Training method for battery fault detection model for distributed battery energy storage system

Abstract

A training method for a battery fault detection model for a distributed battery energy storage system, comprising: acquiring first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system (S101), the first sample feature data being used for describing the operation condition of the first battery in a normal operation state, and the second sample feature data being used for describing the operation condition of the first battery in a fault operation state; then acquiring first annotation information corresponding to the first sample feature data and second annotation information corresponding to the second sample feature data (S102); and on the basis of the first sample feature data, the second sample feature data, the first annotation information and the second annotation information, training an initial fault detection model to obtain a target fault detection model (S103).

Description

Training method for battery fault detection model for distributed battery energy storage system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on and claims the priority of Chinese patent application with application number 2024100966379 and application date January 23, 2024. The entire content of the Chinese patent application is hereby incorporated into this application by reference. Technical Field

[0003] The present disclosure relates to the technical field of battery fault detection, and in particular to a training method for a battery fault detection model for a distributed battery energy storage system. Background Art

[0004] Distributed battery energy storage systems are primarily composed of a battery energy storage system, a battery management system, a power conversion system, and a monitoring system. As an energy provider, it is crucial to accurately detect fault conditions in the battery energy storage system. However, distributed battery energy storage systems are often composed of multiple battery cells connected in series and parallel. During the charging and discharging process, complex nonlinear chemical reactions occur within the battery cells, making battery fault detection difficult. Summary of the Invention

[0005] The present disclosure aims to solve the technical problems in the related art at least to some extent.

[0006] To this end, the purpose of the present disclosure is to propose a training method, device, electronic device, storage medium, computer program product and computer program for a battery fault detection model for a distributed battery energy storage system.

[0007] An embodiment of a first aspect of the present disclosure proposes a training method for a battery fault detection model for a distributed battery energy storage system, including: obtaining first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system, wherein the first sample feature data is used to describe the operating condition of the first battery in a normal operating state, and the second sample feature data is used to describe the operating condition of the first battery in a faulty operating state; obtaining first labeling information corresponding to the first sample feature data, and second labeling information corresponding to the second sample feature data; and training an initial fault detection model based on the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information to obtain a target fault detection model.

[0008] A second aspect embodiment of the present disclosure proposes a battery fault detection method for a distributed battery energy storage system, including: obtaining third operating data of a second battery in the distributed battery energy storage system; inputting the third operating data into a target fault detection model trained by the training method for a fault detection model for a distributed battery energy storage system proposed in a corresponding aspect embodiment, so as to obtain a fault detection result corresponding to the second battery output by the target fault detection model.

[0009] An embodiment of a third aspect of the present disclosure proposes a training device for a battery fault detection model for a distributed battery energy storage system, including: a first acquisition module, used to obtain first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system, wherein the first sample feature data is used to describe the operating conditions of the first battery in a normal operating state, and the second sample feature data is used to describe the operating conditions of the first battery in a faulty operating state; a second acquisition module, used to obtain first labeling information corresponding to the first sample feature data, and second labeling information corresponding to the second sample feature data; a training module, used to train an initial fault detection model based on the first sample feature data, the second sample feature data, the first labeling information and the second labeling information to obtain a target fault detection model.

[0010] An embodiment of a fourth aspect of the present disclosure proposes a battery fault detection device for a distributed battery energy storage system, including: a third acquisition module, used to obtain third operating data of a second battery in the distributed battery energy storage system; a processing module, used to input the third operating data into a target fault detection model trained by a training device for a fault detection model for a distributed battery energy storage system as proposed in the embodiment of the third aspect, to obtain a fault detection result corresponding to the second battery output by the target fault detection model.

[0011] The fifth embodiment of the present disclosure proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements the training method of the battery fault detection model for a distributed battery energy storage system proposed in the first embodiment of the present disclosure, or implements the battery fault detection method for a distributed battery energy storage system proposed in the second embodiment of the present disclosure.

[0012] The sixth aspect embodiment of the present disclosure proposes a non-temporary computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, it implements the training method of the battery fault detection model for a distributed battery energy storage system as proposed in the first aspect embodiment of the present disclosure, or implements the battery fault detection method for a distributed battery energy storage system as proposed in the second aspect embodiment of the present disclosure.

[0013] The seventh aspect embodiment of the present disclosure proposes a computer program product. When the instruction processor in the computer program product is executed, it executes the training method of the battery fault detection model for the distributed battery energy storage system proposed in the first aspect embodiment of the present disclosure, or implements the battery fault detection method for the distributed battery energy storage system proposed in the second aspect embodiment of the present disclosure.

