Method for screening abnormal single batteries in battery pack and related device

Abstract

The invention belongs to the technical field of battery anomaly detection, and discloses a method for screening abnormal single batteries in a battery pack and a related device. The method for screening the abnormal single batteries in the battery pack comprises the following steps: acquiring a charge state and a polarization voltage based on acquired current and external voltage of each single battery at different moments; on the basis of the obtained charge state, obtaining the temperature of each battery monomer by using a change function of the model parameters along with the temperature and the charge state; based on the obtained polarization voltage, obtaining an open-circuit voltage of each battery monomer; and respectively calculating the average deviation, the standard score and the Euclidean distance of the charge state, the temperature and the open-circuit voltage of each battery monomer, obtaining characteristic quantities, and carrying out clustering analysis to obtain a screening result of the abnormal battery monomers. According to the technical scheme disclosed by the invention, the abnormal battery monomers in the battery pack can be screened, and the method has the advantages of relatively high efficiency, relatively low cost and relatively high reliability.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery abnormality detection, and in particular relates to a method for screening abnormal battery cells in a battery pack and a related device. Background Art

[0002] In the application of battery packs in electric vehicles and energy storage systems, the consistency management of battery cells is particularly important; further explained, in electric vehicles and energy storage systems, each battery cell inside the battery pack must maintain a high degree of consistency to ensure the safe and reliable operation of the entire battery pack system. If there are battery cells with abnormal performance, it may have a negative impact on the operation of the entire battery pack and shorten the service life of the battery pack. Therefore, timely identification and screening of abnormal battery cells and early warning and processing are extremely important for ensuring the safe and reliable operation of the battery pack, improving the energy utilization efficiency of the battery cells in the battery pack, and extending the service life of the battery pack.

[0003] At present, most of the existing solutions for screening abnormal battery cells in battery packs are still based on testing methods, which have technical problems such as low efficiency, high cost and poor reliability. In order to effectively screen abnormal battery cells, maintain the overall performance of the battery pack, and ensure the stable and reliable operation of the battery pack under various working conditions, a new method for screening abnormal battery cells in battery packs is urgently needed. Summary of the invention

[0004] The purpose of the present invention is to provide a method and a related device for screening abnormal battery cells in a battery pack to solve one or more of the above-mentioned technical problems. The technical solution disclosed in the present invention is specifically a screening solution for abnormal battery cells in a battery pack that takes into account the battery state of charge, temperature and open circuit voltage, which can realize the screening of abnormal battery cells in a battery pack and has the advantages of high efficiency, low cost and high reliability.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for screening abnormal battery cells in a battery pack, comprising the following steps: Obtain the current and external voltage of each battery cell in the battery pack to be evaluated at different times; Based on the current and external voltage of each battery cell at different times, the state of charge and polarization voltage of each battery cell are obtained; Based on the acquired state of charge of each battery cell, the temperature of each battery cell in the battery pack is obtained by reversely calculating the function of the model parameters changing with temperature and state of charge when the state of charge and model parameters are known; Based on the polarization voltage of each battery cell obtained, the open circuit voltage of each battery cell in the battery pack is obtained by combining the external voltage of the battery cell with the battery equivalent circuit model; Based on the obtained state of charge, temperature and open circuit voltage of each battery cell, the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell are calculated respectively to obtain characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack; Based on the obtained characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack, a cluster analysis is performed to obtain abnormal battery cells in the battery pack to be evaluated.

[0006] A further improvement of the present invention is that the specific steps of obtaining the state of charge and polarization voltage of each battery cell based on the current and external voltage of each battery cell at different times include: Based on the current and external voltage of each battery cell at different times, the unscented Kalman filter algorithm is used to estimate the state of charge and polarization voltage of each battery cell.

[0007] A further improvement of the present invention is that the step of obtaining the function of the model parameters changing with temperature and state of charge comprises: Based on the external characteristics of the battery cell, a battery equivalent circuit model is established; Perform model parameter identification on the battery equivalent circuit model, obtain model component parameters at different temperatures and charge states, and obtain the function of model parameters changing with temperature and charge state; Among them, when identifying the model parameters of the battery equivalent circuit model: temperature and state of charge are used as input, and the model parameter values ​​are used as output; all pre-acquired sample data are randomly divided into training set and verification set according to a certain ratio, the training set is imported and polynomial fitting is used to establish the function of the model parameters changing with temperature and state of charge, the verification set is imported and the fitting accuracy of the change function is verified.

