Method for predicting health state of lithium ion battery for logistics lease

Abstract

The invention provides a method for predicting the health state of a lithium ion battery for logistics leasing. The method comprises the following steps: firstly, extracting performance parameters of each group of batteries in each charging and discharging period, and calculating a maximum value, a minimum value and an average value of the performance parameters as statistical characteristics; and clustering the statistical characteristics of all the battery charging and discharging sections to obtain a specified number of clusters, generating a reference sequence representing a common mode of each cluster, and calculating a dynamic time warping distance between the reference sequence and the standardized charging and discharging section characteristics as a characteristic vector of each charging and discharging period of each group of batteries. And then extracting the residual electric quantity of each group of batteries at the end of each charging and discharging period, evaluating the charging and discharging fullness degree through probability density distribution, and dividing the batteries into three groups of full charging and discharging, medium charging and discharging degree and insufficient charging and discharging. The feature vectors are used for predicting the health states of the batteries of different groups, the prediction accuracy is improved, and the method is suitable for logistics leasing of the lithium ion batteries and other scenes with obvious charge and discharge feature changes.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium-ion batteries, and in particular to a method for predicting the health status of lithium-ion batteries by extracting battery charge and discharge characteristics and grouping the sufficiency level in scenarios where the charge and discharge differences are significant, such as lithium-ion batteries for rental of logistics electric vehicles. Background Art

[0002] Lithium-ion batteries are widely used in modern electronic devices and electric vehicles, and their charge and discharge characteristics directly affect the performance and life of the battery. However, the performance prediction of lithium-ion batteries is a complex problem, mainly because the charge and discharge behaviors of lithium-ion batteries under different usage conditions vary significantly, especially in the complex and changeable working scenarios of logistics electric vehicle rental batteries, the differences will be more obvious.

[0003] Existing battery life prediction methods usually rely on single feature data or overall data for analysis, such as using only the actual values ​​of current and voltage, or directly using all monitoring data for prediction without feature selection. Such methods are often unable to accurately reflect the health status of the battery when dealing with complex and changeable usage conditions, resulting in predictions in actual scenarios that often fail to achieve the accuracy of laboratory test data.

[0004] With the widespread application of electric vehicles and energy storage systems in logistics scenarios, the usage frequency, load, environmental conditions, etc. of leased lithium-ion batteries have become increasingly complex and changeable. Therefore, there is an urgent need for an extraction method that can accurately reflect the charge and discharge characteristics of lithium-ion batteries under complex usage conditions, and to group lithium-ion batteries according to the degree of charge and discharge sufficiency, so as to improve the accuracy and reliability of the performance and life prediction of lithium-ion batteries used randomly under complex working conditions. Summary of the invention

[0005] The purpose of the present invention is to provide a method for predicting the health status of lithium-ion batteries for logistics electric vehicle rental based on charge and discharge feature extraction and sufficiency grouping, so as to solve the problem of low accuracy in predicting the performance and life of lithium-ion batteries in actual complex application scenarios in the prior art.

[0006] To achieve the above object, the present invention provides a method for predicting the health status of a lithium-ion battery for logistics leasing, comprising the following steps: S1: extracting the performance parameters of each lithium-ion battery in each charge and discharge cycle, including current, voltage and temperature, and calculating the maximum, minimum and average values ​​of the current, voltage and temperature as statistical features, clustering the statistical features of all batteries in each charge and discharge segment, obtaining a specified number of clusters after clustering, and generating a reference sequence that can represent the common mode of each cluster; S2: for each charge and discharge cycle of each battery, calculate the minimum cumulative dynamic time warping distance between the standardized feature sequence of the battery in the charge and discharge cycle and the reference sequence of each cluster, and convert the minimum cumulative dynamic time warping distance into a vector as the final dynamic time warping feature vector of the battery in the charge and discharge cycle; S3: extract the remaining power of each battery at the end of each charge and discharge cycle for probability statistics, extract the probability density distribution peak of the probability statistics, obtain the charge and discharge sufficiency of the battery, and group the batteries according to the charge and discharge sufficiency of each battery; S4: Using the dynamic time warped sequence feature vectors of the charge and discharge cycles of each group of lithium-ion batteries as features, the machine learning extreme gradient boosting tree algorithm is used to predict the health status of lithium-ion batteries in different charge and discharge degree groups.

[0007] Preferably, S1 includes: S11: Obtaining an original feature sequence, wherein the original feature sequence of the jth charging segment of the i-th battery is in the following format, and the jth charging segment is a charging segment in the jth charge-discharge cycle: ,in is the maximum current of the jth charging segment, is the minimum current value of the jth charging segment, is the average current of the jth charging segment, is the maximum voltage of the jth charging segment, is the minimum voltage of the jth charging segment, is the average voltage of the jth charging segment, is the maximum temperature of the jth charging segment, is the minimum temperature of the jth charging segment, is the average temperature of the jth charging segment.

