Deep learning-driven dynamic evaluation method and system for residual capacity of battery

Abstract

The invention relates to a deep learning-driven battery residual capacity dynamic evaluation method and system, and relates to the technical field of battery capacity evaluation, and the method comprises the steps: reading a current SOC region and a current prediction error confidence interval; monitoring to obtain a multi-source operation data sequence, performing operation state stability analysis, and outputting a current operation stability coefficient; compensating a preset standard prediction depth to generate an adaptive prediction depth; and formulating an adaptive prediction scheme according to the current operation stability coefficient and the adaptive prediction depth, obtaining adaptive multi-source prediction data, calling a battery residual capacity prediction plug-in, performing battery residual capacity prediction on the target storage battery pack, and outputting a battery residual capacity evaluation result. The method solves the problem that the conventional battery residual capacity evaluation is not adaptive to an actual operation scene, and a unified evaluation scheme is adopted, so that excessive consumption of resources is easily caused or the evaluation precision cannot be matched with an actual demand.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of battery capacity evaluation, in particular to a deep learning driven dynamic battery residual capacity evaluation method and system. BACKGROUND

[0002] With the increasing demand for battery operation reliability in new energy and energy storage fields, the accuracy and resource efficiency of battery residual capacity evaluation have become key technical requirements.

[0003] Currently, traditional battery residual capacity evaluation methods do not adapt to actual operation scenarios, and mostly use a unified solution, which not only easily over-consumes resources, but also may not meet the demand due to insufficient accuracy, affecting the stability of equipment operation and the rationality of decision-making. SUMMARY

[0004] The present application provides a deep learning driven dynamic battery residual capacity evaluation method and system, which improves the status quo of traditional battery residual capacity evaluation not adapting to actual operation scenarios and using a unified solution leading to over-consumption of resources or evaluation accuracy not matching actual demand.

[0005] The embodiments of the present application disclose the following technical solutions: In a first aspect, the embodiments of the present application provide a deep learning driven dynamic battery residual capacity evaluation method, which comprises: reading the current SOC region where the target battery pack is located, and matching to obtain the current prediction error confidence interval of battery residual capacity prediction; monitoring and obtaining a multi-source operation data sequence of the target battery pack in a historical time zone, performing operation state stationarity analysis according to the multi-source operation data sequence, and outputting a current operation stationarity coefficient; compensating for a preset standard prediction depth according to the current prediction error confidence interval and the current operation stationarity coefficient, and generating an adaptive prediction depth; based on a preset prediction index space, formulating an adaptive prediction scheme according to the current operation stationarity coefficient and the adaptive prediction depth, collecting and obtaining adaptive multi-source prediction data according to the adaptive prediction scheme, and calling a battery residual capacity prediction plug-in constructed based on deep learning to perform battery residual capacity prediction on the target battery pack according to the adaptive multi-source prediction data, and outputting a battery residual capacity evaluation result.

[0006] In a second aspect, the embodiments of the present application provide a deep learning driven dynamic battery residual capacity evaluation system, which comprises: an SOC and error reading module for reading the current SOC region where the target battery pack is located, and matching to obtain the current prediction error confidence interval of battery residual capacity prediction; A multi-source data stationarity analysis module is configured to monitor and acquire a multi-source operation data sequence of the target battery pack in a historical time zone, perform operation state stationarity analysis based on the multi-source operation data sequence, and output a current operation stationarity coefficient; A prediction depth compensation module is configured to compensate a preset standard prediction depth based on the current prediction error confidence interval and the current operation stationarity coefficient, and generate an adaptive prediction depth. A prediction scheme execution module is configured to formulate an adaptive prediction scheme based on a preset prediction index space, the current operation stationarity coefficient and the adaptive prediction depth, collect adaptive multi-source prediction data according to the adaptive prediction scheme, and call a battery residual capacity prediction plug-in constructed based on deep learning to perform battery residual capacity prediction on the target battery pack based on the adaptive multi-source prediction data, and output a battery residual capacity evaluation result.

