of best warping distances;

[0080] Find the target best warping distance with the smallest warping distance in the set of best warping distances as the dynamic time warping value.

[0081] Specifically, the calculation process of the dynamic time warping value is as follows:

[0082] Construct an n×m matrix, that is, the distance matrix described in the embodiment of the present application, defined as the distance matrix D base :

[0083] ;

[0084] In the formula, dbase represents the basic distance, and the specific meaning is the Euclidean distance between point p i in time series P and point q j in time series Q. Time series P and time series Q represent the first time series test data and the second time series test data respectively, and the database is:

[0085] ;

[0086] Therefore, the optimal warping path can be defined as:

[0087] ;

[0088] In the formula, F is the length of the optimal warping path; W F =(i, j) represents the corresponding relationship between p i and q j ​

[0089] Specifically, the role of optimizing path bending is to dynamically adjust the alignment of time series on the time axis to find the optimal matching path between two time series. This can help simplify the Euclidean distance between two time series at a specific point in the case of time asynchrony, and the dbase bending path accumulates these Euclidean distances, thus minimizing the overall difference.

[0090] The DTW distance between time series P and Q is calculated as follows:

[0091] ;

[0092] The specific recursive definition is as follows:

[0093] ;

[0094] In the formula, DTW(i, j) represents the optimal bending distance between the first i points of time series P and the first j points of time series Q; i ∈ (1, n], j ∈ (1, m), DTW(1, 1) = dbase(p1, q1), and DTW(n, m) is the best bending distance between the two DTW distances between the two time series P and Q.

[0095] Specifically, the DTW(i, j) set, that is, the best bending distance set, is formed by calculating the best bending distance between any point of the first time series test data and any point of the second time series test data, and by looking up DTW(n, m), that is, looking up the target best bending distance with the smallest bending distance as the dynamic time warping value.

[0096] In step 130, the difference in the average differential voltage between each of the second time series test data and the first time series test data is calculated to obtain the difference in the average differential voltage between each of the target batteries to be tested and the reference battery.

[0097] Specifically, the average differential voltage, denoted as MDV, is used to determine the voltage difference between series-connected batteries. The determination principle is usually to determine the inconsistency by comparing the open-circuit voltage difference between each target battery to be tested and the voltage of the reference battery. Inconsistency refers to the performance difference shown by the battery in the working state, which may affect the safety and reliability of the entire battery pack. Inconsistency emphasizes the potential failure risk that the target battery to be tested may cause, so the purpose is to ensure the overall performance and safety of the battery pack.

[0098] In an embodiment of the present application, the difference in the average differential voltage between a single target battery to be tested and the reference battery is obtained through the following steps:

[0099] Determine the first open-circuit voltage sequence test data of the reference battery according to the first time-series test data, and determine the second open-circuit voltage sequence test data of the target battery under test according to the second time-series test data;

[0100] Determine the first average differential voltage value of the reference battery according to each open-circuit voltage value and the number of each open-circuit voltage value in the first open-circuit voltage sequence test data;

[0101] Determine the second average differential voltage value of the target battery under test according to each open-circuit voltage value and the number of each open-circuit voltage value in the second open-circuit voltage sequence test data;

[0102] Determine the average differential voltage difference based on the first average differential voltage value and the second average differential voltage value.

[0103] The embodiments of the present application use the average differential voltage (MDV) to represent the inconsistency between battery strings. Taking an instant t within a certain time period 0~ t e as an example to illustrate the detection method. The open-circuit voltage OCV of each target battery under test in the target battery pack under test is expressed as:

[0104] ;

[0105] In the formula, n is the number of target batteries under test in the battery system; i represents the i-th target battery unit.

[0106] ;

[0107] In the formula, DVOCV,i is the differential OCV of the i-th battery, and MDVOCV,i is the average value of DVOCV,i.

[0108] Specifically, by comparing the average differential voltage differences between the reference battery and the target battery under test, the inconsistency between the reference battery and the target battery under test can be analyzed.

[0109] In step 140, input each of the dynamic time warping values and each of the average differential voltage differences as sample feature data of each target battery under test into a preset battery fault identification model, and train the battery fault identification model according to each of the sample feature data.