[0008] Preferably, S2 includes: S21: The original feature sequence of each battery Perform feature standardization to obtain a standardized feature sequence; S22: according to each standardized feature sequence, principal component analysis is used for dimensionality reduction, and K-means clustering is performed on the standardized feature sequences of all battery charging segments and discharging segments after dimensionality reduction, to generate a specified number of charging clusters and discharging clusters; S23: for the generated charging clusters and discharging clusters, a reference sequence is generated for each cluster using a time series averaging algorithm based on dynamic time warping; S24: calculating the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster, vectorizing the dynamic time warping distance, and generating a final dynamic time warping feature vector; S25: Obtain the dynamic time warping feature vectors of all batteries in each charge and discharge cycle. For a certain charge and discharge cycle c, calculate the minimum cumulative dynamic time warping distance between the standardized charge feature sequence and the reference sequence of each charge cluster in the charge segment, and calculate the dynamic time warping distance between the standardized discharge feature sequence and the reference sequence of each discharge cluster in the discharge segment. Concatenate the calculated dynamic time warping distances to obtain the final dynamic time warping feature vector of the charge and discharge cycle. The form is: ,in, is the charging stage characteristic, It is the characteristic of the discharge stage.

[0009] Preferably, in S21, the standardized feature sequence comprises the following steps: For each battery, the global mean of the battery is calculated based on each feature of the battery's charge and discharge stages. and standard deviation , k ranges from 1 to 9; in, is the number of all charging or discharging stages of the battery, is the value of the kth feature in the lth charging or discharging segment; After obtaining the global mean and the standard deviation of the battery, each feature of the battery is standardized by the following formula: in The standardized value of the kth feature in the lth charging or discharging segment generates the standardized feature sequence of the battery , .

[0010] Preferably, in S23, generating a reference sequence comprises the following steps: For each cluster, an original feature sequence in the charging or discharging cycle of a lithium-ion battery in the cluster is randomly selected as the current reference sequence. For the original feature sequences of other batteries, dynamic time warping is used to calculate the distance between the original feature sequences of other batteries and the current reference sequence, and the optimal alignment path is found. The average value at each position after alignment is calculated, and a new reference sequence is generated to replace the current reference sequence. This step is repeated until the difference between the current reference sequence and the previous reference sequence is less than a set threshold, which is considered to converge, and a reference sequence generated for the cluster is obtained.

[0011] Preferably, in S24, the following steps are included: The dynamic time warping distance calculation formula is: , in, Represents the standardized charge and discharge characteristic sequence The reference sequence of each cluster The dynamic time warping distance between them; for multidimensional feature vectors and , the distance is defined as: The dynamic time warping distance matrix D is at position Value The calculation of is defined as: At a specific point in time Under this condition, the dynamic time warping distance is , the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster can be calculated. By starting from the first position (1,1) of the matrix, each position in the first row It is accumulated from left to right, and each position in the first column It is accumulated from top to bottom, accumulating the distance of each step in turn until the calculation reaches the lower right corner of the matrix D(P, Q). D(P, Q) represents the minimum dynamic time warping distance of the two sequences. It is calculated from the first position of the matrix and P and Q are the lengths of the two sequences respectively. In this way, a complete dynamic time warping distance matrix can be obtained, in which the element in the lower right corner of the matrix is ​​the required minimum dynamic time warping distance. Therefore, the value of each position is the minimum cumulative distance of the three positions above, left or diagonal.

[0012] Preferably, in S3, the charge and discharge sufficiency grouping method comprises the following sub-steps: S31: Calculate the probability density distribution function of the remaining power of each battery according to the remaining power at the end of the charging and discharging stages in each charging and discharging cycle of each battery; S32: according to the peak position of the battery's remaining power probability density function, obtain the remaining power value corresponding to the peak value of the remaining power probability density distribution, and determine the grouping threshold of the remaining power based on the remaining power value corresponding to the peak value of the remaining power probability density distribution and the healthy charging and discharging experience, the charging and discharging thresholds are 50% and 80% and 40% and 50% respectively; S33: For the charging stage, if the peak value of the probability density distribution of the remaining power is above 80% of the full charge, it is considered to be fully charged; if the peak value of the probability density distribution of the remaining power is at a medium charge, i.e. 50%-80% of the full charge, it is considered to be moderately charged; if the peak value of the probability density distribution of the remaining power is below 50% of the full charge, it is considered to be insufficiently charged; S34: For the discharge stage, if the peak value of the probability density distribution of the remaining power is above 50% of the full rating, the discharge is considered to be insufficient; if the peak value of the probability density distribution of the remaining power is at a medium level, i.e., 40%-50% of the full rating, the discharge is considered to be medium sufficient; if the peak value of the probability density distribution of the remaining power is below 40% of the full rating, the discharge is considered to be sufficient; S35: According to the comprehensive situation of charge and discharge classification, the battery is divided into three groups, namely, fully charged and discharged, medium charge and discharge, and insufficient charge and discharge.