[0014] An eighth aspect embodiment of the present disclosure proposes a computer program, which includes computer program code. When the computer program code is run on a computer, the computer executes the training method of the battery fault detection model for a distributed battery energy storage system proposed in the first aspect embodiment of the present disclosure, or implements the battery fault detection method for a distributed battery energy storage system proposed in the second aspect embodiment of the present disclosure.

[0015] The training method of a battery fault detection model for a distributed battery energy storage system provided by an embodiment of the present disclosure may include the following beneficial effects: obtaining first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system, wherein the first sample feature data is used to describe the operating conditions of the first battery in a normal operating state, and the second sample feature data is used to describe the operating conditions of the first battery in a faulty operating state, then obtaining first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data, and training an initial fault detection model based on the first sample feature data, the second sample feature data, the first labeling information and the second labeling information to obtain a target fault detection model, thereby improving the training effect of the battery fault detection model for the distributed battery energy storage system, so that the trained target battery fault detection model for the distributed battery energy storage system can realize battery fault detection in combination with the operating data of the first battery, thereby improving the prediction accuracy of the battery fault detection model for the distributed battery energy storage system.

[0016] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present disclosure will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0018] FIG1 is a flow chart of a method for training a battery fault detection model for a distributed battery energy storage system according to an embodiment of the present disclosure;

[0019] FIG2 is a flow chart of a method for training a battery fault detection model for a distributed battery energy storage system according to another embodiment of the present disclosure;

[0020] FIG3 is a flow chart of a battery fault detection method for a distributed battery energy storage system according to an embodiment of the present disclosure;

[0021] FIG4 is a schematic structural diagram of a training device for a battery fault detection model of a distributed battery energy storage system according to an embodiment of the present disclosure;

[0022] FIG5 is a schematic structural diagram of a battery fault detection device for a distributed battery energy storage system proposed in one embodiment of the present disclosure;

[0023] FIG6 illustrates a block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure. DETAILED DESCRIPTION

[0024] The following describes in detail embodiments of the present disclosure, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present disclosure and are not to be construed as limiting the present disclosure. On the contrary, the embodiments of the present disclosure include all variations, modifications, and equivalents that fall within the spirit and scope of the appended claims.

[0025] It should be noted that the processes of data acquisition, collection, storage, use, and processing in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0026] FIG1 is a flow chart of a method for training a battery fault detection model for a distributed battery energy storage system according to an embodiment of the present disclosure.

[0027] It should be noted that the execution entity of the training method for a battery fault detection model for a distributed battery energy storage system in the embodiment of the present disclosure is a training device for a battery fault detection model for a distributed battery energy storage system. The device can be implemented by software and / or hardware. The device can be configured in an electronic device, and the electronic device may include but is not limited to a terminal, a server, etc.

[0028] As shown in FIG1 , the training method for a battery fault detection model for a distributed battery energy storage system includes: S101 - S103 .

[0029] S101: Acquire first sample characteristic data and second sample characteristic data of a first battery in a distributed battery energy storage system, wherein the first sample characteristic data is used to describe the operating condition of the first battery in a normal operating state, and the second sample characteristic data is used to describe the operating condition of the first battery in a faulty operating state.

[0030] The first sample characteristic data is used to describe the operating condition of the first battery in a normal operating state, and the second sample characteristic data is used to describe the operating condition of the first battery in a faulty operating state.

[0031] The first battery may be any battery in a distributed battery energy storage system, without limitation.

[0032] In some embodiments, obtaining first sample characteristic data and second sample characteristic data of a first battery in a distributed battery energy storage system may be obtaining first operating data and second operating data of the first battery, performing feature engineering processing on the first operating data to obtain a first data feature, and performing feature engineering processing on the second operating data to obtain a second data feature, and then obtaining the first sample characteristic data and the second sample characteristic data based on the first data feature and the second data feature.

[0033] The first operating data is operating data of the first battery in the distributed battery energy storage system when it is in a normal operating state. Correspondingly, the second operating data is operating data of the first battery in the distributed battery energy storage system when it is in a fault operating state.

[0034] In the embodiment of the present disclosure, the first operating data and the second operating data of the first battery are obtained by collecting the basic battery parameters and operating condition parameters of the first battery during normal operation as the first operating data through sensors, and collecting the basic battery parameters and operating condition parameters of the first battery during faulty operation as the second operating data through sensors.

[0035] The above-mentioned sensors include temperature sensors, optical fiber sensors, voltage sensors, hydrogen sensors, etc., and there is no limitation on this.