[0008] A further improvement of the present invention is that the battery equivalent circuit model includes: a controlled voltage source, an internal resistance and two series-connected resistor-capacitor parallel networks; wherein, The controlled voltage source is used to characterize the relationship between the open circuit voltage and the state of charge of the battery, the internal resistance is used to characterize the self-discharge phenomenon of the battery, and the two series-connected resistor-capacitor parallel networks are used to characterize different dynamic polarization behaviors of the battery; The equation of the battery equivalent circuit model is expressed as:

[0009] In the formula, U t ( k )express k The external battery voltage at the moment;OCV express k The battery open circuit voltage at the moment; U p1 ( k )express k The polarization voltage of the first resistor-capacitor parallel network at time t; U p2 ( k )express k The polarization voltage of the second resistor-capacitor parallel network at time t; I ( k )express k The current flowing through the battery at all times; R o Represents the internal resistance of the model.

[0010] A further improvement of the present invention is that, in the step of respectively calculating the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell based on the obtained state of charge, temperature and open circuit voltage of each battery cell to obtain a characteristic quantity reflecting the degree of discreteness of each battery cell and the overall battery pack, The calculation expression of the average deviation is:

[0011]

[0012] In the formula, X D,i The battery pack i The average deviation of the battery cell values; X D,i,t yes t The battery pack is i The deviation of the battery cell value, n represents the total number at the moment; X i,t yes t Moment i A battery cell value, which is the state of charge, temperature or open circuit voltage of the battery cell; X μ,t yes t The average value of all battery cells in the battery pack at the moment; The calculation expression of the standard score is:

[0013] ; In the formula, X Z,i The battery pack i Standard fraction of each battery cell value; X σ,t yest The standard deviation of the values ​​of all battery cells in the battery pack at the moment; X Z,i,t yes t The battery pack is i Standard fraction of each battery cell value; The calculation expression of Euclidean distance is:

[0014]

[0015] In the formula, X d,i The battery pack i The Euclidean distance of the battery cell values; X d,i,t yes t The battery pack is i The Euclidean distance between the values ​​of the individual cells.

[0016] A further improvement of the present invention lies in that, in the step of performing cluster analysis based on the obtained characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack to obtain abnormal battery cells in the battery pack to be evaluated, before performing the cluster analysis, the characteristic quantities of all battery cells are unified and normalized in terms of correlation direction so that the characteristic quantities of all battery cells are in the same dimension.

[0017] A further improvement of the present invention is that in the step of performing cluster analysis based on the obtained characteristic quantity reflecting the degree of discreteness of each battery cell and the overall battery pack to obtain abnormal battery cells in the battery pack to be evaluated, a DBSCAN clustering algorithm is used for cluster analysis.

[0018] In a second aspect, the present invention provides a system for screening abnormal battery cells in a battery pack, comprising: A data acquisition module is used to obtain the current and external voltage of each battery cell in the battery pack to be evaluated at different times; A first calculation module, used to obtain the state of charge and polarization voltage of each battery cell based on the current and external voltage of each battery cell at different times; A second calculation module is used to obtain the temperature of each battery cell in the battery pack by inversely calculating the change function of the model parameters with temperature and the state of charge based on the acquired state of charge of each battery cell when the state of charge and the model parameters are known; A third calculation module is used to obtain the open circuit voltage of each battery cell in the battery pack based on the obtained polarization voltage of each battery cell, combined with the external voltage of the battery cell and the battery equivalent circuit model; A fourth calculation module is used to calculate the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell based on the obtained state of charge, temperature and open circuit voltage of each battery cell, and obtain a characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack; The analysis and screening module is used to perform cluster analysis based on the obtained characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack, and obtain abnormal battery cells in the battery pack to be evaluated.

[0019] According to a third aspect of the present invention, there is provided an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for screening abnormal battery cells in a battery pack as described in any one of the first aspect of the present invention is implemented.

[0020] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for screening abnormal battery cells in a battery pack as described in any one of the first aspects of the present invention is implemented.