[0007] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: The present application provides a deep learning driven battery residual capacity dynamic evaluation method and system, which realizes accurate and dynamic evaluation of the residual capacity of the target battery pack by acquiring battery operation data in steps, formulating an adaptive prediction scheme, building a multi-model prediction plug-in, and performing dynamic prediction and result optimization. First, the multi-source operation data sequence of the target battery pack in the historical time zone is monitored and acquired, the electrical characteristic fluctuation degree and the load characteristic fluctuation degree are calculated, and the current operation stationarity coefficient is determined. Then, the preset standard prediction depth is compensated based on the current prediction error confidence interval to generate an adaptive prediction depth. Subsequently, based on the preset prediction index space, an adaptive prediction index set is selected and the number of adaptive selected models is determined to form an adaptive prediction scheme. At the same time, an adaptive battery capacity prediction plug-in containing N LSTM models is constructed with the adaptive prediction index as a constraint. Finally, adaptive multi-source prediction data is collected according to the adaptive prediction scheme, a randomly selected model in the adaptive battery capacity prediction plug-in is called for prediction, and the result is fitted by mean value to output the battery residual capacity evaluation result.

[0008] The technical solutions of the present application dynamically adjust the prediction index and the number of models, rely on the multi-model plug-in to improve the generalization ability, and combine the operation state to optimize the prediction depth, thereby solving the problems of data redundancy or information loss caused by fixed indexes, difficulty of single model to adapt to different operation conditions, and large influence of prediction accuracy by environment and operation state in traditional battery residual capacity evaluation. The evaluation deviation caused by improper index selection, weak model generalization ability or mismatched prediction depth is avoided, and the accuracy, dynamic adaptability and reliability of battery residual capacity evaluation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to make the technical solutions in the embodiments of the present application clearer, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0010] Figure 1 A flowchart of a deep learning driven battery residual capacity dynamic evaluation method provided by the embodiments of the present application is shown in the figure. Figure 2 A structural diagram of a deep learning driven battery residual capacity dynamic evaluation system provided by the embodiments of the present application is shown in the figure.

[0011] In the drawings, the components represented by the respective reference numerals are described as follows. The SOC and error reading module 01, the multi-source data stationary analysis module 02, the prediction depth compensation module 03, and the prediction scheme execution module 04. DETAILED DESCRIPTION

[0012] The present application provides a deep learning driven battery residual capacity dynamic evaluation method and system, which is used to solve the technical problem that the traditional battery residual capacity evaluation in the prior art is not adapted to the actual operation scene and uses a unified evaluation scheme, which is prone to cause excessive consumption of resources or evaluation precision that cannot match the actual demand.

[0013] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort fall within the scope of protection of the present application.

[0014] In the description of the present application, the terms "first" and "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0015] In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration." Any embodiment described as "for example" in this application is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for the purposes of explanation, numerous details are set forth in order to provide a thorough understanding. It should be apparent to those skilled in the art that the present application can be practiced without the use of these specific details. In other instances, well known structures and processes are not elaborated upon in order to avoid obscuring the description of the present application. Thus, the present application is not intended to be limited by the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0016] Embodiment one, as shown in the accompanying drawings Figure 1 The present application provides a deep learning driven battery residual capacity dynamic evaluation method, which comprises the following steps: S110: reading the current SOC region where the target battery group is located, and matching to obtain the current prediction error confidence interval of the battery residual capacity prediction; In the embodiments of the present application, in the scenario where the battery residual capacity prediction needs to adapt to different operating states, in order to accurately obtain the prediction error range corresponding to different SOC regions, the SOC region where the target battery group is currently located needs to be determined first, and then the prediction error confidence interval in this region is calculated accordingly, so as to guarantee the accuracy and reliability of subsequent residual capacity evaluation.

[0017] Wherein, the SOC region is a numerical division interval of the battery state of charge, indicating the proportion range of the current residual capacity of the battery to the rated capacity. The difference in electrochemical characteristics of the battery in different intervals will directly affect the error performance of the residual capacity prediction.