[0013] The present invention also provides a health status prediction system for lithium-ion batteries for logistics leasing, which executes the above-mentioned health status prediction method for lithium-ion batteries for logistics leasing, and specifically includes a capacity calculation module, a data processing module, a feature extraction module based on charge and discharge, a grouping module based on charge and discharge sufficiency, and a lithium-ion battery grouping prediction module; The capacity calculation module is used to calculate the specific capacity of the charge and discharge segment as a representation of the life attenuation of the lithium-ion battery; the data processing module is used to process the required features (such as current, voltage, temperature, etc.) and the target (capacity) according to certain rules. The feature extraction module based on charge and discharge is responsible for extracting the available features that can be used for the grouping module and prediction module based on the charge and discharge sufficiency through machine learning algorithms; the grouping module based on the charge and discharge sufficiency groups the batteries according to the extracted remaining power of the charge and discharge segment, and divides the batteries into different groups for the grouping prediction module; the lithium-ion battery grouping prediction module. Finally, the remaining capacity and health status of the lithium-ion battery are predicted according to different groups, and accurate prediction results can be obtained for each group.

[0014] Furthermore, the specific process of the capacity calculation module includes: First, calculate the capacity change in the charge / discharge stage. The specific calculation formula is: ,in, is the capacity change, and are the initial and end time points of the capacity segment to be calculated, respectively. For current.

[0015] According to the initial capacity and capacity change, the capacity value of the current state can be obtained. The specific calculation formula is: , C is the current capacity, The capacity at the end of the previous cycle.

[0016] Furthermore, the specific process of the data extraction module includes: Extract the performance parameters of each lithium-ion battery in each charge and discharge cycle, including current, voltage and temperature, and calculate the maximum value, minimum value and average value of the current, voltage and temperature as statistical features. Furthermore, the specific steps of the feature extraction module include: First, based on the existing data, the charging and discharging segments are separated to obtain the data of all charging and discharging segments. For each charging and discharging segment, a set of consistent feature sequences is obtained by calculating statistical features. The features are expressed as ,in is the maximum current of the jth charging segment, is the minimum current value of the jth charging segment, is the average current of the jth charging segment, is the maximum voltage of the jth charging segment, is the minimum voltage of the jth charging segment, is the average voltage of the jth charging segment, is the maximum temperature of the jth charging segment, is the minimum temperature of the jth charging segment, is the average temperature of the jth charging segment.

[0017] Next, in order to eliminate the influence of different feature scales and ensure that each feature contributes equally to the dynamic time warping (DTW) distance calculation, the original feature sequence of each battery is Perform feature standardization to obtain a standardized feature sequence. To standardize the original features, first calculate the global mean of each feature of the battery charging and discharging segment in all usage cycles. and standard deviation , the specific calculation method is shown in the formula: , .in, is the number of all charging and discharging stages of the battery, is the value of the kth feature in the lth charge and discharge segment. After obtaining the mean and standard deviation, each feature is standardized using the following formula: ,in is the normalized value of the kth feature in the lth charging or discharging segment, is the original feature value, generating a standardized feature sequence , .

[0018] Subsequently, in order to set the DTW reference sequence for each battery, the statistical features of all battery charge and discharge segments are clustered to obtain a specified number of clusters, each of which represents a similar charging and discharging behavior pattern of lithium-ion batteries. For the results of the generated cluster clusters, a time series averaging algorithm based on dynamic time warping is used to generate a reference sequence for each cluster as the reference sequence for subsequent DTW matching. The time series averaging algorithm based on dynamic time warping (DTWBarycenter Averaging, DBA) is a method for generating the center of a time series, which makes the sequence in the cluster closest to its center through step-by-step iteration. Specifically, DBA can generate a representative center sequence that can maximize the capture of the common features of all time series in the cluster. For each cluster cluster, the DBA algorithm is applied to calculate the reference sequence of the cluster. Specifically, an original feature sequence in a lithium-ion battery charging or discharging cycle in the current cluster is randomly selected as the initial reference sequence. R 1 = [r1, r2, …, r9], for each battery charge and discharge sequence, use dynamic time warping to calculate its distance from the current original reference sequence, find the optimal alignment path, calculate the average value at each position after alignment, generate a new reference sequence, repeat this step until the difference between the current reference sequence and the previous reference sequence is less than a certain set threshold, and it is considered converged, and the final reference sequence corresponding to the original eigenvalue of the normalized charge and discharge cycle of the lithium-ion battery in each cluster is obtained R q。 The specific formula is ,in, is the sequence of charge and discharge segments in the cluster , is the reference sequence generated by the i-th cluster.