[0036] In the embodiment of the present disclosure, after obtaining the first operating data and the second operating data of the first battery, the first operating data and the second operating data can be processed to remove erroneous, incomplete, duplicate, useless, inconsistent, non-compliant and other problem data.

[0037] In the embodiment of the present disclosure, after obtaining the first operating data and the second operating data and performing data cleaning on the first operating data and the second operating data, feature engineering processing can be performed on the first operating data to obtain first data features, and feature engineering processing can be performed on the second operating data to obtain second data features.

[0038] In the embodiments of the present disclosure, feature engineering processing methods include but are not limited to data transformation, encoding, binning, classification statistics and feature combination, and are not limited to these.

[0039] In the embodiment of the present disclosure, the first data feature and the second data feature include at least one of the following: battery model data, battery brand data, working time data, battery temperature distribution data, voltage, current, power, power, voltage range and standard deviation, temperature range and standard deviation, number of charges and discharges, charge and discharge efficiency, energy loss rate, SOH, vibration frequency and amplitude, variance between the current charge and discharge voltage and the previous charge and discharge voltage, change trend of relative charging time, etc., and there is no restriction on this.

[0040] In an embodiment of the present disclosure, after feature engineering is performed on the first operating data to obtain the first data feature, and feature engineering is performed on the second operating data to obtain the second data feature, first sample feature data and second sample feature data can be obtained based on the first data feature and the second data feature.

[0041] In some embodiments, first sample feature data and second sample feature data are obtained based on the first data feature and the second data feature. The first data feature can be subjected to dimensionality reduction processing to obtain a third data feature, and the second data feature can be subjected to dimensionality reduction processing to obtain a fourth data feature, and then the first quantity of the third data feature and the second quantity of the fourth data feature are determined, and then the first sample feature data and the second sample feature data are obtained based on the first quantity, the second quantity, the third data feature and the fourth data feature.

[0042] In the embodiment of the present disclosure, the dimensionality reduction processing is performed on the first data feature, and principal component analysis processing can be performed on the first data feature, so as to determine a data feature with a high information rate from the first data feature as the third data feature.

[0043] In the embodiment of the present disclosure, the dimensionality reduction processing is performed on the second data feature, and principal component analysis processing can be performed on the second data feature, so as to determine a data feature with a high information rate from the second data feature as the fourth data feature.

[0044] In the embodiment of the present disclosure, after performing dimensionality reduction processing on the first data feature to obtain the third data feature, and performing dimensionality reduction processing on the second data feature to obtain the fourth data feature, the first quantity of the third data feature and the second quantity of the fourth data feature can be determined, and then first sample feature data and second sample feature data can be obtained based on the first quantity, the second quantity, the third data feature, and the fourth data feature.

[0045] In some embodiments, first sample feature data and second sample feature data are obtained based on the first quantity, the second quantity, the first data feature, and the second data feature. This can be based on synthetic minority class oversampling technology, processing the fourth data feature according to the first quantity and the second quantity to obtain the fifth data feature, and then standardizing the fifth data feature to obtain the second sample feature data, and then standardizing the third data feature to obtain the first sample feature data.

[0046] That is to say, in the embodiment of the present disclosure, the weight coefficient G can be determined according to the first quantity and the second quantity, G=(m l -m s )×β, where m l is the first quantity, m s is the second quantity, β is a random number between [0,1]. If β is equal to 1, the ratio of the third data feature to the fifth data feature after increment is 1:1. Set the proximity index K and calculate the proportion of normal samples in the K nearest neighbors of each sample r i =Δ i / K,Δ i is the number of majority class samples in K nearest neighbors, i=1,2,3,...,m s , for r i Normalization is performed to obtain Finally, the number of new samples generated for a few fault samples is After determining the number, for each fourth data feature x in the minority class, calculate the distance between the point and other sample points in the minority class to obtain the nearest k neighbors. Then, set a sampling ratio according to the sample imbalance ratio to determine the sampling rate. For each fourth data feature x in the minority class, randomly select several samples from its k nearest neighbors. Assume that the selected nearest neighbor is x'. For each randomly selected nearest neighbor x', respectively, construct the fifth data feature with the fourth data feature according to the following formula: new =x+rand(0,1)*(x′-x).