[0018] In the method provided by the embodiments of the present application, the SOC region includes a high SOC region, an intermediate SOC region and a low SOC region, wherein the high SOC region is SOC greater than or equal to 80%, the intermediate SOC region is SOC greater than or equal to 20% and less than 80%, and the low SOC region is SOC less than 20%, and each SOC region is identified with a prediction error confidence interval.

[0019] Specifically, first, the current SOC value of the target battery group is read and judged in real time to determine the SOC region where the target battery group is located. For example, when the current SOC of the target battery group is detected to be 88%, it is determined that it is in the high SOC region; if the current SOC is 45%, it corresponds to the intermediate SOC region; if the current SOC is 18%, it belongs to the low SOC region.

[0020] Meanwhile, each SOC region is identified with a prediction error confidence interval to quantify the fluctuation range of the remaining capacity prediction result in the current SOC region.

[0021] In the method provided by the embodiments of the application, the calculation process of the prediction error confidence interval comprises: Randomly selecting any SOC region as the first SOC region from the high SOC region, the intermediate SOC region and the low SOC region; obtaining a current aging index of the target battery pack, and expanding the current aging index according to a preset tolerance interval to obtain an aging index interval, wherein the current aging index is determined based on a historical number of charge and discharge cycles and a historical number of deep discharge times; retrieving historical operation records of the same type of battery pack of the target battery pack to obtain a sample predicted SOC value set and a sample real SOC value set under different operating conditions, with the first SOC region and the aging index interval as constraints; based on the sample predicted SOC value set and the sample real SOC value set, taking the difference between each sample predicted SOC value and the corresponding sample real SOC value as a sample error to obtain a sample error set, and calculating an error mean and an error standard deviation of the sample error set; determining a corresponding standard score according to a preset confidence level, taking the error mean minus the product of the standard score and the error standard deviation as a lower limit of the confidence interval, and taking the error mean plus the product of the standard score and the error standard deviation as an upper limit of the confidence interval to obtain a first prediction error confidence interval of the first SOC region.

[0022] Specifically, first, one of the high SOC region, the intermediate SOC region and the low SOC region is randomly selected as the first SOC region. For example, the intermediate SOC region (20%-80%) is selected as the first SOC region, and the corresponding prediction error confidence interval can be calculated for the SOC region subsequently.

[0023] Further, a current aging index of the target battery pack is obtained, which is evaluated based on a historical number of charge and discharge cycles and a historical number of deep discharge times. For example, if the number of charge and discharge cycles of the target battery pack is 500 times and the number of deep discharge times is 30 times, the current aging index is evaluated to be 0.6. The aging index interval is obtained as 0.56 to 0.64 by expanding according to the preset tolerance interval ±0.04.

[0024] Further, the historical operation records of the same type of battery pack are retrieved to obtain a sample predicted SOC value set and a sample real SOC value set under different operating conditions, with the selected first SOC region and the expanded aging index interval as constraints.

[0025] wherein the operating conditions include different temperatures, different charge-discharge rates, etc., such as retrieving the operating data of the same type of battery under the conditions of 25°C, 1C charge-discharge rate, 30°C, 0.8C charge-discharge rate, etc., in the middle SOC region, and the aging index in the interval of 0.56-0.64, and the sample predicted SOC value and the sample true SOC value under the above conditions are extracted.

[0026] Further, based on the obtained sample data, the sample error is obtained by subtracting the corresponding sample true SOC value from each sample predicted SOC value, and the sample error set is formed by integrating all sample errors.

[0027] For example, the predicted SOC value of a certain group of samples is 60%, the true SOC value is 59.2%, and the corresponding sample error is 0.8%. In addition, the predicted SOC value of another group of samples is 45%, the true SOC value is 45.5%, and the corresponding sample error is -0.5%.

[0028] Similarly, after collecting a plurality of sample errors, the error average and error standard deviation of the sample error set are calculated, such as calculating the error average of 0.3% and the error standard deviation of 0.6%.