[0021] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a method for screening abnormal battery cells in a battery pack. Based on the function of the model parameters changing with temperature and state of charge, the temperature of the battery cell can be estimated more accurately, the number of temperature sensors in the battery pack can be reduced, and the cost can be reduced. The temperature and open circuit voltage of the battery cell are obtained based on the acquired state of charge and polarization voltage of the battery cell, which increases the judgment criteria for battery inconsistency and improves the accuracy and reliability of screening. Based on the acquired characteristic quantity, a clustering algorithm is used to obtain the abnormal battery corresponding to the outlier, and the screening of abnormal battery cells in the battery pack is realized. In summary, the technical solution of the present invention, after obtaining the state of charge, battery temperature and open circuit voltage data, completes the screening of abnormal battery cells in the battery pack by analyzing the discreteness of the state of charge, battery temperature and open circuit voltage data, and has the advantages of low cost and high accuracy and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below; obviously, the drawings described below are some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 is a schematic flow chart of a method for screening abnormal battery cells in a battery pack according to an embodiment of the present invention; Figure 2 It is a flow chart of a method for screening abnormal battery cells in a battery pack based on model parameter identification and state of charge estimation in a specific embodiment of the present invention; Figure 3 is a schematic diagram of a battery equivalent circuit model in an embodiment of the present invention; Figure 4 is a schematic diagram of changes in model parameters with temperature and state of charge in an embodiment of the present invention; Figure 5 is a flow chart of establishing a function of model parameters changing with temperature and state of charge in an embodiment of the present invention; Figure 6 is a schematic diagram of a process for finding abnormal battery cells based on feature quantities and a DBSCAN clustering algorithm in an embodiment of the present invention; Figure 7 It is a schematic diagram of a system for screening abnormal battery cells in a battery pack according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments and technical solutions are only part of the embodiments of the present invention, not all of the embodiments.

[0025] All other embodiments obtained by those of ordinary skill in the art without creative work based on the technical solutions disclosed in the embodiments of the present invention belong to the scope of protection of the present invention. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] See also Figure 1 , an embodiment of the present invention provides a method for screening abnormal battery cells in a battery pack, comprising the following steps: Step 1, obtaining the current and external voltage of each battery cell in the battery pack to be evaluated at different times; Step 2, based on the current and external voltage of each battery cell at different times obtained in step 1, the state of charge and polarization voltage of each battery cell are obtained; wherein, the state of charge and polarization voltage of each battery cell can be estimated using an Unscented Kalman Filter (UKF) algorithm; Step 3, based on the state of charge of each battery cell obtained in step 2, the temperature of each battery cell in the battery pack is obtained by reverse calculation using the function of the change of model parameters with temperature and state of charge when the state of charge and model parameters are known; Step 4, based on the polarization voltage of each battery cell obtained in step 2, combined with the external voltage of the battery cell and the battery equivalent circuit model, the open circuit voltage of each battery cell in the battery pack is obtained; wherein the battery equivalent circuit model is established based on the external characteristics of the battery cell; Step 5, based on the obtained state of charge, temperature and open circuit voltage of each battery cell, respectively calculate the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell, and obtain the characteristic quantity reflecting the degree of discreteness of each battery cell and the overall battery pack; in a further preferred technical solution, the characteristic quantities of all battery cells are unified and normalized in the correlation direction, so that the characteristic quantities of all battery cells are in the same dimension, avoiding the situation of large errors in the subsequent calculation process; Step 6: Based on the characteristic quantity reflecting the degree of discreteness of each battery cell and the overall battery pack obtained in step 5, a selected clustering algorithm is used to perform cluster analysis on the characteristic quantities of charge state, temperature and open circuit voltage of the battery cells to obtain abnormal battery cells in the battery pack; wherein the selected clustering algorithm may be a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm.

[0027] In the technical solution disclosed in the embodiment of the present invention, by using the unscented Kalman filter algorithm, the state of charge and polarization voltage of the battery cell can be estimated more accurately, and high-precision state of charge and polarization voltage estimation can be achieved; explanatory, the UKF algorithm performs well in processing nonlinear systems and can better adapt to changes in the dynamic characteristics of the battery, thereby improving the accuracy of the estimation. In addition, based on the state of charge and polarization voltage, combined with the function of the model parameters changing with temperature and state of charge, the temperature of the battery cell can be accurately inferred; at the same time, using the battery equivalent circuit model, combined with the polarization voltage and external voltage, the open circuit voltage of the battery cell can be accurately obtained, and the temperature and open circuit voltage can be accurately inferred, which helps to fully understand the state of the battery cell. Furthermore, by calculating the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of the battery cells, multiple feature quantities reflecting the degree of discreteness of each battery cell and the overall battery pack are obtained. These feature quantities reflect the state of the battery cell from multiple dimensions, which helps to more comprehensively evaluate the performance of the battery cell; the feature quantities of all battery cells are unified and normalized in the correlation direction, so that the feature quantities of all battery cells are in the same dimension, which helps to eliminate the dimensional differences and numerical differences between different feature quantities and avoid large errors in subsequent calculations. Finally, the DBSCAN clustering algorithm can be used to perform cluster analysis on the feature quantities of battery cells, which can effectively identify abnormal battery cells in the battery pack; the DBSCAN algorithm performs clustering based on density, can find clusters of any shape, and can handle noise data, so it performs well in anomaly detection.