[0047] In the embodiment of the present disclosure, the fifth data feature is normalized to obtain the second sample feature data, and the third data feature is then normalized to obtain the first sample feature data. The z-score normalization method may be used. Specifically, the average value of the data features (the third data feature and the fifth data feature) may be obtained. and standard deviation s, where n is the number of data features, x j is the data feature, j = 1, 2, 3, ... n, and then normalized to obtain the sample feature data y after normalization j (first sample feature data and second sample feature data),

[0048] Alternatively, in the embodiment of the present disclosure, the fifth data feature can be standardized using the maximum and minimum standardization method to obtain second sample feature data, and then the third data feature can be standardized to obtain first sample feature data. There is no limitation on this.

[0049] S102: Acquire first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data.

[0050] Among them, the annotation information can be used to determine the specific convergence timing of the model during the model training process. The annotation information corresponding to the first sample feature is the first annotation information, and the annotation information corresponding to the second sample feature is the second annotation information.

[0051] In the embodiment of the present disclosure, obtaining first annotation information corresponding to the first sample characteristic data may be determining a battery fault detection result corresponding to the first sample characteristic data. Correspondingly, obtaining second annotation information corresponding to the second sample characteristic data may be determining a battery fault detection result corresponding to the second sample characteristic data, and using the battery fault detection result corresponding to the first sample characteristic data as the first annotation information, and using the battery fault detection result corresponding to the second sample characteristic data as the second annotation information.

[0052] S103: Training an initial fault detection model according to the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information to obtain a target fault detection model.

[0053] After obtaining the first labeling information corresponding to the first sample feature data and the second labeling information corresponding to the second sample feature data, the embodiment of the present disclosure can train an initial fault detection model based on the first sample feature data, the second sample feature data, the first labeling information and the second labeling information to obtain a target fault detection model.

[0054] In the embodiment of the present disclosure, the initial fault detection model adopts a fusion model. The fusion model is not only a fusion of the results of different single models, but also can bring the loss functions, parameter adjustment results and fitting processes of different models as training sets into other models for deeper fusion. The initial fault detection model includes: elastic network regression model (ElasticNet regression), kernel ridge regression model (Kernel Ridge Regression, KRR), lightweight gradient boosting machine model (Light Gradient Boosting Machine, LightGBM) and back propagation neural network model, and there is no restriction on this.

[0055] In some embodiments, an initial fault detection model is trained based on the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information to obtain a target fault detection model. The first sample feature data can be input into the initial fault detection model to obtain the first prediction information output by the initial fault detection model, and the second sample feature data can be input into the initial fault detection model to obtain the second prediction information output by the initial fault detection model. Then, based on the loss function preset for the initial fault detection model, a first loss value between the first prediction information and the first labeling information, and a second loss value between the second prediction information and the second labeling information are determined. Then, a weighted summation process is performed on the first loss value and the second loss value to obtain a target loss value. When the target loss value is less than the loss threshold, it is determined that the initial fault detection model has converged, and the trained initial fault detection model is used as the target fault detection model. There is no restriction on this.

[0056] In an embodiment of the present disclosure, first sample feature data and second sample feature data of a first battery in a distributed battery energy storage system are obtained, wherein the first sample feature data is used to describe the operating condition of the first battery in a normal operating state, and the second sample feature data is used to describe the operating condition of the first battery in a faulty operating state. Then, first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data are obtained, and an initial fault detection model is trained based on the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information to obtain a target fault detection model, thereby improving the training effect of the battery fault detection model for the distributed battery energy storage system, so that the trained target battery fault detection model for the distributed battery energy storage system can realize battery fault detection in combination with the operating data of the first battery, thereby improving the prediction accuracy of the battery fault detection model for the distributed battery energy storage system.

[0057] FIG2 is a flow chart of a method for training a battery fault detection model for a distributed battery energy storage system according to another embodiment of the present disclosure.

[0058] As shown in FIG2 , the training method for the battery fault detection model for the distributed battery energy storage system includes: S201 - S203 .

[0059] S201: Acquire first sample characteristic data and second sample characteristic data of a first battery in a distributed battery energy storage system, wherein the first sample characteristic data is used to describe the operating condition of the first battery in a normal operating state, and the second sample characteristic data is used to describe the operating condition of the first battery in a faulty operating state.

[0060] S202: Acquire first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data.

[0061] The detailed description of S201 - S202 can be found in the above embodiment and will not be repeated here.

[0062] S203: Input the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information into the initial fault detection model, and iteratively optimize the model parameters of the initial fault detection model using convergence and divergence operations until the initial fault detection model converges to obtain a target fault detection model.