[0029] Finally, the corresponding standard score is determined according to the preset confidence level, and the upper and lower limits of the confidence interval are calculated. Specifically, if the preset confidence level is 95% and the corresponding standard score is 1.96, the lower limit of the confidence interval is 0.3%-1.96x0.6%=-0.876%, and the upper limit of the confidence interval is 0.3%+1.96x0.6%=1.476%, thereby obtaining the first prediction error confidence interval of the middle SOC region as [-0.876%, 1.476%].

[0030] Through the above steps, the prediction error confidence interval corresponding to the selected SOC region can be obtained, which provides a clear reference basis for the control of the subsequent remaining capacity prediction.

[0031] Further, after obtaining the first prediction error confidence interval corresponding to the selected SOC region, the interval needs to be accurately matched with the SOC region currently occupied by the target battery pack to determine the current prediction error confidence interval of the battery remaining capacity prediction.

[0032] Specifically, if the first SOC region randomly selected and calculated is consistent with the current SOC region determined in real time by the target battery pack, for example, both are the middle SOC region (20%-80%), and the interval of the aging index of the target battery pack after expansion is consistent with the interval of the aging index used when calculating the first prediction error confidence interval, then the first prediction error confidence interval calculated previously can be directly used as the current prediction error confidence interval of the target battery pack.

[0033] Conversely, if the SOC region in which the target battery pack is currently located is different from the previously selected first SOC region, if the previously calculated is the first prediction error confidence interval of the high SOC region (≥80%), and the target battery pack is currently actually located in the low SOC region (<20%), the above calculation process needs to be re-performed with the low SOC region as the new first SOC region, in combination with the current aging index of the target battery pack and the expanded aging index interval, to search the historical data of the same type of battery pack under different operating conditions in the low SOC region, corresponding to the aging index interval, and to re-calculate the prediction error confidence interval corresponding to the low SOC region. The interval is the current prediction error confidence interval of the target battery pack.

[0034] Through the above matching process, it can be ensured that the current prediction error confidence interval obtained is highly adapted to the actual operating state (current SOC region, current aging state) of the target battery pack, laying a foundation for subsequent adjustment of the remaining capacity prediction strategy based on the error interval and guaranteeing the prediction accuracy.

[0035] S120: Monitor and obtain a multi-source operating data sequence of the target battery pack in a historical time zone, perform operating state stability analysis based on the multi-source operating data sequence, and output a current operating stability coefficient; In the embodiment of the application, in order to accurately analyze the operating stability of the target battery pack, the multi-source operating data sequence in the historical time zone needs to be monitored and obtained first, and then the operating state stability analysis is carried out based on these data and the current operating stability coefficient is outputted, to provide a reliable state basis for the generation of the adaptive prediction depth and the formulation of the adaptive prediction scheme.

[0036] Specifically, first, the multi-source operating data sequence of the target battery pack in the historical time zone is monitored and obtained. The multi-source operating data sequence includes an electrical characteristic data sequence and a load characteristic data sequence, the electrical characteristic data sequence covers a voltage sequence, a current sequence and a power sequence, and the load characteristic data sequence covers a load current sequence and a load power sequence. These two types of sequence data need to be comprehensively collected to avoid incomplete judgment of the operating state due to data missing.

[0037] Further, the index fluctuation is calculated based on the obtained electrical characteristic data sequence. For the voltage sequence, the current sequence and the power sequence, the corresponding fluctuation calculation method is used to obtain the voltage fluctuation coefficient, the current fluctuation coefficient and the power fluctuation coefficient, respectively. Then, according to the influence degree of each coefficient on the overall stability of the electrical characteristics, the corresponding weight is given and weighted summation is performed to obtain the electrical characteristic fluctuation degree, to quantitatively reflect the operating fluctuation of the battery in the electrical layer.

[0038] Meanwhile, the index fluctuation is calculated based on the obtained load characteristic data sequence. For the load current sequence and the load power sequence, the corresponding fluctuation coefficients are obtained through the fluctuation calculation method respectively, and then the load characteristic fluctuation degree is obtained by weighted summation according to the influence weight of each fluctuation coefficient on the stability of the load characteristic, so as to quantitatively reflect the running fluctuation of the battery load level.