[0028] In summary, the technical solution of the embodiment of the present invention comprehensively considers multiple performance indicators such as the state of charge, temperature and open circuit voltage of the battery cell, and realizes effective screening of abnormal battery cells in the battery pack through cluster analysis, which helps to improve the overall performance and safety of the battery pack and extend the service life of the battery pack. The technical solution of the embodiment of the present invention has the advantages of high precision, multi-dimensional feature quantity analysis, standardization and unification of feature quantity processing, and abnormal battery cell identification based on DBSCAN, which can realize effective screening and performance evaluation of abnormal battery cells in the battery pack.

[0029] In one embodiment of the present invention, the step of obtaining the function of the model parameters changing with temperature and state of charge comprises: Step 1), based on the external characteristics of the battery cell, establish a battery equivalent circuit model; in a specific exemplary technical solution, the battery equivalent circuit model adopts a second-order RC model; Step 2), based on the pre-acquired open circuit voltage and current data of the battery cell, the model parameter identification is performed to obtain the model component parameters at different temperatures and charge states, that is, to obtain the function of the model parameters changing with temperature and charge state; for example, the values ​​of the controlled voltage source, internal resistance, two polarization internal resistances and two polarization capacitors at different temperatures and charge states in the second-order RC model are obtained through parameter identification; Among them, the steps of model parameter identification include: taking temperature and state of charge as input and model parameter values ​​as output, randomly dividing all data into training set and verification set according to a certain ratio; importing the training set, and using polynomial fitting to establish a function of the change of model parameters with temperature and state of charge; importing the verification set, and verifying the fitting accuracy of the change function; further explanatory, since the division of training set and verification set is random, multiple fittings are performed. If the accuracy of the result of each verification set meets the requirements, it is considered that the fitting result can be trusted; if the result of the verification set deviates far from the actual value, the number of polynomial fittings should be changed until the most suitable fitting result is found.

[0030] In the exemplary explanatory technical solution, when fitting the variation function, the training set and the validation set are randomly divided according to a certain ratio, and the overfitting or underfitting is avoided by testing the validation set multiple times to find the most suitable polynomial fitting times. In addition, after the state of charge of the battery cell is known, the temperature of the battery cell can be quickly determined by the variation function of the model parameters with temperature and state of charge, which reduces the number of temperature sensors in the battery pack, reduces costs and improves efficiency.

[0031] In one embodiment of the present invention, the battery equivalent circuit model includes: a controlled voltage source, an internal resistance and two series-connected resistor-capacitor parallel networks; wherein the controlled voltage source is used to characterize the relationship between the open circuit voltage and the state of charge of the battery, and the internal resistance is used to characterize the self-discharge phenomenon of the battery; the two series-connected resistor-capacitor parallel networks are used to characterize different dynamic polarization behaviors of the battery, and can simultaneously consider ohmic polarization, electrochemical polarization and concentration polarization, so that the model can be closer to the actual operating characteristics of the battery; The equation of the battery equivalent circuit model is expressed as:

[0032] In the formula, U t ( k )express k The external battery voltage at the moment; OCV express k The battery open circuit voltage at the moment; U p1 ( k )express k The polarization voltage of the first resistor-capacitor parallel network at time t;U p2 ( k )express k The polarization voltage of the second resistor-capacitor parallel network at time t; I ( k )express k The current flowing through the battery at all times; R o Represents the internal resistance of the model.

[0033] The battery equivalent circuit model of the embodiment of the present invention can better simulate the dynamic characteristics of the battery during the charging and discharging process through the combination of a controlled voltage source, an internal resistance and two series resistor-capacitor parallel networks. Specifically, based on the battery equivalent circuit model, the parameters of the battery at different temperatures and different states of charge are identified, and after the function of the change of model parameters with temperature and state of charge is established, the function can more accurately estimate the average temperature of the battery cell under the premise of known state of charge through the parameters of the model components.

[0034] In one embodiment of the present invention, the specific steps of obtaining the current and external voltage of each battery cell in the battery pack to be evaluated at different times are: The battery cell data of the battery pack to be evaluated are obtained at a sampling interval of 1 second through the set voltage sensor and current sensor, including the current and external voltage of the battery cells in the battery pack to be evaluated at different times; Based on the unscented Kalman filter algorithm, the state of charge and polarization voltage of the battery cell are estimated according to the acquired external voltage and current data; based on the acquired state of charge, the battery temperature is inferred when the state of charge and model parameters are known, in combination with the change function of the model parameters with temperature and state of charge, to obtain the temperature of each battery cell in the battery pack; based on the acquired polarization voltage, the open circuit voltage of the battery cell is inferred in combination with the external voltage of the battery cell and the equivalent circuit model, to obtain the open circuit voltage of each battery cell in the battery pack; Among them, when using the unscented Kalman filter algorithm for calculation, the state equation of the battery is expressed as:

