Battery management method, device and apparatus
By calculating the characteristic index of battery voltage data and dynamically updating the weight coefficients, the problem of the lack of adaptability of weight coefficients in battery management is solved, thereby improving the accuracy and reliability of battery management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WEICHAI POWER CO LTD
- Filing Date
- 2026-02-10
- Publication Date
- 2026-05-29
AI Technical Summary
In existing battery management technologies, the weighting coefficient settings lack adaptability, resulting in low accuracy and reliability of battery management, especially when the battery operating conditions change or the sensor fails, the evaluation accuracy decreases.
By calculating the characteristic index of battery voltage data, and dynamically updating the weighting coefficients based on the weighting coefficients of the characteristic indexes, the first, second, and third indicators of voltage data, the battery consistency score is calculated to achieve adaptive battery management.
It improves the accuracy and reliability of battery management, adapts to changes in battery operating conditions and sensor failures, and ensures accurate assessment.
Smart Images
Figure CN122109841A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of battery technology, and more specifically to a battery management method, apparatus, and device. Background Technology
[0002] In the field of battery technology, battery packs, as important energy storage devices in the new energy sector, play a crucial role in energy management in scenarios such as electric vehicles and energy storage power stations. Through battery management, battery life can be extended.
[0003] Existing battery management technologies generally suffer from the following significant drawbacks: Firstly, the weighting coefficients lack adaptability. Secondly, existing multi-feature fusion schemes often use fixed weighting coefficients determined based on experience or offline experiments. When battery conditions change or certain sensors malfunction, leading to a decline in data quality, this fixed weighting coefficient strategy can severely reduce evaluation accuracy or even result in incorrect judgments. These shortcomings lead to low accuracy and reliability in practical engineering applications. Summary of the Invention
[0004] In view of this, embodiments of the present invention aim to provide a battery management method, apparatus, and device to solve the problem that the lack of adaptability in the setting of weight coefficients in the prior art leads to low accuracy and reliability of battery management.
[0005] This invention provides a battery management method, the method comprising: Based on the battery voltage data, a feature index of the voltage data is calculated. The number of feature indices is greater than 1, and each feature index has a corresponding weighting coefficient. Based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, an adaptive weight coefficient is calculated. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. Based on the adaptive weight coefficient, the weight coefficient corresponding to the feature index is updated to obtain the target weight coefficient corresponding to the feature index. The battery consistency score is calculated based on the characteristic index and the corresponding target weight coefficient. The battery is managed based on the battery consistency score.
[0006] In one possible embodiment, before calculating the adaptive weighting coefficient based on the weighting coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, the method further includes: Obtain the coefficient of variation, and calculate the first index of the voltage data based on the coefficient of variation; Obtain the standard score value of the battery, and calculate the second index of the voltage data based on the absolute value of the standard score value of the battery; The effective ratio and data smoothness of the voltage data are obtained, and a third index of the voltage data is calculated based on the effective ratio and data smoothness.
[0007] In one possible embodiment, the calculation of adaptive weighting coefficients based on the weighting coefficients corresponding to the feature index, and the first, second, and third indicators of the voltage data includes: The adaptive weighting coefficient is obtained by adding the weighting coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data.
[0008] In one possible embodiment, updating the weight coefficients corresponding to the feature index based on the adaptive weight coefficients to obtain the target weight coefficients corresponding to the feature index includes: The target weight coefficients are normalized to obtain a normalized target weight coefficient vector; Multiply the target weight coefficient vector with the weight coefficient corresponding to the feature index to obtain the target weight coefficient corresponding to the feature index.
[0009] In one possible embodiment, calculating the battery consistency score based on the feature index and the corresponding target weight coefficient includes: Multiply the feature index by the corresponding target weight coefficient to obtain the product of each feature index and its corresponding target weight coefficient; The battery consistency score is calculated by adding the products of each of the aforementioned characteristic indices.
[0010] In one possible embodiment, the characteristic index includes instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier proportion; the step of calculating the characteristic index of the voltage data based on the battery voltage data includes: Calculate the mean, standard deviation, and range of the voltage data; The instantaneous dispersion is calculated based on the mean, standard deviation, and range. The trend inconsistency is calculated based on the mean and standard deviation. Based on the voltage data at different time points in the voltage data, the voltage difference is calculated, and the volatility index is calculated based on the voltage difference; The embedding dimension and similarity tolerance are set, and the distribution entropy is calculated based on the voltage data, embedding dimension, and similarity tolerance. Based on the data type of the voltage data, the EMA period is set, and the EMA deviation is calculated based on the EMA period; Obtain abnormal data from the voltage data, divide the number of abnormal data by the total voltage data, and obtain the proportion of abnormal values.