[0110] Specifically, in the present application, each of the dynamic time warping values and each of the average differential voltage differences are input into a preset battery fault identification model as sample feature data of each of the target batteries to be tested. According to the input sample feature data, it can be used as a training sample for the battery fault identification model, so that the battery fault identification model can quickly identify whether the target battery to be tested has a fault and what faults have occurred during subsequent identification, and can accurately identify multiple faults.

[0111] In an embodiment of the present application, the battery fault identification model includes a first-layer Gaussian mixture model and a second-layer Gaussian mixture model. Training the battery fault identification model according to each of the sample feature data includes:

[0112] Obtaining the dynamic time warping values and average differential voltage differences between each of the target batteries to be tested and the reference battery according to each of the sample feature data;

[0113] Inputting each of the dynamic time warping values and each of the average differential voltage differences into the first-layer Gaussian mixture model to obtain a first fault analysis result for each of the target batteries to be tested;

[0114] Inputting each of the dynamic time warping values and each of the average differential voltage differences into the second-layer Gaussian mixture model to obtain a second fault analysis result for each of the target batteries to be tested;

[0115] Training the battery fault identification model according to the first fault analysis result and the second fault analysis result.

[0116] In an embodiment of the present application, the first-layer Gaussian mixture model includes a first probability density function, which is used to determine the first probability distribution points of each of the target batteries to be tested, and based on each of the first probability distribution points, determine whether the target battery to be tested has a short-circuit fault; the second-layer Gaussian mixture model includes a second probability density function, which is used to determine the second probability distribution points of each of the target batteries to be tested, and based on each of the second probability distribution points, determine whether the target battery to be tested has an abnormal degradation fault. The abscissa and ordinate of the first probability distribution point and the second probability distribution point are respectively the dynamic time warping value and the average differential voltage difference.

[0117] Specifically, as Figure 4 shown, Figure 4(a)The respective first probability distribution points obtained from the first probability density function used in the first-layer Gaussian mixture model, where the first probability distribution points represented by the blue dots are the normal target batteries under test without faults, and the first probability distribution points represented by the yellow dots are the target batteries under test with SC faults (short circuit faults).

[0118] Figure 4 (b)The respective second probability distribution points obtained from the second probability density function used in the second-layer Gaussian mixture model, where the second probability distribution points represented by the blue triangles are the normal target batteries under test without faults, and the second probability distribution points represented by the yellow triangles are the target batteries under test with abnormal degradation faults.

[0119] It should be noted that the embodiments of the present application can also identify SC faults and abnormal degradation faults through a two-layer clustering method. First, the normal batteries and the batteries with SC faults are distinguished through the first-layer Gaussian mixture model, and then the normal batteries identified by the first-layer Gaussian mixture model are further identified for abnormal degradation faults. Figure 4 It can be seen that when the DTW value (Dynamic Time Warping value) is high, SC faults and abnormal degradation faults are likely to occur. When the MDV value (Mean Differential Voltage value) is high, abnormal degradation faults are likely to occur, but SC faults do not necessarily occur.

[0120] Further, assuming that the Gaussian mixture model consists of G Gaussian models (in the embodiments of the present application, the value of G is 2), the probability density function of the Gaussian mixture model is as follows:

[0121] ;

[0122] where p(g)=πg is the weight of the g-th Gaussian model; p(x|g)= N(x|ug, ∑g|) is the probability density function of the g-th Gaussian model. The first-layer probability density function can be obtained through the above formula, and the formula of the second-layer probability density function is specifically expanded as:

[0123] ;

[0124] where D represents the number of variables in each Gaussian model.

[0125] Whether the target battery under test has a short circuit fault can be determined through the following steps:

[0126] Obtain the abscissa value and ordinate value of the first probability distribution point of the target battery under test. The abscissa value is the dynamic time warping value between the target battery under test and the reference battery, and the ordinate value is the mean differential voltage difference between the target battery under test and the reference battery.

[0127] If the abscissa value is greater than the first preset threshold and / or the ordinate value is greater than the second preset threshold, it is determined that a short - circuit fault occurs in the target battery to be tested.