[0019] The dynamic time warping distance calculation formula is: in, Represents the standardized charge and discharge characteristic sequence The reference sequence of each cluster The dynamic time warping distance between them; for multidimensional feature vectors and , the distance is defined as: The dynamic time warping distance matrix D is at position Value The calculation of is defined as: At a specific point in time Under this condition, the dynamic time warping distance is , the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster can be calculated. By starting from the first position (1,1) of the matrix, each position in the first row It is accumulated from left to right, and each position in the first column It is accumulated from top to bottom, and the distance of each step is accumulated in turn until the lower right corner of the matrix D(P, Q) is calculated. D(P, Q) represents the minimum dynamic time warping distance of the two sequences. Starting from the first position of the matrix, P and Q are the lengths of the two sequences respectively. From this, a complete dynamic time warping distance matrix can be obtained, in which the element in the lower right corner of the matrix is ​​the required minimum dynamic time warping distance. Therefore, the value of each position is obtained by selecting the minimum cumulative distance of the three positions above, left or diagonal. And the minimum cumulative distance is converted into a vector to generate the final DTW feature vector, which contains the charge and discharge characteristics within the cycle.

[0020] For a certain charge and discharge cycle c, the final eigenvector The form is ,in, is the charging stage characteristic, It is the characteristic of the discharge stage.

[0021] Furthermore, the specific steps of grouping modules based on the remaining power are as follows: According to the probability density distribution of the remaining power, the charging and discharging process is divided into different groups; first, the probability density function of the remaining power of each battery is calculated. The specific calculation formula is: ,in is the estimated density function, expressed in The probability density at ; is the number of samples, i.e., the total number of measurements of remaining power in the data set; is each sample data point, i.e., each measured remaining power value in the data set; is the bandwidth (smoothing parameter), which controls the smoothness of the kernel density estimate. The larger the bandwidth, the smoother the estimated curve; It is a kernel function. This design uses the Gaussian kernel function. The main function of the Gaussian kernel function is to assign a weighted value to each data point, and the size of the weight is determined by the distance from the point to the target location. The closer the distance, the greater the weight; the farther the distance, the smaller the weight. The specific calculation formula is: ,in is the value of the Gaussian kernel function; Indicates the target location With sample data points The standardized distance between represents the exponential function; The remaining power grouping threshold is determined comprehensively based on the probability density function of the remaining power of the battery and the experience of healthy charging and discharging. The grouping threshold of the remaining power in the charging stage is set to 50% and 80%, and the grouping threshold of the remaining power in the discharging stage is set to 40% and 50%. For the charging stage, if the peak value of the probability density distribution of the remaining power is above 80% of the full value, it is considered to be fully charged; if the peak value of the remaining power distribution is at the medium power (50%-80% of the full value), it is considered to be moderately charged; if the peak value of the remaining power distribution is below 50% of the full value, it is considered to be insufficiently charged; for the discharging stage, if the peak value of the probability density distribution of the remaining power is above 50% of the full value, it is considered to be insufficiently discharged; if the peak value of the remaining power distribution is at the medium power (40%-50% of the full value), it is considered to be moderately discharged; if the peak value of the remaining power distribution is below 40% of the full value, it is considered to be fully discharged; According to the comprehensive situation of charge and discharge classification, the battery is divided into three groups as a whole, namely, fully charged and discharged, medium charge and discharge, and insufficient charge and discharge. They respectively represent the charge and discharge saturation of the battery, from high to low, that is, fully charged and discharged means that the battery is fully charged and discharged, the medium charge and discharge group means that one of the charge or discharge is insufficient, and the insufficient charge and discharge group means that the charge and discharge are insufficient, or even poor; Furthermore, for the group prediction module, it specifically includes the following contents: For an ungrouped battery, we first analyze it to determine which group it belongs to. Then, we use machine learning algorithms such as Random Forest (RF), XGB, and Long Short-Term Memory (LSTM) to predict the battery life of the three groups of data, and train a most suitable model for each group. Then, we use the model of this group to train the battery.

[0022] The above method can effectively improve the accuracy and robustness of battery life and performance prediction. Experimental results show that the method of the present invention significantly improves the accuracy of the prediction model, thereby extending the battery life and improving safety.

[0023] The method and system of the present invention not only improve the accuracy of battery life and performance prediction, but also enhance the robustness of the model, and are suitable for complex and changeable charging and discharging conditions such as logistics scenarios.

[0024] In summary, the present invention proposes a new battery performance and life prediction method by extracting the charging and discharging characteristics of lithium-ion batteries and grouping the charging and discharging degrees, which effectively solves the prediction accuracy and robustness problems in the prior art and has broad application prospects and significant practical significance.

[0025] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes a novel feature extraction method that takes into account both dynamic time series features and automatic data pattern extraction. This method combines the dynamic matching capability of DTW and the pattern recognition capability of clustering methods, and can extract representative fixed-length feature vectors from a large number of variable-length sequences. This feature extraction method can effectively process large-scale variable-length sequence data and avoid information loss caused by interpolation or clipping. The time series averaging algorithm (DBA) that generates representative sequences using DTW clustering provides an objective, data-driven feature extraction solution, providing a new idea for extracting features for predicting battery health status in actual scenarios.