[0011] In one possible embodiment, managing the battery based on the battery consistency score includes: Based on the battery consistency score, obtain the battery management action; The battery is managed according to the management actions described.
[0012] In one possible embodiment, before calculating the characteristic index of the voltage data based on the battery voltage data, wherein the number of characteristic indices is greater than 1 and each characteristic index has a corresponding weighting coefficient, the method further includes: Real-time acquisition of the battery's current voltage data; Based on the current voltage data and historical voltage data, the data reliability of the current voltage data is calculated; Configure the historical depth of the voltage data based on the reliability of the data.
[0013] In a second aspect, the present invention provides a battery management device, the device comprising: The first calculation module is used to calculate the feature index of the voltage data based on the battery voltage data. The number of the feature indexes is greater than 1, and each feature index has a corresponding weight coefficient. The second calculation module is used to calculate adaptive weight coefficients based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. The update module is used to update the weight coefficient corresponding to the feature index according to the adaptive weight coefficient, so as to obtain the target weight coefficient corresponding to the feature index. The third calculation module is used to calculate the battery consistency score based on the feature index and the corresponding target weight coefficient. The management module is used to manage the battery based on the battery consistency score.
[0014] Thirdly, the present invention provides an electronic device, the device comprising: a memory and a processor; the memory being used to store related program code; the processor being used to call the program code to execute the battery management method described in any of the implementations of the first aspect.
[0015] Fourthly, the present invention provides a computer-readable storage medium for storing a computer program for executing the battery management method described in any implementation of the first aspect.
[0016] Fifthly, the present invention provides a computer program product, the computer program product comprising a computer program / instruction, which, when executed by a processor, implements the battery management method described in any of the implementations of the first aspect.
[0017] In the above implementation of the present invention, a feature index of the voltage data is calculated based on the battery voltage data. The number of feature indices is greater than 1, and each feature index has a corresponding weight coefficient. An adaptive weight coefficient is calculated based on the weight coefficient corresponding to the feature index, a first index, a second index, and a third index of the voltage data. The first index characterizes the feature reliability of the voltage data, the second index characterizes the data quality of the voltage data, and the third index characterizes the information gain of the voltage data. The weight coefficient corresponding to the feature index is updated according to the adaptive weight coefficient to obtain a target weight coefficient corresponding to the feature index. A battery consistency score is calculated based on the feature index and the corresponding target weight coefficient. The battery is managed according to the battery consistency score. This adaptive weight coefficient update achieves adaptive battery management, improving the accuracy and reliability of battery management. Attached Figure Description
[0018] Figure 1 This is a flowchart of a battery management method provided in an embodiment of the present invention.
[0019] Figure 2 Another flowchart of a battery management method provided in an embodiment of the present invention.
[0020] Figure 3 A schematic diagram of a battery management device provided in an embodiment of the present invention.
[0021] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] In the field of battery technology, battery packs, as important energy storage devices in the new energy sector, play a crucial role in energy management in scenarios such as electric vehicles and energy storage power stations. Through battery management, battery life can be extended.
[0024] Existing battery management technologies generally suffer from the following significant drawbacks: the weighting coefficients lack adaptability. Current multi-feature fusion schemes often use fixed weighting coefficients determined based on experience or offline experiments. When battery conditions change or certain sensors malfunction, leading to a decline in data quality, this fixed weighting coefficient strategy can severely reduce evaluation accuracy or even result in incorrect judgments. These shortcomings lead to low accuracy and reliability in practical engineering applications.
[0025] Therefore, in this invention, the weight coefficients of the feature index are updated based on the first, second, and third indicators of the voltage data to achieve adaptive updating of the weight coefficients, thereby improving the accuracy and reliability of battery management.
[0026] Specifically, one embodiment of the present invention provides a battery management method. This method calculates a feature index of the battery voltage data, where the number of feature indices is greater than one, and each feature index has a corresponding weight coefficient. Based on the weight coefficients corresponding to the feature indices, and a first, second, and third indicator of the voltage data, an adaptive weight coefficient is calculated. The first indicator characterizes the feature reliability of the voltage data, the second indicator characterizes the data quality of the voltage data, and the third indicator characterizes the information gain of the voltage data. The weight coefficients corresponding to the feature indices are updated according to the adaptive weight coefficients to obtain target weight coefficients for the feature indices. A battery consistency score is calculated based on the feature indices and the corresponding target weight coefficients. The battery is then managed based on the battery consistency score. This adaptive weight coefficient update achieves adaptive battery management, improving the accuracy and reliability of battery management.