[0128] Specifically, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold can be set according to actual needs and are not limited herein. In the embodiments of the present application, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold can be specifically 50, 0.05, 8, and 0.02 respectively. The units of the second preset threshold and the fourth preset threshold are volts.

[0129] In an embodiment of the present application, whether an abnormal degradation fault occurs in the target battery to be tested can be determined through the following steps:

[0130] Obtain the abscissa value and the ordinate value of the second probability distribution point of the target battery to be tested. The abscissa value is the dynamic time warping value between the target battery to be tested and the reference battery, and the ordinate value is the average differential voltage difference between the target battery to be tested and the reference battery.

[0131] If the abscissa value is greater than the third preset threshold or the ordinate value is greater than the fourth preset threshold, it is determined that an abnormal degradation fault occurs in the target battery to be tested.

[0132] Specifically, if the dynamic time warping value between the target battery to be tested and the reference battery is greater than 8 or the average differential voltage value of the target battery to be tested exceeds 0.02, it indicates that an abnormal degradation fault occurs in the target battery to be tested. Therefore, the target battery to be tested can be labeled with an abnormal degradation fault, and the dynamic time warping value and the average differential voltage difference of the target battery to be tested can be used as training samples and input into the battery fault recognition model. In this way, when the battery fault recognition model identifies a battery with battery parameters within this range next time, it can directly output the analysis result that the current battery has an abnormal degradation fault.

[0133] Based on the selected fault features, the present application uses a two - layer Gaussian mixture model clustering algorithm for multi - fault diagnosis of the battery system. The clustering algorithm is an unsupervised learning method, which means that the data applied is usually unlabeled. It should be emphasized that although unlabeled data is used for clustering, due to previous experiments, the label of each sample is known, that is, the label is given through the sample feature data given in the embodiments of the present application. Specifically, the clustering of the first - layer Gaussian mixture model is used to separate the short - circuit (SC) fault from other faults. The result is as Figure 4 shown in (a). Based on the first - layer clustering result, the samples considered to be SC faults are removed, and the remaining samples are subjected to a second clustering. Just as Figure 4(b) As shown, abnormally degraded batteries can be accurately identified.

[0134] In this application, a cycling experiment was designed by connecting normal batteries and degraded batteries. Meanwhile, a short - circuit (SC) fault was artificially triggered through an equivalent circuit model during this process. This technology is developed based on an equivalent circuit model (ECM), similarity metric, and unsupervised learning framework. First, combined with the recursive least - squares method, an equivalent circuit model was constructed to model the battery and identify the open - circuit voltage (OCV). Then, dynamic time warping (DTW) was used to measure the similarity between OCVs. In addition, the mean differential voltage (MDV) was defined to evaluate the inconsistency of the battery system. Both DTW and MDV were extracted as fault features for subsequent diagnosis. Finally, a double Gaussian mixture model was applied to cluster multiple faults. The first step is used to detect short - circuit faults, and the second step is used to distinguish abnormally degraded batteries from normal batteries. During the whole process, short - circuit faults and abnormally degraded faults can be quantified, classified, and located in a timely and effective manner.

[0135] In summary, in this application, by separately performing serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested, the first - time - series test data of the reference battery and the second - time - series test data of each target battery to be tested can be used to characterize the battery characteristics of the reference battery and the target batteries to be tested over time.

[0136] Furthermore, the dynamic time warping values of each target battery to be tested and the reference battery can be used to reflect the difference between the target battery to be tested and the reference battery (normal battery), and the difference in the mean differential voltage between each target battery to be tested and the reference battery can reflect the inconsistency between the target battery to be tested and the reference battery (normal battery). The difference and inconsistency obtained through analysis can be used as sample feature data of the target battery to be tested and input into the battery fault recognition model to train the model, so that the trained battery fault recognition model can accurately identify whether there is too much difference or inconsistency between the target battery to be tested and the normal reference battery, and further analyze and identify whether the target battery to be tested has a fault, improving the accuracy of the model's fault recognition.