[0063] In the embodiment of the present disclosure, the model parameters can be represented by the information matrix P, P = [p1, p2, p i , p m ], p m Represents the probability of the number of nodes m appearing in the hidden layer; m represents the number of nodes in the hidden layer, and satisfies

[0064] Among them, the model parameter winning subgroup and temporary subgroup perform convergence operations in parallel:

[0065] That is, is a winner, and the corresponding information matrix is ​​p * , guided by the winner, the individuals generated are (a′1a′2…a′ i …a′ n ), with information matrix p * To guide the generation of the node a′ at the corresponding position in the individual i .like Then go directly to the next generation; if Then, new individuals are generated according to the following steps:

[0066] (1) On each layer of the winner, the probability value S of each node number is accumulated in turn i , the last cumulative value is S m , and S m =1;

[0067] (2) Generate a uniformly distributed random number R in the interval [0,1];

[0068] (3) Use S in sequence i Compared with R, S appears for the first time i Nodes with ≥R are selected;

[0069] (4) Repeat (2) and (3) until the entire subpopulation is generated.

[0070] Then, the winner information matrix is ​​generated, and the winner is (a′1a′2…a′ i …a′ n ), when the number of nodes of the winner is 'a' i The position b′ corresponding to the information matrix p′ on the local bulletin board j When the number of nodes represented is equal, the probability value of this position increases, otherwise it decreases. Let p * The influence coefficient on p′ is λ s ,λ s ∈(0,1), j∈[1,m], let Then when the number of hidden layer nodes j∈[1,L+1]:

[0071] When the number of hidden layer nodes j∈[L+2,m]:

[0072] The alienation operation process includes:

[0073] (1) When a temporary subgroup u matures, its information matrix is ​​denoted as U. At this time, the local bulletin board information matrix is ​​denoted as U′, and the global bulletin board information matrix is ​​denoted as U * , let U * The probability of a row in is The information on the corresponding position of the local bulletin board information matrix U′ of u is (p′1p′2…p′ i …p′ m ), let the average score of all previously mature subgroups be S * , u's score is s, then:

[0074] Where j∈[1,m] is an integer and λ>0 is the sensitivity coefficient. The above formula can make the subgroups with higher scores have a greater influence on the global bulletin board, while weakening the influence of the subgroups with lower scores on the global bulletin board.

[0075] (2) Let (p1p2…p i …p m ) is the information corresponding to a node to be generated, and (p′1p′2…p′ i …p′ m ) are the node information of the global bulletin board and the winning subgroup, an integer in L∈[0,m]. Then:

[0076] Among them, ω g is the coefficient for absorbing global bulletin board information; ω b To absorb the information coefficient of the current winning subgroup; ω r is the innovation coefficient, and ωg +ω b +ω r =1.

[0077] (3) Repeat (1)-(2) until the score of the optimal winning sub-group no longer changes, then determine the model parameters and complete the iterative optimization.

[0078] In an embodiment of the present disclosure, first sample feature data and second sample feature data of a first battery in a distributed battery energy storage system are obtained, wherein the first sample feature data is used to describe the operating condition of the first battery in a normal operating state, and the second sample feature data is used to describe the operating condition of the first battery in a faulty operating state. Then, first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data are obtained. Then, the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information are input into an initial fault detection model. The model parameters of the initial fault detection model are iteratively optimized using convergence and divergence operations until the initial fault detection model converges to obtain a target fault detection model, thereby improving the training effect of the battery fault detection model for the distributed battery energy storage system, so that the trained target battery fault detection model for the distributed battery energy storage system can realize battery fault detection in combination with the operating data of the first battery, thereby improving the prediction accuracy of the battery fault detection model for the distributed battery energy storage system.

[0079] FIG3 is a flow chart of a battery fault detection method for a distributed battery energy storage system according to an embodiment of the present disclosure.

[0080] It should be noted that the execution subject of the battery fault detection method for a distributed battery energy storage system in the embodiment of the present disclosure is a battery fault detection device for a distributed battery energy storage system. The device can be implemented by software and / or hardware. The device can be configured in an electronic device, and the electronic device may include but is not limited to a terminal, a server, etc.

[0081] As shown in FIG3 , the battery fault detection method for a distributed battery energy storage system includes: S301 - S302 .

[0082] S301: Acquire third operating data of a second battery in a distributed battery energy storage system.

[0083] For the explanation of the same terms in the embodiments of the present disclosure and the above embodiments, please refer to the above embodiments for details and will not be repeated here.

[0084] The second battery is a battery in the distributed battery energy storage system whose fault is to be detected.