[0027] Please see Figure 1 In one exemplary embodiment, a battery management method is provided, which is applied to a battery pack. The method may include the following steps: S101: Based on the battery voltage data, calculate the characteristic index of the voltage data. The number of the characteristic indexes is greater than 1, and each characteristic index has a corresponding weight coefficient.
[0028] Specifically, characteristic indices of the voltage data can be calculated based on the voltage data. These indices can include instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier proportion. Each index has a corresponding weighting coefficient. The weighting coefficients for instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier proportion are set to A1, A2, A3, A4, A5, and A6, respectively. The sum of A1, A2, A3, A4, A5, and A6 is 1. The specific values of A1, A2, A3, A4, A5, and A6 can be set empirically and are not restricted here.
[0029] Specifically, the step of calculating the characteristic index of the voltage data based on the battery voltage data includes: Calculate the mean, standard deviation, and range of the voltage data; The instantaneous dispersion is calculated based on the mean, standard deviation, and range. The trend inconsistency is calculated based on the mean and standard deviation. Based on the voltage data at different time points in the voltage data, the voltage difference is calculated, and the volatility index is calculated based on the voltage difference; The embedding dimension and similarity tolerance are set, and the distribution entropy is calculated based on the voltage data, embedding dimension, and similarity tolerance. Based on the data type of the voltage data, the EMA period is set, and the EMA deviation is calculated based on the EMA period; Obtain abnormal data from the voltage data, divide the number of abnormal data by the total voltage data, and obtain the proportion of abnormal values.
[0030] In the specific implementation process, the original voltage data of N cells at time t is read, and the original voltage data is checked for data validity. Invalid cells are marked and removed, leaving M valid cells. The average value μ, standard deviation σ, and range ΔV of the voltage set of the valid cells are calculated. The coefficient of variation CV is calculated based on the average value and standard deviation, with the specific formula: CV = σ / μ. The instantaneous dispersion index is calculated based on the standard deviation, range, and coefficient of variation. Specifically, when the instantaneous dispersion index consists of the standard deviation σ, range ΔV, and coefficient of variation, the instantaneous dispersion index is (σ_t, ΔV_t, CV_t).
[0031] Select voltage data points with the same number of cycles. Calculate the standard deviation (σ) and mean (μ) of the capacity of all batteries at that point based on the selected data. Calculate the relative standard deviation (C1) of the battery capacity based on the standard deviation and mean. The formula for calculating the relative standard deviation of capacity is: C1 = (σ / μ) 100%. Among them, the larger the C1 value, the worse the capacity consistency between batteries at that moment, and the more obvious the trend inconsistency.
[0032] Volatility indicators can include volatility. Volatility can be calculated from voltage differences. The formula for calculating the voltage difference is: rt = xt - x_{t-1}, where xt represents the voltage value observed at time t, and x{t-1} represents the voltage value of the same battery observed at the previous time t-1. The formula for calculating volatility A is: A = sqrt( [Σ(rt - μr)²] / (N-1) ), where μr is the average of all rt values, and N is the length of the rt sequence.
[0033] The calculation process for distribution entropy includes: acquiring the voltage sequence V(t) = [3.200, 3.205, 3.210, ..., 4.200] within a preset sampling period from the voltage data. Setting the embedding dimension m, typically 3, which is the length of the segment being compared. Setting the similarity tolerance r, typically 0.1 to 0.25 times the standard deviation of the voltage sequence. Statistically calculating the proportion of vectors of length m and m+1 in the entire sequence whose distance is less than r, yields the distribution entropy.
[0034] For voltage data, the EMA period is selected as N. The choice of N can be determined based on the voltage data. For example, if the voltage data is high-frequency, such as voltage per second, N can be a larger value, such as N=60, representing the typical trend over the past minute. If the voltage data is low-frequency, such as capacity per cycle, then N can be a smaller value, such as N=3, representing the trend over the last three cycles.
[0035] The smoothing coefficient is calculated based on N, where the formula for calculating the smoothing coefficient C2 is as follows: C 2 = 2 / (N + 1).