[0137] Figure 3 FIG. 300 is a block diagram of a battery thermal runaway state warning device 300 according to an embodiment of the present application. According to an embodiment of the present application, the battery thermal runaway state warning device 300 includes: a test unit 301, a first calculation unit 302, a second calculation unit 303, and a training unit 304.

[0138] A test unit 301 is configured to perform serialization tests related to battery characteristics on a preset reference battery and each target battery to be tested respectively, so as to obtain first time-series test data of the reference battery and second time-series test data of each of the target batteries to be tested;

[0139] A first calculation unit 302 is configured to perform dynamic time warping on each of the second time-series test data and the first time-series test data one by one, so as to obtain dynamic time warping values of each of the target batteries to be tested and the reference battery;

[0140] A second calculation unit 303 is configured to calculate the difference in average differential voltage between each of the second time-series test data and the first time-series test data one by one, so as to obtain the average differential voltage difference between each of the target batteries to be tested and the reference battery;

[0141] A training unit 304 is configured to input each of the dynamic time warping values and each of the average differential voltage differences as sample feature data of each of the target batteries to be tested into a preset battery fault identification model, and train the battery fault identification model according to each of the sample feature data.

[0142] As another aspect, the present application also provides a computer-readable storage medium, on which a program product capable of implementing the method provided in the above description of this specification is stored. In some possible implementation manners, various aspects of the present application can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary implementation manners of the present application described in the "Embodiment Method" part of this specification.

[0143] The program product for implementing the above method according to the embodiment of the present application may adopt a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present application is not limited thereto. In this document, the readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0144] The program product may employ any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the readable storage medium (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0145] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which the readable program code is carried. Such a propagated data signal may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0146] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0147] The program code for performing the operations of this application may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0148] In addition, the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the time sequence of these processes. Additionally, it is also easy to understand that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0149] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A battery fault identification model training method, characterized in that: The method comprises: Performing serial tests related to battery characteristics on a preset reference battery and each target battery to be tested, respectively, to obtain first time series test data of the reference battery and second time series test data of each target battery to be tested; Performing dynamic time warping on each of the second time series test data and the first time series test data one by one to obtain dynamic time warping values ​​of each of the target battery to be tested and the reference battery; Calculate the difference of average differential voltage between each of the second time series test data and the first time series test data one by one, to obtain the average differential voltage difference between each of the target test batteries and the reference battery; Inputting each of the dynamic time warping values ​​and each of the average differential voltage difference values ​​as sample feature data of each of the target batteries to be tested into a preset battery fault identification model, and training the battery fault identification model according to each of the sample feature data; The average differential voltage difference between the single target battery to be tested and the reference battery is obtained by the following steps: Determine first open circuit voltage sequence test data of the reference battery according to the first time series test data, and determine second open circuit voltage sequence test data of the target battery to be tested according to the second time series test data; Determining a first average differential voltage value of the reference battery according to each open circuit voltage value and the number of each open circuit voltage value in the first open circuit voltage sequence test data; Determining a second average differential voltage value of the target battery to be tested according to each open circuit voltage value and the number of each open circuit voltage value in the second open circuit voltage sequence test data; The average differential voltage difference value is determined based on the first average differential voltage value and the second average differential voltage value.

2. The battery fault identification model training method according to claim 1, characterized in that: The dynamic time warping values ​​of the single target battery to be tested and the reference battery are obtained by the following steps: Constructing a distance matrix according to the first time series test data of the reference battery and the second time series test data of the target battery to be tested; Calculate the best bending distance between any point of the first time series test data and any point of the second time series test data in the distance matrix according to the dynamic time warping method, and generate a best bending distance set; A target optimal bending distance with the smallest bending distance is searched in the optimal bending distance set as the dynamic time warping value.