[0026] 2. When the present invention uses deep learning to predict the life of lithium-ion batteries, the probability density distribution statistics of the remaining power of the battery at the end of each charge and discharge cycle are performed. According to the degree of charge and discharge, the batteries are divided into three groups: fully charged and discharged, medium charge and discharge, and insufficient charge and discharge, which helps to classify the attenuation of the batteries, so that batteries with the same attenuation can be trained together, thereby improving the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0028] Figure 1 This is a flow chart of a life prediction method based on charge and discharge feature grouping proposed by the present invention.

[0029] Figure 2 This is a flow chart of DTW feature extraction in Example 1.

[0030] Figure 3 This is a probability density distribution diagram of the remaining power of 30 battery charging and discharging stages in Example 1.

[0031] Figure 4 This is the capacity attenuation diagram of 30 groups of batteries in Example 1.

[0032] Figure 5 Schematic diagram of battery grouping strategy.

[0033] Figure 6 This is a comparison chart of the effects of using DTW feature life prediction and other feature prediction in Example 1.

[0034] Figure 7 This is a comparison chart of the life prediction of grouped batteries and the overall prediction of ungrouped batteries in Example 1. Specific implementation methods

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Embodiment 1: The present invention provides a grouping method based on the charging and discharging characteristics of lithium-ion batteries, such as Figure 1 As shown, the following steps are included: Lithium-ion batteries are widely used as delivery vehicle batteries in actual logistics scenarios. The analysis of battery status and predicted battery life in logistics scenarios can provide a basis for battery management and has strong economic value. This example uses a battery management system (BMS) to obtain characteristic data at the end of each charge and discharge cycle of the battery, including remaining power, voltage, current, temperature, etc., for 30 lithium-ion batteries in actual logistics scenarios.

[0037] The specific steps of feature generation are as follows: Figure 2 First, based on the existing data, the charging and discharging segments are separated to obtain the data of all charging and discharging segments. For each charging and discharging segment, the performance parameters in the charging and discharging cycle, including current, voltage and temperature, are extracted, and the maximum, minimum and average values ​​of the current, voltage and temperature are calculated as statistical features to obtain a set of consistent feature sequences, which are expressed as ,in is the maximum current of the jth charging segment, is the minimum current value of the jth charging segment, is the average current of the jth charging segment, is the maximum voltage of the jth charging segment, is the minimum voltage of the jth charging segment, is the average voltage of the jth charging segment, is the maximum temperature of the jth charging segment, is the minimum temperature of the jth charging segment, is the average temperature of the jth charging segment. These feature arrays cover the key indicators in the charging and discharging process, ensuring that the feature vector dimensions of each charging and discharging segment are consistent.

[0038] Next, in order to eliminate the influence of different feature scales and ensure that each feature contributes equally to the DTW distance calculation, the features are standardized. To standardize the original features, first calculate the global mean of each feature in the battery charging and discharging segment of all usage cycles. and standard deviation , the specific calculation method is shown in the formula: , .in, is the number of all charging and discharging stages of the battery, is the value of the kth feature in the lth charge and discharge segment. After obtaining the mean and standard deviation, each feature is standardized using the following formula: ,in is the normalized value of the kth feature in the lth charging or discharging segment, is the original feature value, generating a standardized feature sequence , .

[0039] Subsequently, in order to set the DTW reference sequence, the statistical features of all battery charging and discharging segments are clustered to generate a specified number of clusters. In this example, the number of charging and discharging clusters is set to 5, and a reference sequence that can represent the common pattern of each cluster is generated as the reference sequence for subsequent DTW matching. DBA is a method for generating the center of a time series by gradually iterating to make the sequence in the cluster closest to its center. Specifically, DBA can generate a representative center sequence that can maximize the capture of the common features of all time series in the cluster. For each cluster cluster, the DBA algorithm is applied to calculate the reference sequence of the cluster. Specifically, a sequence is first randomly selected as the initial reference sequence. R 1 = [r1, r2, …, r9], for each input time series, use DTW to calculate its distance from the current reference sequence, find the optimal alignment path, calculate the average value at each position after alignment, generate a new reference sequence, and repeat the steps until the reference sequence converges. Thus, the final reference sequence of each cluster is obtained. The specific formula is ,in, is the sequence of charge and discharge segments in the cluster , is the reference sequence generated by the i-th cluster.

[0040] Calculate the DTW distance between the original feature vector and the reference sequence, specifically: ,in, Represents the standardized charge and discharge sequence (including the characteristics of the charge and discharge stages) and the final reference sequence The DTW distance between them.

[0041] For multidimensional feature vectors and , the distance is defined as: . Cumulative distance matrix The calculation of is defined as: At a specific point in time Under this condition, the dynamic time warping distance is , the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster can be calculated. By starting from the first position (1,1) of the matrix, each position in the first row It is accumulated from left to right, and each position in the first column It is accumulated from top to bottom, accumulating the distance of each step in turn until the lower right corner of the matrix D(P, Q) is calculated. D(P, Q) represents the minimum dynamic time warping distance of the two sequences. Starting from the first position of the matrix and calculating backward continuously, P and Q are the lengths of the two sequences respectively. In this way, a complete dynamic time warping distance matrix can be obtained, in which the element in the lower right corner of the matrix is ​​the required minimum dynamic time warping distance. Therefore, the value of each position is generated by selecting the minimum cumulative distance of the three positions above, left or diagonal, and vectorizing it to generate the final DTW feature vector, which contains the charge and discharge characteristics within the cycle.