[0036] The EMA is calculated recursively based on the smoothing coefficient. The formula for calculating the EMA is as follows: EMA_V(t) = C2 V(t) + (1 – C2) EMA_V(t-1) The initial value EMA_V(0) is typically taken as the SMA of the first data point or the first few points. V(t) represents the voltage data. SMA refers to the arithmetic mean of all voltage data points within a fixed-length window.
[0037] The instantaneous deviation is calculated based on the EMA. The formula for calculating the instantaneous deviation is: Instantaneous deviation (t) = V(t) - EMA_V(t-1).
[0038] The EMA deviation is calculated based on the instantaneous deviation. The formula for calculating the EMA deviation is: EMA_Deviation(t) = C_dev |Instantaneous deviation(t)| + (1 - C_dev) EMA_Deviation(t-1).
[0039] The outlier ratio can be obtained by dividing the number of outliers in the voltage data by the total amount of voltage data.
[0040] Through the above calculation process, instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier ratio can be obtained.
[0041] Furthermore, the historical depth can be configured based on the data quality of the current voltage data. That is, based on the battery voltage data, a feature index of the voltage data is calculated. The number of feature indices is greater than one, and each feature index has a corresponding weighting coefficient. Before this, the following steps are also included: Real-time acquisition of the battery's current voltage data; Based on the current voltage data and historical voltage data, the data reliability of the current voltage data is calculated; Configure the historical depth of the voltage data based on the reliability of the data.
[0042] In the specific implementation process, the current voltage data of the battery is collected in real time; based on the current voltage data and historical voltage data, the reliability of the current voltage data is calculated; wherein, the historical voltage data is historical voltage data of a preset number of frames, for example, a historical feature sequence of 50 frames of voltage data. The reliability of the current voltage data is calculated based on the historical feature sequence of 50 frames of voltage data and the current voltage data. Specifically, based on the historical feature sequence of 50 frames of voltage data and the current voltage data, the mean μ of the current voltage data is calculated, and the standard deviation σ is calculated based on the mean. The absolute deviation is calculated as: d = |V_t - μ|. A threshold is set, and the reliability R is calculated based on the absolute deviation and the standard deviation. Specifically, if d <= k1 If σ, then R = 1.0, if d <= k2 If σ is positive, then R = 0.5; otherwise, R = 0.0. Where k1 = 2, k2 = 3. Configure the historical depth of the voltage data based on the data reliability. For example, assuming R = 1, it indicates high data quality, and a shorter historical window of voltage data can be used for analysis. In this case, the historical depth of the voltage data can be configured as 20 or 30, i.e., less than 50. Assuming R = 0.5, it indicates that the current voltage data itself is unreliable, and the historical window needs to be lengthened. For example, setting the historical depth of voltage data to 100 frames allows for a sliding window design to capture the temporal variation characteristics of the data, fully utilizing the temporal characteristics of the voltage data. Furthermore, the historical depth can be configured based on data quality to improve the reliability of the voltage data, resulting in a higher accuracy of the calculated battery consistency score.
[0043] S102: Based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, an adaptive weight coefficient is calculated. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data.
[0044] In this embodiment, after calculating the characteristic index of the voltage data, a first index, a second index, and a third index of the voltage data can be further calculated. The first index characterizes the characteristic reliability of the voltage data, the second index characterizes the data quality of the voltage data, and the third index characterizes the information gain of the voltage data. Adaptive weighting coefficients are then calculated based on the first, second, and third indices. Therefore, before calculating the adaptive weighting coefficients, the following steps may also be included: Obtain the coefficient of variation, and calculate the first index of the voltage data based on the coefficient of variation; Obtain the standard score value of the battery, and calculate the second index of the voltage data based on the absolute value of the standard score value of the battery; The effective ratio and data smoothness of the voltage data are obtained, and a third index of the voltage data is calculated based on the effective ratio and data smoothness.
[0045] In the specific implementation process, the coefficient of variation (CV) is obtained by calculating the mean and standard deviation of the voltage data. The specific calculation formula is: CV = σ / μ, where μ is the mean of the voltage data and σ is the standard deviation of the voltage data. The calculation formula for the first indicator is: Where R(f) represents the first indicator, and CV is the coefficient of variation. The smaller the variation, the higher the reliability and the higher the weight. The mapping range of R(f) is [0.5, 1.5].