3. The battery fault identification model training method according to claim 1, characterized in that: The battery fault identification model includes a first-layer Gaussian mixture model and a second-layer Gaussian mixture model, and the training of the battery fault identification model according to each of the sample feature data includes: Acquire a dynamic time warping value and an average differential voltage difference between each of the target batteries to be tested and the reference battery according to each of the sample characteristic data; Inputting each of the dynamic time warping values ​​and each of the average differential voltage difference values ​​into the first-layer Gaussian mixture model to obtain a first fault analysis result of each of the target batteries to be tested; Inputting each of the dynamic time warping values ​​and each of the average differential voltage difference values ​​into the second-layer Gaussian mixture model to obtain a second fault analysis result of each of the target batteries to be tested; The battery fault identification model is trained according to the first fault analysis result and the second fault analysis result.

4. The battery fault identification model training method according to claim 3 is characterized in that: The first-layer Gaussian mixture model includes a first probability density function, which is used to determine a first probability distribution point of each of the target batteries to be tested, and to determine whether the target batteries to be tested have a short circuit fault based on each of the first probability distribution points; the second-layer Gaussian mixture model includes a second probability density function, which is used to determine a second probability distribution point of each of the target batteries to be tested, and to determine whether the target batteries to be tested have an abnormal degradation fault based on each of the second probability distribution points.

5. The battery fault identification model training method according to claim 4, characterized in that: The horizontal coordinates and vertical coordinates of the first probability distribution point and the second probability distribution point are respectively the dynamic time warping value and the average differential voltage difference.

6. The battery fault identification model training method according to claim 5, characterized in that: Whether the target battery to be tested has a short circuit fault can be determined by the following steps: Acquire a horizontal coordinate value and a vertical coordinate value of a first probability distribution point of the target battery to be tested, wherein the horizontal coordinate value is a dynamic time warping value of the target battery to be tested and the reference battery, and the vertical coordinate value is an average differential voltage difference between the target battery to be tested and the reference battery; If the horizontal coordinate value is greater than a first preset threshold value and / or the vertical coordinate value is greater than a second preset threshold value, it is determined that a short circuit fault occurs in the target battery to be tested.

7. The battery fault identification model training method according to claim 5, characterized in that: Whether the target battery to be tested has an abnormal degradation fault can be determined by the following steps: Obtaining a horizontal coordinate value and a vertical coordinate value of a second probability distribution point of the target battery to be tested, wherein the horizontal coordinate value is a dynamic time warping value of the target battery to be tested and the reference battery, and the vertical coordinate value is an average differential voltage difference between the target battery to be tested and the reference battery; If the horizontal coordinate value is greater than the third preset threshold or the vertical coordinate value is greater than the fourth preset threshold, it is determined that the target battery to be tested has an abnormal degradation fault.

8. The battery fault identification model training method according to claim 1, characterized in that: The serialized test related to the battery characteristics is to test the voltage characteristics of the battery in sequence according to the time sequence in the preset time period.

9. A battery fault identification model training device, characterized in that: The device comprises: A testing unit, used to perform serial tests related to battery characteristics on a preset reference battery and each target battery to be tested, respectively, to obtain first time series test data of the reference battery and second time series test data of each target battery to be tested; A first calculation unit, configured to perform dynamic time warping on each of the second time series test data and the first time series test data one by one, to obtain a dynamic time warping value of each of the target battery to be tested and the reference battery; A second calculation unit, configured to calculate the difference of average differential voltage between each of the second time series test data and the first time series test data one by one, to obtain the average differential voltage difference between each of the target test cells and the reference cell; A training unit, used for inputting each of the dynamic time warping values ​​and each of the average differential voltage difference values ​​as sample feature data of each of the target batteries to be tested into a preset battery fault identification model, and training the battery fault identification model according to each of the sample feature data; The average differential voltage difference between the single target battery to be tested and the reference battery is obtained by the following steps: Determine first open circuit voltage sequence test data of the reference battery according to the first time series test data, and determine second open circuit voltage sequence test data of the target battery to be tested according to the second time series test data; Determining a first average differential voltage value of the reference battery according to each open circuit voltage value and the number of each open circuit voltage value in the first open circuit voltage sequence test data; Determining a second average differential voltage value of the target battery to be tested according to each open circuit voltage value and the number of each open circuit voltage value in the second open circuit voltage sequence test data; The average differential voltage difference value is determined based on the first average differential voltage value and the second average differential voltage value.

Images