[0085] That is, in the embodiment of the present disclosure, the third operating data of the second battery in the distributed battery energy storage system may be collected based on the sensor, and then fault detection may be performed on the second battery in the distributed battery energy storage system based on the third operating data.

[0086] S302: Inputting the third operating data into a target fault detection model trained by a training method for a fault detection model of a distributed battery energy storage system to obtain a fault detection result corresponding to the second battery output by the target fault detection model.

[0087] After obtaining the third operating data of the second battery in the distributed battery energy storage system, the embodiment of the present disclosure can input the third operating data into a target fault detection model trained by a training method for a fault detection model of a distributed battery energy storage system, so as to obtain a fault detection result corresponding to the second battery output by the target fault detection model.

[0088] In an embodiment of the present disclosure, third operating data of a second battery in a distributed battery energy storage system is obtained, and then the third operating data is input into a target fault detection model trained by a training method for a fault detection model for a distributed battery energy storage system, so as to obtain a fault detection result corresponding to the second battery output by the target fault detection model. This can effectively improve the fault detection effect of batteries in the distributed battery energy storage system when performing battery fault detection for the distributed battery energy storage system based on the target battery fault detection model for the distributed battery energy storage system.

[0089] FIG4 is a schematic structural diagram of a training device for a battery fault detection model for a distributed battery energy storage system according to an embodiment of the present disclosure.

[0090] As shown in FIG4 , the training device 40 for a battery fault detection model of a distributed battery energy storage system includes:

[0091] A first acquisition module 401 is configured to acquire first sample characteristic data and second sample characteristic data of a first battery in a distributed battery energy storage system, wherein the first sample characteristic data is used to describe the operation of the first battery in a normal operating state, and the second sample characteristic data is used to describe the operation of the first battery in a faulty operating state;

[0092] A second acquisition module 402 is configured to acquire first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data;

[0093] The training module 403 is used to train an initial fault detection model according to the first sample feature data, the second sample feature data, the first labeling information and the second labeling information to obtain a target fault detection model.

[0094] In some embodiments of the present disclosure, the second obtaining module 402 is further configured to:

[0095] acquiring first operating data and second operating data of the first battery;

[0096] Performing feature engineering on the first operating data to obtain a first data feature, and performing feature engineering on the second operating data to obtain a second data feature;

[0097] First sample feature data and second sample feature data are acquired according to the first data feature and the second data feature.

[0098] In some embodiments of the present disclosure, the second obtaining module 402 is further configured to:

[0099] Performing dimensionality reduction processing on the first data feature to obtain a third data feature, and performing dimensionality reduction processing on the second data feature to obtain a fourth data feature;

[0100] determining a first quantity of a third data feature and a second quantity of a fourth data feature;

[0101] First sample feature data and second sample feature data are obtained according to the first quantity, the second quantity, the third data feature, and the fourth data feature.

[0102] In some embodiments of the present disclosure, the second obtaining module 402 is further configured to:

[0103] processing the fourth data feature according to the first quantity and the second quantity based on a synthetic minority class oversampling technique to obtain a fifth data feature;

[0104] performing standardization processing on the third data feature to obtain first sample feature data;

[0105] The fifth data feature is normalized to obtain second sample feature data.

[0106] In some embodiments of the present disclosure, the training module 403 is further configured to:

[0107] The first sample feature data, the second sample feature data, the first labeling information and the second labeling information are input into the initial fault detection model, and the model parameters of the initial fault detection model are iteratively optimized using convergence and divergence operations until the initial fault detection model converges to obtain a target fault detection model.

[0108] Corresponding to the training method for the battery fault detection model for a distributed battery energy storage system provided in the embodiments of Figures 1 to 2 above, the present disclosure also provides a training device for a battery fault detection model for a distributed battery energy storage system. Since the training device for the battery fault detection model for a distributed battery energy storage system provided in the embodiments of the present disclosure corresponds to the training method for the battery fault detection model for a distributed battery energy storage system provided in the embodiments of Figures 1 to 2 above, the implementation method of the training method for the battery fault detection model for a distributed battery energy storage system is also applicable to the training device for the battery fault detection model for a distributed battery energy storage system provided in the embodiments of the present disclosure, and will not be described in detail in the embodiments of the present disclosure.