[0046] Obtain the standard score of the battery and calculate the absolute value of the standard score. Based on the absolute value of the standard score, calculate the second index of the voltage data. Specifically, the formula for calculating the second index is: Here, G(f) is the second indicator, and z-score is the standard score value of the battery corresponding to the current voltage data. The mapping range of G(f) is [0.5, 1.5]. The larger the value of G(f) is, the greater the deviation, the greater the information content, and the corresponding weight is increased.
[0047] The effective proportion and data smoothness of the voltage data are obtained, and a third index is calculated based on these parameters. The specific formula for calculating the third index is as follows: Where Q(f) is the third index, valid_ratio is the effective ratio of the current voltage data, and smoothness is the smoothness of the current voltage data.
[0048] After calculating the first, second, and third indicators of the voltage data, adaptive weighting coefficients can be calculated based on the weighting coefficients corresponding to the feature indices and the first, second, and third indicators of the voltage data. Specifically, the basic weights of the voltage data are further obtained. These basic weights are the sum of A1, A2, A3, A4, A5, and A6. The basic weights are then added to the first, second, and third indicators to obtain the adaptive weighting coefficients of the voltage data. The specific formula for calculating the adaptive weighting coefficients is as follows: Among them, W base (f) is the base weight.
[0049] S103: Update the weight coefficients corresponding to the feature index according to the adaptive weight coefficients to obtain the target weight coefficients corresponding to the feature index.
[0050] In the specific implementation process, after calculating the adaptive weighting coefficients of the voltage data, these coefficients can be further normalized to obtain the normalized target weighting coefficient vector. The specific normalization formula is as follows: Multiply the target weight coefficient vector with the weight coefficient corresponding to the feature index to obtain the target weight coefficient corresponding to the feature index.
[0051] Specifically, the step of updating the weight coefficients corresponding to the feature index based on the adaptive weight coefficients to obtain the target weight coefficients corresponding to the feature index includes: The target weight coefficients are normalized to obtain a normalized target weight coefficient vector; Multiply the target weight coefficient vector with the weight coefficient corresponding to the feature index to obtain the target weight coefficient corresponding to the feature index.
[0052] In this embodiment, the adaptive weight coefficients are normalized to obtain a normalized target weight coefficient vector. The target weight coefficient vector is then multiplied by the weight coefficients corresponding to the feature indices to obtain the target weight coefficients corresponding to the feature indices. For example, the weights set for instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier ratio are A1, A2, A3, A4, A5, and A6, respectively. The target weight coefficient vector is then multiplied by A1, A2, A3, A4, A5, and A6, respectively, to obtain the target weight coefficients A11, A22, A33, A44, A55, and A66 corresponding to the feature indices.
[0053] S104: Calculate the battery consistency score based on the characteristic index and the corresponding target weight coefficient.
[0054] In the specific implementation process, the battery consistency score is obtained by combining the characteristic indices with their corresponding target weight coefficients. When the characteristic indices include instantaneous dispersion, trend inconsistency, volatility indicators, distribution entropy, EMA deviation, and outlier proportion, and the corresponding target weight coefficients are A11, A22, A33, A44, A55, and A66, the battery consistency score equals the instantaneous dispersion. A11+Trend Inconsistency; A22+Volatility Indicator; A33+Distribution Entropy; A44+EMA Bias; A55+Outlier Ratio; A66.
[0055] S105: Manage the battery based on the battery consistency score.
[0056] In practice, batteries are managed based on their consistency scores. Specifically, the steps for managing batteries based on their consistency scores include: Based on the battery consistency score, obtain the battery management action; The battery is managed according to the management actions described.
[0057] In practice, management actions are obtained based on the battery consistency score, and the battery is managed accordingly. Taking a battery pack as an example, the specific management based on the calculated battery consistency score is explained in detail. Assume the battery pack consists of 96 series-connected cells, and the battery consistency score S ranges from 0 to 100, with higher scores indicating better consistency. A hierarchical management strategy is then implemented based on this score. The specific management strategy can be found in the table below.