[0035] When using the unscented Kalman filter algorithm, the battery measurement equation is expressed as:

[0036] In the technical solution of the embodiment of the present invention, the unscented Kalman filter algorithm is used to obtain the state of charge and polarization voltage of the battery cell, and then the temperature and open circuit voltage of the battery cell are obtained, which increases the judgment criteria for battery inconsistency. Explanatory, the unscented Kalman filter algorithm is a model-driven state of charge estimation algorithm. After establishing a battery equivalent circuit model, the state equation and measurement equation of the battery are obtained, and then the state of charge of the battery at the current moment is calculated based on the external voltage and current data, and the polarization voltage of the two resistor-capacitor networks is obtained at the same time. Among them, the temperature of the battery cell is obtained by inversely deducing the function of the model parameters changing with the temperature and state of charge after the battery state of charge is known; the open circuit voltage of the battery cell is obtained by inversely deducing the known polarization voltage combined with the external voltage and equivalent circuit of the battery cell.

[0037] In one embodiment of the present invention, after obtaining the state of charge, temperature and open circuit voltage of each battery cell, three characteristic quantities of the state of charge, temperature and open circuit voltage are calculated; the characteristic quantities of all battery cells are unified and normalized in correlation direction so that all characteristic quantities are in the same dimension; the DBSCAN clustering algorithm is used to analyze the discrete degree of the characteristic quantities, find outlier data, and then find the abnormal battery corresponding to the outlier data; Among them, the three references for measuring the degree of data dispersion, namely, mean deviation, standard score and Euclidean distance, are used as feature quantities; explanatory, mean deviation is used to describe the difference between each data point and the center point (usually the mean) in the data set. The deviation shows the specific gap between each data point and the mean, thus reflecting the degree of dispersion of the data point. The calculation formula is:

[0038] In the formula, X D,i,t yes t The battery pack is i The numerical deviation of each battery cell, X i,t yes t Moment i The value of each battery cell, X μ,t yes t The average value of all battery cells in the battery pack at the moment; In order to further reflect the discreteness of the battery cells, it is necessary to obtain the battery cells at all times during this period. n The average of the mean deviations is:

[0039] In addition, the standard score understands the position of a data point in the entire data distribution by dividing the deviation of the data point by the standard deviation of the data set. The standard score is often used to identify outliers because their position in the standard distribution is often very extreme. The calculation formula is:

[0040] In the formula, X σ,t yes t The standard deviation of the values ​​of all battery cells in the battery pack at the moment. In order to reflect the dispersion of battery cells over a long period of time, it is necessary to obtain the standard deviation of the values ​​of all battery cells at the moment of this period. n The average of the standard scores is:

[0041] Euclidean distance can be understood as the "straight-line" distance between two points in Euclidean space, that is, the most direct path distance, and the calculation formula is:

[0042] In order to reflect the discreteness of the battery cell, it is necessary to obtain the discreteness of the battery cell at all times. n The Euclidean distance is:

[0043] In summary, in the preferred solution of the embodiment of the present invention, nine characteristic quantities of three judgment criteria, namely state of charge, temperature and open circuit voltage, are selected to screen and find abnormal battery cells, which can improve the accuracy of screening.

[0044] In one embodiment of the present invention, the specific steps of performing cluster analysis using the DBSCAN clustering algorithm include: Step (1), performing correlation direction unification and normalization processing on the feature quantities of all battery cells so that all feature quantities are in the same dimension; Step (2): Select a battery cell object point h 1. The selection criterion is that it is not classified into a cluster or marked as an outlier; then, for the object point h 1 is observed in the neighborhood. If the number of other object points in the neighborhood L 1≥minpoints, then all object points in the neighborhood are added to the temporary cluster Q ; If the number of points in the neighborhood L 1<minpoints, then the object point h 1 is marked as an outlier; Step (3), for the temporary cluster Q All unselected battery object points h2, if the number of other object points in the neighborhood L 2≥minpoints, then add these points to the temporary cluster Q ; If the battery object point h 2 is not included in any cluster, so it is included in the cluster C ; Step (4), repeat step (3) and continue to check the temporary cluster Q Object points that have not been selected; Step (5), repeat steps (2) to (4) until all battery object points are classified into a cluster or marked as outliers; Step (6), confirm the battery serial number represented by the outlier point, and obtain the abnormal battery cell in the battery pack.

[0045] In the technical solution of the embodiment of the present invention, three characteristic quantities are proposed to measure the discreteness of data, and the DBSCAN clustering algorithm is used to find outliers, and finally the abnormal batteries corresponding to the outliers are obtained, and the screening results have high accuracy and reliability.