[0109] In an embodiment of the present disclosure, first sample feature data and second sample feature data of a first battery in a distributed battery energy storage system are obtained, wherein the first sample feature data is used to describe the operating condition of the first battery in a normal operating state, and the second sample feature data is used to describe the operating condition of the first battery in a faulty operating state. Then, first labeling information corresponding to the first sample feature data and second labeling information corresponding to the second sample feature data are obtained, and an initial fault detection model is trained based on the first sample feature data, the second sample feature data, the first labeling information, and the second labeling information to obtain a target fault detection model, thereby improving the training effect of the battery fault detection model for the distributed battery energy storage system, so that the trained target battery fault detection model for the distributed battery energy storage system can realize battery fault detection in combination with the operating data of the first battery, thereby improving the prediction accuracy of the battery fault detection model for the distributed battery energy storage system.

[0110] FIG5 is a schematic structural diagram of a battery fault detection device for a distributed battery energy storage system proposed in one embodiment of the present disclosure.

[0111] As shown in FIG5 , the battery fault detection device 50 for a distributed battery energy storage system includes:

[0112] A third acquisition module 501 is used to acquire third operating data of a second battery in the distributed battery energy storage system;

[0113] The processing module 502 is configured to input the third operating data into a target fault detection model trained by a training device for a fault detection model of a distributed battery energy storage system, so as to obtain a fault detection result corresponding to the second battery output by the target fault detection model.

[0114] In an embodiment of the present disclosure, third operating data of a second battery in a distributed battery energy storage system is obtained, and then the third operating data is input into a target fault detection model trained by a training method for a fault detection model for a distributed battery energy storage system, so as to obtain a fault detection result corresponding to the second battery output by the target fault detection model. This can effectively improve the fault detection effect of batteries in the distributed battery energy storage system when performing battery fault detection for the distributed battery energy storage system based on the target battery fault detection model for the distributed battery energy storage system.

[0115] Corresponding to the battery fault detection method for a distributed battery energy storage system provided in the embodiment of Figure 3 above, the present disclosure also provides a battery fault detection device for a distributed battery energy storage system. Since the battery fault detection device for a distributed battery energy storage system provided in the embodiment of the present disclosure corresponds to the battery fault detection method for a distributed battery energy storage system provided in the embodiment of Figure 3 above, the implementation of the battery fault detection method for a distributed battery energy storage system is also applicable to the battery fault detection device for a distributed battery energy storage system provided in the embodiment of the present disclosure, and will not be described in detail in the embodiment of the present disclosure.

[0116] In order to implement the above embodiments, the present disclosure further proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the training method for a battery fault detection model for a distributed battery energy storage system proposed in the above embodiments of the present disclosure is implemented, or the battery fault detection method for a distributed battery energy storage system proposed in the above embodiments of the present disclosure is implemented.

[0117] In order to implement the above embodiments, the present disclosure also proposes a non-transitory computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the training method of the battery fault detection model for a distributed battery energy storage system proposed in the above embodiments of the present disclosure is implemented, or the battery fault detection method for a distributed battery energy storage system proposed in the above embodiments of the present disclosure is implemented.

[0118] In order to implement the above embodiments, the present disclosure also proposes a computer program product. When the instruction processor in the computer program product is executed, it executes the training method of the battery fault detection model for the distributed battery energy storage system proposed in the above embodiments of the present disclosure, or implements the battery fault detection method for the distributed battery energy storage system proposed in the above embodiments of the present disclosure.

[0119] In order to implement the above embodiments, the present disclosure also proposes a computer program, which includes computer program code. When the computer program code is run on a computer, it enables the computer to execute the training method of the battery fault detection model for the distributed battery energy storage system proposed in the above embodiments of the present disclosure, or implement the battery fault detection method for the distributed battery energy storage system proposed in the above embodiments of the present disclosure.

[0120] Figure 6 shows a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present disclosure. The electronic device shown in Figure 6 is only an example and should not limit the functions and scope of use of the embodiments of the present disclosure.

[0121] 6 , the electronic device is implemented as a general-purpose computing device. Components of the electronic device may include, but are not limited to, one or more processors or processing units 16 , a system memory 28 , and a bus 18 connecting various system components (including the system memory 28 and the processing unit 16 ).

[0122] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnection (PCI) bus.

[0123] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0124] Memory 28 may include computer-readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device may further include other removable / non-removable, volatile / non-volatile computer-readable storage media. By way of example only, storage system 34 may be configured to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 6 , and commonly referred to as a "hard drive").

[0125] Although not shown in FIG6 , a disk drive for reading and writing to a removable non-volatile disk (e.g., a floppy disk) and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a Compact Disc Read Only Memory (CD-ROM), a Digital Video Disc Read Only Memory (DVD-ROM), or other optical media) can be provided. In these cases, each drive can be connected to the bus 18 via one or more data media interfaces. The memory 28 can include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present disclosure.