[0042] For a certain charge and discharge cycle c, the final eigenvector The form is ,in, is the charging stage characteristic, It is the characteristic of the discharge stage.

[0043] Therefore, a feature vector of length [1, 2q] is obtained for each charge and discharge cycle. Specifically, in this example, a feature vector of [1, 10] is obtained, which is used as the feature of each charge and discharge cycle to train the model.

[0044] According to the probability density distribution of the remaining power, the charging and discharging process is divided into different groups. First, the probability density function is calculated. The probability density function of the remaining power is calculated using the software. The specific calculation formula of the probability density is: ,in is the estimated density function, expressed in The probability density at ; is the number of samples, i.e., the total number of measurements of remaining power in the data set; is each sample data point, i.e., each measured remaining power value in the data set; is the bandwidth (smoothing parameter), which controls the smoothness of the kernel density estimate. The larger the bandwidth, the smoother the estimated curve; It is a kernel function. This design uses the Gaussian kernel function. The main function of the Gaussian kernel function is to assign a weighted value to each data point, and the size of the weight is determined by the distance from the point to the target location. The closer the distance, the greater the weight; the farther the distance, the smaller the weight. The Gaussian kernel function helps to smoothly estimate the probability density, especially when the data points are sparse or noisy. Kernel density estimation is a non-parametric method, which means that it does not assume that the data follows a specific distribution (such as a normal distribution). The Gaussian kernel function is only used to smoothly estimate the distribution of the data, rather than assuming the true distribution of the data. The specific calculation formula of the Gaussian kernel function is: ,in is the value of the Gaussian kernel function; Indicates the target location With sample data points The standardized distance between represents an exponential function; the probability density distribution diagram of 30 groups of batteries can be obtained, such as Figure 3 shown.

[0045] The remaining power grouping threshold is determined comprehensively based on the probability density function of the remaining power of the battery and the experience of healthy charging and discharging. The grouping threshold of the remaining power in the charging stage is set to 50% and 80%, and the grouping threshold of the remaining power in the discharging stage is set to 40% and 50%. The specific logic of the overall grouping is as follows Figure 5 As shown, for the charging stage, if the peak value of the probability density distribution of the remaining power is above 80% of the full amount, it is considered that the charge is fully charged; if the peak value of the remaining power distribution is at the medium power (50%-80% of the full amount), the charging is considered to be moderately sufficient; if the peak value of the remaining power distribution is below 50% of the full amount, it is considered that the charge is insufficient; for the discharging stage, if the peak value of the probability density distribution of the remaining power is above 50% of the full amount, it is considered that the discharge is insufficient; if the peak value of the probability density distribution of the remaining power is at the medium power (40%-50% of the full amount), it is considered that the discharge is moderately sufficient; if the peak value of the remaining power distribution is below 40% of the full amount, it is considered that the discharge is sufficient; According to the comprehensive situation of charge and discharge classification, the battery is divided into three groups, namely, full charge and discharge, medium charge and discharge, and insufficient charge and discharge. They respectively represent the charge and discharge saturation of the battery, from high to low, that is, full charge and discharge means full charge and discharge, general charge and discharge means that one of the charge or discharge is insufficient, and insufficient charge and discharge means that both charge and discharge are insufficient, or even poor. There are 14, 6, and 7 batteries in the three groups respectively. Another three batteries were not online during this period, and no valuable data was collected, so they were not grouped. The capacity decay curves of the three groups are as follows Figure 4 shown.

[0046] The machine learning algorithm is used to predict the battery life and performance of data from different groups; the present invention selects Random Forest, Gradient Boosting Machine and Long Short-Term Memory Network (LSTM) as machine learning algorithms; the machine learning model is trained using the grouped data. In order to verify the effectiveness of DTW feature extraction, we plotted the prediction results of the best feature subsets obtained by selecting DTW features, all features, and conventional methods after grouping. Four groups of batteries were selected as examples. They are from different groups. The prediction results are shown in the figure below. Figure 6 As shown in the figure, the prediction results of DTW features are the best, and compared with other methods, the prediction effect of DTW method is improved by 80% on average, and the average root mean square error reaches 1.48. The model is trained with the training set and tuned with the validation set to ensure the accuracy and robustness of the model; the new data is input into the trained model to predict the remaining life and performance of the battery, and the prediction results after grouping are compared with the overall prediction results without grouping. Figure 7 As shown in the figure, it is found that the prediction effect of group prediction is nearly doubled compared to the overall prediction.