[0046] In summary, an embodiment of the present invention discloses a method for screening abnormal battery cells in a battery pack based on model parameter identification and state of charge estimation, which can be used to monitor the performance and health status of the battery. It establishes a function of how model parameters change with temperature and state of charge based on model parameter identification, estimates the state of charge of the battery based on a Kalman filter algorithm, infers the temperature of the battery based on the change function, infers the open circuit voltage of the battery based on the battery model and polarization voltage, and obtains abnormal battery cells corresponding to outliers based on a DBSCAN clustering algorithm and three characteristic quantities.

[0047] See also Figures 2 to 6 The embodiment of the present invention provides a method for screening abnormal battery cells in a battery pack based on model parameter identification and state of charge estimation, comprising the following steps: According to the external characteristics of the single battery, a second-order RC equivalent circuit model is established; Based on the second-order RC equivalent circuit model, the parameters of the battery at different temperatures and different states of charge are identified, and a function of the variation of the model parameters with temperature and state of charge is established. Then, the model parameters under other temperature and state of charge conditions are selected to verify the fitting accuracy of the function to temperature and state of charge, and the function is further adjusted according to the results to avoid underfitting or overfitting of the function, so that the function can more accurately estimate the average temperature of the battery cell under the premise of known state of charge through the model component parameters; A voltage sensor and a current sensor are set to obtain the data of the battery pack to be evaluated at a sampling interval of 1 second, including the current and external voltage of the battery cells in the resistor group to be evaluated at different times; based on the unscented Kalman filter algorithm, the state of charge and polarization voltage of the battery are estimated according to the acquired external voltage and current data, and the state of charge of each battery cell in the battery pack is obtained; Based on the acquired state of charge, the battery temperature is inferred in combination with the function of the model parameters changing with temperature and state of charge when the state of charge and the model parameters are known, and the temperature of each battery cell in the battery pack is obtained; based on the acquired polarization voltage, the open circuit voltage of the battery cell is inferred in combination with the external voltage of the battery cell and the equivalent circuit, and the open circuit voltage of each battery cell in the battery pack is obtained; Based on the obtained battery cell state of charge data, temperature data and open circuit voltage data, the average deviation, standard score and Euclidean distance of each battery cell state of charge, battery temperature and open circuit voltage are calculated to obtain three characteristic quantities reflecting the degree of discreteness of the battery cell and the overall battery pack, and the characteristic quantities of all battery cells are unified and normalized in the correlation direction so that all characteristic quantities are in the same dimension to avoid large errors in the subsequent calculation process; Based on the DBSCAN clustering algorithm, the state of charge, temperature and open circuit voltage characteristics of the battery cells are clustered to find the outliers, confirm the battery serial number represented by the outliers, and obtain the abnormal battery cells in the battery pack.

[0048] See also Figure 3 , Figure 3 The battery equivalent circuit model of the embodiment of the present invention is shown, and the equation of the battery equivalent circuit model is:

[0049] Therefore, after obtaining the state of charge and polarization voltage of the battery, the open circuit voltage of the battery cell is:

[0050] In the formula, OCV is the open circuit voltage of the battery cell.

[0051] See also Figure 4 , Figure 4 It shows that the model parameters vary with temperature and state of charge. After model parameter identification, the following is obtained: Figure 4 The surface shown in FIG. 1 can be used to obtain the battery temperature by looking up a table after the model parameters and the state of charge are known; at the same time, the model has multiple components, and the temperatures obtained by the change functions of different components can verify each other, thereby obtaining an accurate battery temperature.

[0052] See also Figure 5 , Figure 5The process of establishing the function of model parameters changing with temperature and state of charge is shown. Temperature and state of charge are used as inputs, and model parameter values ​​are used as outputs. All values ​​are randomly divided into training set and validation set according to a certain ratio. Generally, the proportion of training set is 0.8, and the proportion of validation set is 0.2. The training set is imported, and the function of model parameters changing with temperature and state of charge is established by polynomial fitting. The validation set is imported to verify the fitting accuracy of the function. It is considered that R 2 When it is greater than or equal to 0.8, a higher fitting accuracy is achieved, but attention should be paid to whether overfitting or underfitting occurs; since the division of the training set and the validation set is random, multiple fittings are performed; in the embodiment of the present invention, the number of times is set to 10 times, and if the accuracy of the result of each validation set of 10 fittings meets the requirements, it is considered that the fitting result can be trusted; if the result of the validation set deviates far from the actual value, the number of polynomial fittings should be changed until the most suitable fitting result is found.