[0126] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally implement the functions and / or methods of the embodiments described herein.

[0127] The electronic device can also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication can occur via an input / output (I / O) interface 22. Furthermore, the electronic device can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device via a bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0128] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the training method of the battery fault detection model for the distributed battery energy storage system mentioned in the above embodiment.

[0129] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow from the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0130] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

[0131] It should be noted that, in the description of this disclosure, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of this disclosure, unless otherwise specified, the meaning of "plurality" is two or more.

[0132] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.

[0133] It should be understood that various parts of the present disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0134] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0135] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0136] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0137] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0138] Although the embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are illustrative and are not to be construed as limitations on the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A training method for a battery fault detection model of a distributed battery energy storage system, comprising: Obtaining first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system, wherein the first sample feature data is used to describe the operation of the first battery in a normal operating state, and the second sample feature data is used to describe the operation of the first battery in a fault operating state; Obtaining first annotation information corresponding to the first sample feature data and second annotation information corresponding to the second sample feature data; Training the initial fault detection model according to the first sample feature data, the second sample feature data, the first annotation information and the second annotation information to obtain a target fault detection model.

2. The method according to claim 1, wherein, The obtaining of the first sample feature data and the second sample feature data of the first battery in the distributed battery energy storage system comprises: Obtaining first operation data and second operation data of the first battery; Performing feature engineering processing on the first operation data to obtain first data features, and performing feature engineering processing on the second operation data to obtain second data features; Obtaining the first sample feature data and the second sample feature data according to the first data features and the second data features.

3. The method according to claim 2, wherein, The obtaining of the first sample feature data and the second sample feature data according to the first data features and the second data features comprises: Performing dimensionality reduction processing on the first data features to obtain third data features, and performing dimensionality reduction processing on the second data features to obtain fourth data features; Determining a first quantity of the third data features and a second quantity of the fourth data features; Obtaining the first sample feature data and the second sample feature data according to the first quantity, the second quantity, the third data features and the fourth data features.

4. The method according to claim 3, wherein, The obtaining of the first sample feature data and the second sample feature data according to the first quantity, the second quantity, the third data features and the fourth data features comprises: Based on the synthetic minority over-sampling technique, processing the fourth data features according to the first quantity and the second quantity to obtain fifth data features; Performing standardization processing on the third data features to obtain the first sample feature data; Performing standardization processing on the fifth data features to obtain the second sample feature data.

5. The method according to any one of claims 1 to 4, wherein, The training of the initial fault detection model according to the first sample feature data, the second sample feature data, the first annotation information and the second annotation information to obtain a target fault detection model comprises: Inputting the first sample feature data, the second sample feature data, the first annotation information and the second annotation information into the initial fault detection model, and iteratively optimizing the model parameters of the initial fault detection model by using convergence and divergence operations until the initial fault detection model converges to obtain the target fault detection model.

6. A battery fault detection method for a distributed battery energy storage system, comprising: Obtain the third operation data of the second battery in the distributed battery energy storage system; Input the third operation data into the target fault detection model trained by the training method of the battery fault detection model for the distributed battery energy storage system according to any one of claims 1-5, so as to obtain the fault detection result corresponding to the second battery output by the target fault detection model.

7. A training device for a battery fault detection model for a distributed battery energy storage system, comprising: A first acquisition module, configured to acquire first sample feature data and second sample feature data of a first battery in the distributed battery energy storage system, wherein the first sample feature data is used to describe the operation of the first battery in a normal operation state, and the second sample feature data is used to describe the operation of the first battery in a fault operation state; A second acquisition module, configured to acquire first annotation information corresponding to the first sample feature data and second annotation information corresponding to the second sample feature data; A training module, configured to train the initial fault detection model according to the first sample feature data, the second sample feature data, the first annotation information, and the second annotation information, so as to obtain a target fault detection model.

8. A battery fault detection device for a distributed battery energy storage system, comprising: A third acquisition module, configured to acquire third operation data of a second battery in the distributed battery energy storage system; A processing module, configured to input the third operation data into the target fault detection model trained by the training device of the battery fault detection model for the distributed battery energy storage system according to claim 7, so as to obtain the fault detection result corresponding to the second battery output by the target fault detection model.

9. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-6.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

11. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1-6 is implemented.

12. A computer program, the computer program comprising computer program code, wherein when the computer program code runs on a computer, the computer is caused to execute the method according to any one of claims 1-6.

Images