[0058] Further, please refer to Figure 2 This illustrates a flow diagram of an embodiment of a battery management method according to this application. The battery management method includes the following steps: S201, Real-time acquisition of the current voltage data of the battery, and calculation of the data reliability of the current voltage data based on the current voltage data and historical voltage data; Real-time acquisition of current battery voltage data; calculation of the reliability of the current voltage data based on the current voltage data and historical voltage data; wherein, the historical voltage data is a preset number of historical voltage data, for example, a historical feature sequence of 50 frames of voltage data. The reliability of the current voltage data is calculated based on the historical feature sequence of 50 frames of voltage data and the current voltage data. Specifically, the mean μ of the voltage data is calculated, and the standard deviation σ is calculated based on the mean. The absolute deviation is calculated: d = |V_t - μ|. A threshold is set, and the reliability R is calculated based on the absolute deviation and the standard deviation.
[0059] S202, Based on the data reliability, configure the historical depth of the voltage data, and extract voltage data from the historical voltage data based on the historical depth; Specifically, if d <= k1 If σ, then R = 1.0, if d <= k2 If σ is positive, then R = 0.5; otherwise, R = 0.0. Where k1 = 2, k2 = 3. Configure the historical depth of the voltage data based on the data reliability. For example, assuming R = 1, it indicates high data quality, and a shorter historical window of voltage data can be used for analysis. In this case, the historical depth of the voltage data can be configured as 20 or 30, i.e., less than 50. Assuming R = 0.5, it indicates that the current voltage data itself is unreliable, and the historical window needs to be lengthened. For example, setting the historical depth of voltage data to 100 frames allows for a sliding window design to capture the temporal variation characteristics of the data, fully utilizing the temporal characteristics of the voltage data. Furthermore, the historical depth can be configured based on data quality to improve the reliability of the voltage data, resulting in a higher accuracy of the calculated battery consistency score.
[0060] S203, Based on the battery voltage data, calculate the characteristic index of the voltage data, wherein the number of the characteristic index is greater than 1, and each characteristic index has a corresponding weight coefficient; Specifically, characteristic indices of the voltage data can be calculated based on the voltage data. These indices can include instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier ratio. Each indices have corresponding weighting coefficients. The weighting coefficients for instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier ratio are A1, A2, A3, A4, A5, and A6, respectively. The sum of A1, A2, A3, A4, A5, and A6 is 1. The specific values of A1, A2, A3, A4, A5, and A6 can be set empirically and are not restricted here.
[0061] S204, based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, an adaptive weight coefficient is calculated. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. After calculating the characteristic index of the voltage data, the first, second, and third indicators of the voltage data can be further calculated. Based on the weight coefficients corresponding to the first, second, and third indicators and the characteristic index, the adaptive weight coefficients can be calculated.
[0062] S205, Update the weight coefficient corresponding to the feature index according to the adaptive weight coefficient to obtain the target weight coefficient corresponding to the feature index; In the specific implementation process, after calculating the adaptive weight coefficients of the voltage data, the adaptive weight coefficients can be further normalized to obtain the target weight coefficient vector. The target weight coefficient vector is then multiplied by the weight coefficients corresponding to the feature index to obtain the target weight coefficients corresponding to the feature index.
[0063] S206, Calculate the battery consistency score based on the feature index and the corresponding target weight coefficient; The battery consistency score is calculated by using the characteristic index and the corresponding target weight coefficient.
[0064] S207, manage the battery according to the battery consistency score.
[0065] Based on the battery consistency score, obtain the battery management action, and manage the battery according to the management action.
[0066] Based on the method provided in the above embodiments, a feature index of the battery voltage data is calculated, wherein the number of feature indices is greater than 1, and each feature index has a corresponding weight coefficient. An adaptive weight coefficient is calculated based on the weight coefficient corresponding to the feature index, and a first, second, and third indicator of the voltage data. The weight coefficient corresponding to the feature index is updated according to the adaptive weight coefficient to obtain a target weight coefficient corresponding to the feature index. A battery consistency score is calculated based on the feature index and the corresponding target weight coefficient. The battery is managed according to the battery consistency score. This adaptive weight coefficient update achieves adaptive battery management, improving the accuracy and reliability of battery management.
[0067] Based on the above method embodiments, this invention also provides a battery management device. See also... Figure 3 The diagram shown is a schematic of a battery management device provided in an embodiment of the present invention.
[0068] The device 300 includes: The first calculation module 301 is used to calculate the characteristic index of the voltage data based on the battery voltage data, wherein the number of the characteristic index is greater than 1, and each characteristic index has a corresponding weight coefficient. The second calculation module 302 is used to calculate an adaptive weight coefficient based on the weight coefficient corresponding to the feature index, the first index, the second index and the third index of the voltage data. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. The update module 303 is used to update the weight coefficient corresponding to the feature index according to the adaptive weight coefficient, so as to obtain the target weight coefficient corresponding to the feature index. The third calculation module 304 is used to calculate the battery consistency score based on the feature index and the corresponding target weight coefficient. The management module 305 is used to manage the battery based on the battery consistency score.