[0053] See also Figure 6 , Figure 6 The process of screening abnormal battery cells based on feature quantities and DBSCAN clustering algorithm is shown, including: first, three feature quantities need to be calculated, taking the state of charge as an example,

[0054] In the formula, SOC D,i,t yes t The battery pack is i The charge state value deviation of each battery, SOC i,t yes t Moment i The state of charge value of each battery, SOC μ,t yes t The average state of charge of all battery cells in the battery pack at this moment; In order to further reflect the discreteness of the battery cell, it is necessary to obtain the battery cell at all times during this period n The average of the mean deviations is:

[0055] The standard score understands the position of a data point in the entire data distribution by dividing the deviation of the data point by the standard deviation of the data set. The standard score is often used to identify outliers because their position in the standard distribution is often very extreme. It is applied to state of charge, temperature and open circuit voltage. The calculation formula is:

[0056] Taking the state of charge as an example, where SOC σ,t yest The standard deviation of the state of charge of all battery cells in this battery pack at this moment. In order to reflect the discrete degree of battery cells over a long period of time, it is necessary to obtain the state of charge of all battery cells at this time. n The average of the standard scores is:

[0057] The Euclidean distance can be understood as the "straight line" distance between two points in Euclidean space, that is, the most direct path distance, which is applied to the state of charge, temperature and open circuit voltage, and the calculation formula is:

[0058] In order to reflect the discreteness of the battery cell, it is necessary to obtain the discreteness of the battery cell at all times. n The Euclidean distance is:

[0059] After obtaining the feature quantity, the feature quantities of all battery cells are unified and normalized in terms of correlation direction, so that all feature quantities are in the same dimension; finally, the DBSCAN algorithm is used to find outliers. If there are no outliers, there are no abnormal battery cells. If there are outliers, the corresponding abnormal battery is found according to the outlier battery number, and the abnormal feature quantity and the collection point sequence number when it appears are output to determine the abnormal type and abnormal time of the battery.

[0060] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0061] See also Figure 7 In an embodiment of the present invention, a system for screening abnormal battery cells in a battery pack is provided, comprising: A data acquisition module is used to obtain the current and external voltage of each battery cell in the battery pack to be evaluated at different times; A first calculation module, used to obtain the state of charge and polarization voltage of each battery cell based on the current and external voltage of each battery cell at different times; A second calculation module is used to obtain the temperature of each battery cell in the battery pack by inversely calculating the change function of the model parameters with temperature and the state of charge based on the acquired state of charge of each battery cell when the state of charge and the model parameters are known; A third calculation module is used to obtain the open circuit voltage of each battery cell in the battery pack based on the obtained polarization voltage of each battery cell, combined with the external voltage of the battery cell and the battery equivalent circuit model; A fourth calculation module is used to calculate the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell based on the obtained state of charge, temperature and open circuit voltage of each battery cell, and obtain a characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack; The analysis and screening module is used to perform cluster analysis based on the obtained characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack, and obtain abnormal battery cells in the battery pack to be evaluated.

[0062] In one embodiment of the present invention, a computer device is provided, the computer device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used to perform the operation of the abnormal battery cell screening method in the battery pack.

[0063] In one embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM (Random Access Memory) memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the abnormal battery cell screening method in the battery pack in the above embodiment.

[0064] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program codes.

[0065] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0066] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0067] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for screening abnormal battery cells in a battery pack, characterized in that: The following steps are involved: Obtain the current and external voltage of each battery cell in the battery pack to be evaluated at different times; Based on the current and external voltage of each battery cell at different times, the state of charge and polarization voltage of each battery cell are obtained; Based on the acquired state of charge of each battery cell, the temperature of each battery cell in the battery pack is obtained by reversely calculating the function of the model parameters changing with temperature and state of charge when the state of charge and model parameters are known; Based on the polarization voltage of each battery cell obtained, the open circuit voltage of each battery cell in the battery pack is obtained by combining the external voltage of the battery cell with the battery equivalent circuit model; Based on the obtained state of charge, temperature and open circuit voltage of each battery cell, the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell are calculated respectively to obtain characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack; Based on the obtained characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack, a cluster analysis is performed to obtain abnormal battery cells in the battery pack to be evaluated.

2. A method for screening abnormal battery cells in a battery pack according to claim 1, characterized in that: The specific steps of obtaining the state of charge and polarization voltage of each battery cell based on the current and external voltage of each battery cell at different times include: Based on the current and external voltage of each battery cell at different times, the unscented Kalman filter algorithm is used to estimate the state of charge and polarization voltage of each battery cell.