[0047] The specific evaluation indicators of each battery are shown in Table 1, and RMSE is used as the evaluation indicator: Table 1 Battery / Method / RMSE (%) Battery No. 1 Battery No. 2 Battery 3 Battery 4 Original Features 9.81 9.15 7.18 11.59 The best subset of conventional methods 1.88 1.76 3.75 1.65 DTW Features 0.49 0.62 4.43 0.41 Calculate the average RMSE for each method: Original feature: (9.81 + 9.15 + 7.18 + 11.59) / 4 = 9.43 Best subset of conventional method: (1.88+1.76+3.75+1.65) / 4=2.26 DTW feature: (0.49+0.62+4.43+0.41) / 4=1.49 1. The reduction ratio of DTW compared to the original feature RMSE: (9.43-1.49) / 9.43×100%=84.2% 2. The reduction ratio of RMSE of the best subset of DTW compared with the conventional method: (2.26-1.49) / 2.26×100%=34.2% result: DTW reduces RMSE by 84.2% compared to the original feature DTW reduces the RMSE of the best subset compared to the conventional method by 34.2%

[0048] The embodiments and / or implementation methods described above are only used to illustrate the preferred embodiments and / or implementation methods for realizing the technology of the present invention, and are not intended to limit the implementation methods of the technology of the present invention in any form. Any technical personnel in this field may make slight changes or modifications to other equivalent embodiments without departing from the scope of the technical means disclosed in the content of the present invention, but they should still be regarded as technologies or embodiments that are essentially the same as the present invention.

Claims

1. A method for predicting the health status of lithium-ion batteries for logistics leasing, characterized in that: The following steps are involved: S1: extracting the performance parameters of each lithium-ion battery in each charge and discharge cycle, including current, voltage and temperature, and calculating the maximum, minimum and average values ​​of the current, voltage and temperature as statistical features, clustering the statistical features of all batteries in each charge and discharge segment, obtaining a specified number of clusters after clustering, and generating a reference sequence that can represent the common mode of each cluster; S2: for each charge and discharge cycle of each battery, calculate the minimum cumulative dynamic time warping distance between the standardized feature sequence of the battery in the charge and discharge cycle and the reference sequence of each cluster, and convert the minimum cumulative dynamic time warping distance into a vector as the final dynamic time warping feature vector of the battery in the charge and discharge cycle; S3: extract the remaining power of each battery at the end of each charge and discharge cycle for probability statistics, extract the probability density distribution peak of the probability statistics, obtain the charge and discharge sufficiency of the battery, and group the batteries according to the charge and discharge sufficiency of each battery; S4: Using the dynamic time warped sequence feature vectors of the charge and discharge cycles of each group of lithium-ion batteries as features, the machine learning extreme gradient boosting tree algorithm is used to predict the health status of lithium-ion batteries in different charge and discharge degree groups.

2. The method according to claim 1, characterized in that The S1 includes: S11: Obtaining an original feature sequence, wherein the original feature sequence of the jth charging segment of the i-th battery is in the following format, and the jth charging segment is a charging segment in the jth charge-discharge cycle: ,in is the maximum current of the jth charging segment, is the minimum current value of the jth charging segment, is the average current of the jth charging segment, is the maximum voltage of the jth charging segment, is the minimum voltage of the jth charging segment, is the average voltage of the jth charging segment, is the maximum temperature of the jth charging segment, is the minimum temperature of the jth charging segment, is the average temperature of the jth charging segment.

3. The method according to claim 2, characterized in that The S2 includes: S21: The original feature sequence of each battery Perform feature standardization to obtain a standardized feature sequence; S22: according to each standardized feature sequence, principal component analysis is used for dimensionality reduction, and K-means clustering is performed on the standardized feature sequences of all battery charging segments and discharging segments after dimensionality reduction, to generate a specified number of charging clusters and discharging clusters; S23: for the generated charging clusters and discharging clusters, a reference sequence is generated for each cluster using a time series averaging algorithm based on dynamic time warping; S24: calculating the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster, vectorizing the dynamic time warping distance, and generating a final dynamic time warping feature vector; S25: Obtain the dynamic time warping feature vectors of all batteries in each charge and discharge cycle. For a certain charge and discharge cycle c, calculate the minimum cumulative dynamic time warping distance between the standardized charge feature sequence and the reference sequence of each charge cluster in the charge segment, and calculate the dynamic time warping distance between the standardized discharge feature sequence and the reference sequence of each discharge cluster in the discharge segment. Concatenate the calculated dynamic time warping distances to obtain the final dynamic time warping feature vector of the charge and discharge cycle. The form is: ,in, is the charging stage characteristic, It is the characteristic of the discharge stage.

4. The method according to claim 3, characterized in that: In S21, the standardized feature sequence includes the following steps: For each battery, the global mean of the battery is calculated based on each feature of the battery's charge and discharge stages. and standard deviation , k ranges from 1 to 9; in, is the number of all charging or discharging stages of the battery, is the value of the kth feature in the lth charging or discharging segment; After obtaining the global mean and the standard deviation of the battery, each feature of the battery is standardized by the following formula: in The standardized value of the kth feature in the lth charging or discharging segment generates the standardized feature sequence of the battery , .