[0069] In one possible implementation, before calculating the adaptive weighting coefficient based on the weighting coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, the method further includes: Obtain the coefficient of variation, and calculate the first index of the voltage data based on the coefficient of variation; Obtain the standard score value of the battery, and calculate the second index of the voltage data based on the absolute value of the standard score value of the battery; The effective ratio and data smoothness of the voltage data are obtained, and a third index of the voltage data is calculated based on the effective ratio and data smoothness.
[0070] In one possible implementation, the calculation of adaptive weighting coefficients based on the weighting coefficients corresponding to the feature index, and the first, second, and third indicators of the voltage data, includes: The adaptive weighting coefficient is obtained by adding the weighting coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data.
[0071] In one possible implementation, updating the weight coefficients corresponding to the feature index based on the adaptive weight coefficients to obtain the target weight coefficients corresponding to the feature index includes: The target weight coefficients are normalized to obtain a normalized target weight coefficient vector; Multiply the target weight coefficient vector with the weight coefficient corresponding to the feature index to obtain the target weight coefficient corresponding to the feature index.
[0072] In one possible implementation, calculating the battery consistency score based on the feature index and the corresponding target weight coefficient includes: Multiply the feature index by the corresponding target weight coefficient to obtain the product of each feature index and its corresponding target weight coefficient; The battery consistency score is calculated by adding the products of each of the aforementioned characteristic indices.
[0073] In one possible implementation, the characteristic index includes instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier proportion; the step of calculating the characteristic index of the voltage data based on the battery voltage data includes: Calculate the mean, standard deviation, and range of the voltage data; The instantaneous dispersion is calculated based on the mean, standard deviation, and range. The trend inconsistency is calculated based on the mean and standard deviation. Based on the voltage data at different time points in the voltage data, the voltage difference is calculated, and the volatility index is calculated based on the voltage difference; The embedding dimension and similarity tolerance are set, and the distribution entropy is calculated based on the voltage data, embedding dimension, and similarity tolerance. Based on the data type of the voltage data, the EMA period is set, and the EMA deviation is calculated based on the EMA period; Obtain abnormal data from the voltage data, divide the number of abnormal data by the total voltage data, and obtain the proportion of abnormal values.
[0074] In one possible implementation, managing the battery based on the battery consistency score includes: Based on the battery consistency score, obtain the battery management action; The battery is managed according to the management actions described.
[0075] In one possible implementation, before calculating the feature index of the battery voltage data, wherein the number of feature indices is greater than 1 and each feature index has a corresponding weight coefficient, the method further includes: Real-time acquisition of the battery's current voltage data; Based on the current voltage data and historical voltage data, the data reliability of the current voltage data is calculated; Configure the historical depth of the voltage data based on the reliability of the data.
[0076] See Figure 4 , Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention.
[0077] The electronic device 400 includes a memory 401 and a processor 402; the memory 401 is used to store relevant program code; the processor 402 is used to call the program code to execute the battery management method described in the above method embodiments.
[0078] Furthermore, embodiments of the present invention also provide a computer-readable storage medium for storing a computer program for executing the battery management method described in the above method embodiments.
[0079] This invention also provides a computer program product, which includes a computer program / instructions. When the computer program / instructions are executed by a processor, they implement the battery management method described in the above method embodiments.
[0080] It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.
[0081] The computer program product can be written in any combination of one or more programming languages to perform the operations of the embodiments of the present invention. The programming languages include object-oriented programming languages such as Java and C++, as well as conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0082] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. In particular, for system or device embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The device embodiments described above are merely illustrative. The units or modules described as separate components may or may not be physically separate. The components shown as units or modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the units or modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0083] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functions, and operations that may be implemented according to various embodiments of the invention, including methods, apparatus, and devices. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0084] It should be understood that in this invention, "at least one (item)" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0085] It should also be noted that, in this invention, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0086] The steps of the methods or algorithms described in conjunction with the embodiments disclosed in this invention can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A battery management method, characterized in that, The method includes: Based on the battery voltage data, a feature index of the voltage data is calculated. The number of feature indices is greater than 1, and each feature index has a corresponding weighting coefficient. Based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data, an adaptive weight coefficient is calculated. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. Based on the adaptive weight coefficient, the weight coefficient corresponding to the feature index is updated to obtain the target weight coefficient corresponding to the feature index. The battery consistency score is calculated based on the characteristic index and the corresponding target weight coefficient. The battery is managed based on the battery consistency score.