3. A method for screening abnormal battery cells in a battery pack according to claim 1, characterized in that: The step of obtaining the function of the model parameters changing with temperature and state of charge comprises: Based on the external characteristics of the battery cell, a battery equivalent circuit model is established; Perform model parameter identification on the battery equivalent circuit model, obtain model component parameters at different temperatures and charge states, and obtain the function of model parameters changing with temperature and charge state; Among them, when identifying the model parameters of the battery equivalent circuit model: temperature and state of charge are used as input, and the model parameter values ​​are used as output; all pre-acquired sample data are randomly divided into training set and verification set according to a certain ratio, the training set is imported and polynomial fitting is used to establish the function of the model parameters changing with temperature and state of charge, the verification set is imported and the fitting accuracy of the change function is verified.

4. A method for screening abnormal battery cells in a battery pack according to claim 3, characterized in that: The battery equivalent circuit model includes: a controlled voltage source, an internal resistance and two series-connected resistor-capacitor parallel networks; wherein, The controlled voltage source is used to characterize the relationship between the open circuit voltage and the state of charge of the battery, the internal resistance is used to characterize the self-discharge phenomenon of the battery, and the two series-connected resistor-capacitor parallel networks are used to characterize different dynamic polarization behaviors of the battery; The equation of the battery equivalent circuit model is expressed as: In the formula, U t ( k )express k The external battery voltage at the moment; OCV express k The battery open circuit voltage at the moment; U p1 ( k )express k The polarization voltage of the first resistor-capacitor parallel network at time t; U p2 ( k )express k The polarization voltage of the second resistor-capacitor parallel network at time t; I ( k )express k The current flowing through the battery at all times; R o Represents the internal resistance of the model.

5. The method for screening abnormal battery cells in a battery pack according to claim 1, characterized in that: In the step of respectively calculating the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell based on the obtained state of charge, temperature and open circuit voltage of each battery cell to obtain a characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack, The calculation expression of the average deviation is: In the formula, X D,i The battery pack i The average deviation of the battery cell values; X D,i,t yes t The battery pack is i The deviation of the battery cell value, n represents the total number at the moment; X i,t yes t Moment i A battery cell value, which is the state of charge, temperature or open circuit voltage of the battery cell; X μ,t yes t The average value of all battery cells in the battery pack at the moment; The calculation expression of the standard score is: ； In the formula, X Z,i The battery pack i Standard fraction of each battery cell value; X σ,t yes t The standard deviation of the values ​​of all battery cells in the battery pack at the moment; X Z,i,t yes t The battery pack is i Standard fraction of each battery cell value; The calculation expression of Euclidean distance is: In the formula, X d,i The battery pack i The Euclidean distance of the battery cell values; X d,i,t yes t The battery pack is i The Euclidean distance between the values ​​of the individual cells.

6. A method for screening abnormal battery cells in a battery pack according to claim 1, characterized in that: In the step of performing cluster analysis based on the obtained characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack to obtain abnormal battery cells in the battery pack to be evaluated, before performing cluster analysis, the characteristic quantities of all battery cells are unified and normalized in terms of correlation direction so that the characteristic quantities of all battery cells are in the same dimension.

7. A method for screening abnormal battery cells in a battery pack according to claim 1, characterized in that: In the step of performing cluster analysis based on the obtained characteristic quantity reflecting the degree of discreteness of each battery cell and the entire battery pack to obtain abnormal battery cells in the battery pack to be evaluated, the DBSCAN clustering algorithm is used to perform cluster analysis.

8. A system for screening abnormal battery cells in a battery pack, characterized in that: include: A data acquisition module is used to obtain the current and external voltage of each battery cell in the battery pack to be evaluated at different times; A first calculation module, used to obtain the state of charge and polarization voltage of each battery cell based on the current and external voltage of each battery cell at different times; A second calculation module is used to obtain the temperature of each battery cell in the battery pack by inversely calculating the change function of the model parameters with temperature and the state of charge based on the acquired state of charge of each battery cell when the state of charge and the model parameters are known; A third calculation module is used to obtain the open circuit voltage of each battery cell in the battery pack based on the obtained polarization voltage of each battery cell, combined with the external voltage of the battery cell and the battery equivalent circuit model; A fourth calculation module is used to calculate the average deviation, standard score and Euclidean distance of the state of charge, temperature and open circuit voltage of each battery cell based on the obtained state of charge, temperature and open circuit voltage of each battery cell, and obtain a characteristic quantity reflecting the degree of dispersion of each battery cell and the overall battery pack; The analysis and screening module is used to perform cluster analysis based on the obtained characteristic quantities reflecting the degree of discreteness of each battery cell and the overall battery pack, and obtain abnormal battery cells in the battery pack to be evaluated.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for screening abnormal battery cells in a battery pack according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for screening abnormal battery cells in a battery pack according to any one of claims 1 to 7 is implemented.

Images