5. The method according to claim 3, characterized in that: In S23, generating a reference sequence includes the following steps: For each cluster, an original feature sequence in the charging or discharging cycle of a lithium-ion battery in the cluster is randomly selected as the current reference sequence. For the original feature sequences of other batteries, dynamic time warping is used to calculate the distance between the original feature sequences of other batteries and the current reference sequence, and the optimal alignment path is found. The average value at each position after alignment is calculated, and a new reference sequence is generated to replace the current reference sequence. This step is repeated until the difference between the current reference sequence and the previous reference sequence is less than a set threshold, which is considered to converge, and a reference sequence generated for the cluster is obtained.

6. The method according to claim 3, characterized in that In S24, the following steps are included: The dynamic time warping distance calculation formula is: , in, Represents the standardized charge and discharge characteristic sequence The reference sequence of each cluster The dynamic time warping distance between them; for multidimensional feature vectors and , the distance is defined as: The dynamic time warping distance matrix D is at position Value The calculation of is defined as: At a specific point in time Under this condition, the dynamic time warping distance is , the minimum cumulative dynamic time warping distance between the standardized feature sequence of each battery and the reference sequence of each cluster can be calculated by starting from the first position (1,1) of the matrix, and each position in the first row It is accumulated from left to right, each position in the first column It is accumulated from top to bottom, accumulating the distance of each step in turn until the calculation reaches the lower right corner of the matrix D(P, Q). D(P,Q) represents the minimum dynamic time warping distance of the two sequences. It is calculated from the first position of the matrix and P and Q are the lengths of the two sequences respectively. In this way, a complete dynamic time warping distance matrix can be obtained, in which the element in the lower right corner of the matrix is ​​the required minimum dynamic time warping distance. Therefore, the value of each position is the minimum cumulative distance of the three positions above, left or diagonal.

7. The method according to claim 1, characterized in that In S3, the charge and discharge sufficiency degree grouping method includes the following sub-steps: S31: Calculate the probability density distribution function of the remaining power of each battery according to the remaining power at the end of the charging and discharging stages in each charging and discharging cycle of each battery; S32: according to the peak position of the battery's remaining power probability density function, obtain the remaining power value corresponding to the peak value of the remaining power probability density distribution, and determine the grouping threshold of the remaining power based on the remaining power value corresponding to the peak value of the remaining power probability density distribution and the healthy charging and discharging experience, the charging and discharging thresholds are 50% and 80% and 40% and 50% respectively; S33: For the charging stage, if the peak value of the probability density distribution of the remaining power is above 80% of the full charge, it is considered to be fully charged; if the peak value of the probability density distribution of the remaining power is at a medium charge, i.e., 50%-80% of the full charge, it is considered to be moderately charged; if the peak value of the probability density distribution of the remaining power is below 50% of the full charge, it is considered to be insufficiently charged; S34: For the discharge stage, if the peak value of the probability density distribution of the remaining power is above 50% of the full rating, the discharge is considered to be insufficient; if the peak value of the probability density distribution of the remaining power is at the medium power level, i.e. 40%-50% of the full rating, the discharge is considered to be medium sufficient; if the peak value of the probability density distribution of the remaining power is below 40% of the full rating, the discharge is considered to be sufficient; S35: According to the comprehensive situation of charge and discharge classification, the battery is divided into three groups, namely, fully charged and discharged, medium charge and discharge, and insufficient charge and discharge.

8. A health status prediction system for lithium-ion batteries for logistics leasing, executing the health status prediction method for lithium-ion batteries for logistics leasing as claimed in any one of claims 1 to 7, characterized in that: include: Charge and discharge feature extraction module: extracts the performance parameters of each lithium-ion battery in each charge and discharge cycle, including current, voltage and temperature, and calculates the maximum, minimum and average values ​​of the current, voltage and temperature as statistical features, clusters the statistical features of all batteries in each charge and discharge segment, obtains a specified number of clusters after clustering, and generates a reference sequence that can represent the common pattern of each cluster; Data processing module: for each charge and discharge cycle of each battery, calculate the minimum cumulative dynamic time warping distance between the standardized feature sequence of the battery in the charge and discharge cycle and the reference sequence of each cluster, and convert the minimum cumulative dynamic time warping distance into a vector as the final dynamic time warping feature vector of the battery in the charge and discharge cycle; Grouping module based on charge and discharge sufficiency: extract the remaining power of each battery at the end of each charge and discharge cycle for probability statistics, extract the probability density distribution peak of the probability statistics, obtain the charge and discharge sufficiency of the battery, and group the batteries according to the charge and discharge sufficiency of each battery; Lithium-ion battery grouping prediction module: Using the dynamic time warped sequence feature vector in the charge and discharge cycle of each group of lithium-ion batteries as the feature, the machine learning extreme gradient boosting tree algorithm is used to predict the health status of lithium-ion batteries in different charge and discharge degree groups.

Images