2. The method according to claim 1, characterized in that, Before calculating the adaptive weighting coefficient based on the weighting coefficients corresponding to the feature index, and the first, second, and third indicators of the voltage data, the method further includes: Obtain the coefficient of variation, and calculate the first index of the voltage data based on the coefficient of variation; Obtain the standard score value of the battery, and calculate the second index of the voltage data based on the absolute value of the standard score value of the battery; The effective ratio and data smoothness of the voltage data are obtained, and a third index of the voltage data is calculated based on the effective ratio and data smoothness.
3. The method according to claim 1, characterized in that, The adaptive weighting coefficients are calculated based on the weighting coefficients corresponding to the feature index, and the first, second, and third indicators of the voltage data, including: The adaptive weighting coefficient is obtained by adding the weighting coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data.
4. The method according to claim 1, characterized in that, The step of updating the weight coefficients corresponding to the feature index based on the adaptive weight coefficients to obtain the target weight coefficients corresponding to the feature index includes: The target weight coefficients are normalized to obtain a normalized target weight coefficient vector; Multiply the target weight coefficient vector with the weight coefficient corresponding to the feature index to obtain the target weight coefficient corresponding to the feature index.
5. The method according to claim 1, characterized in that, The calculation of the battery consistency score based on the feature index and the corresponding target weight coefficient includes: Multiply the feature index by the corresponding target weight coefficient to obtain the product of each feature index and its corresponding target weight coefficient; The battery consistency score is calculated by adding the products of each of the aforementioned characteristic indices.
6. The method according to claim 1, characterized in that, The characteristic indices include instantaneous dispersion, trend inconsistency, volatility index, distribution entropy, EMA deviation, and outlier ratio. The calculation of the characteristic indices of the voltage data based on the battery voltage data includes: Calculate the mean, standard deviation, and range of the voltage data; The instantaneous dispersion is calculated based on the mean, standard deviation, and range. The trend inconsistency is calculated based on the mean and standard deviation. Based on the voltage data at different time points in the voltage data, the voltage difference is calculated, and the volatility index is calculated based on the voltage difference; The embedding dimension and similarity tolerance are set, and the distribution entropy is calculated based on the voltage data, embedding dimension, and similarity tolerance. Based on the data type of the voltage data, the EMA period is set, and the EMA deviation is calculated based on the EMA period; Obtain abnormal data from the voltage data, divide the number of abnormal data by the total voltage data, and obtain the proportion of abnormal values.
7. The method according to claim 1, characterized in that, The management of the battery based on the battery consistency score includes: Based on the battery consistency score, obtain the battery management action; The battery is managed according to the management actions described.
8. The method according to claim 1, characterized in that, Before calculating the characteristic index of the voltage data based on the battery voltage data, wherein the number of characteristic indices is greater than 1 and each characteristic index has a corresponding weight coefficient, the method further includes: Real-time acquisition of the battery's current voltage data; Based on the current voltage data and historical voltage data, the data reliability of the current voltage data is calculated; Configure the historical depth of the voltage data based on the reliability of the data.
9. A battery management device, characterized in that, The device includes: The first calculation module is used to calculate the feature index of the voltage data based on the battery voltage data. The number of the feature indexes is greater than 1, and each feature index has a corresponding weight coefficient. The second calculation module is used to calculate adaptive weight coefficients based on the weight coefficients corresponding to the feature index, the first index, the second index, and the third index of the voltage data. The first index is used to characterize the feature reliability of the voltage data, the second index is used to characterize the data quality of the voltage data, and the third index is used to characterize the information gain of the voltage data. The update module is used to update the weight coefficient corresponding to the feature index according to the adaptive weight coefficient, so as to obtain the target weight coefficient corresponding to the feature index. The third calculation module is used to calculate the battery consistency score based on the feature index and the corresponding target weight coefficient. The management module is used to manage the battery based on the battery consistency score.
10. An electronic device, characterized in that, The device includes: a memory and a processor; the memory is used to store relevant program code; the processor is used to call the program code to execute the battery management method according to any one of claims 1 to 8.