[0039] Finally, the current running stability coefficient is determined by weighted evaluation according to the electrical characteristic fluctuation degree and the load characteristic fluctuation degree. In the process of weighted evaluation, reasonable weight distribution is set by combining the attention focus of the battery application scene on the electrical characteristic and the load characteristic. The current running stability coefficient is obtained by weighted summation of the reciprocals of the electrical characteristic fluctuation degree and the load characteristic fluctuation degree, and the numerical value is negatively correlated with the electrical characteristic fluctuation degree and the load characteristic fluctuation degree.

[0040] This step converts the abstract running state of the battery into a quantifiable current running stability coefficient through multi-source running data acquisition and fluctuation degree analysis, which provides a clear numerical reference for subsequent link to adjust the prediction strategy based on the running stability.

[0041] The step S120 in the method provided by the embodiment of the application comprises: monitoring and obtaining a multi-source running data sequence of the target battery pack in a historical time zone, wherein the multi-source running data sequence comprises an electrical characteristic data sequence and a load characteristic data sequence, the electrical characteristic data sequence comprises a voltage sequence, a current sequence and a power sequence, and the load characteristic data sequence comprises a load current sequence and a load power sequence; performing index fluctuation calculation according to the voltage sequence, the current sequence and the power sequence respectively, obtaining a voltage fluctuation coefficient, a current fluctuation coefficient and a power fluctuation coefficient, and performing weighted summation to obtain an electrical characteristic fluctuation degree; performing index fluctuation calculation according to the load current sequence and the load power sequence respectively, and performing weighted summation to obtain a load characteristic fluctuation degree; determining a current running stability coefficient by weighted evaluation according to the electrical characteristic fluctuation degree and the load characteristic fluctuation degree, wherein the current running stability coefficient is negatively correlated with the electrical characteristic fluctuation degree and the load characteristic fluctuation degree.

[0042] In the embodiment of the application, in order to avoid the accuracy deviation or resource waste of the prediction scheme formulated based on the fuzzy state judgment, the historical running data is monitored, the fluctuation index is calculated, and finally the stability coefficient is determined, so as to provide a reliable state basis for subsequent adaptive prediction depth generation and prediction scheme formulation.

[0043] Specifically, first, the multi-source operation data sequence of the target battery pack in the historical time zone is monitored. The setting of the historical time zone needs to be determined in combination with the actual use scene of the battery and the data validity requirement. For example, for the battery pack of an energy storage power station, the past 72 hours can be selected as the historical time zone, which covers the complete charging and discharging cycle of the battery and can also reflect the trend of the short-term operation state.

[0044] In addition, for a new energy vehicle power battery, the time corresponding to the past 5 driving cycles can be selected as the historical time zone to ensure that the data can match the operation characteristics in the actual use of the vehicle.

[0045] At the same time, during the data monitoring process, the integrity and accuracy of the multi-source operation data sequence need to be ensured. The electrical characteristic data sequence needs to continuously collect real-time change data of voltage, current and power to form sequence information in the time dimension, for example, recording the voltage value once every 10 seconds, and continuously for 72 hours to form a voltage sequence containing 25920 data points. The load characteristic data sequence needs to continuously collect load current and load power data to form the corresponding sequence to avoid incomplete judgment of the operation state due to data interruption or loss.

[0046] Further, after obtaining the multi-source operation data sequence, the voltage sequence, current sequence and power sequence in the electrical characteristic data sequence are used to calculate the index volatility respectively. The ratio of standard deviation to mean (coefficient of variation) can be used as the core calculation method for index volatility calculation. For example, for the voltage sequence, the mean and standard deviation of all data points in the sequence are calculated first, and then the voltage fluctuation coefficient is obtained by dividing the standard deviation by the mean. The larger the coefficient, the more violent the voltage fluctuation in the historical time zone.

[0047] Similarly, the current fluctuation coefficient of the current sequence and the power fluctuation coefficient of the power sequence are calculated in the same way. Then, according to the demand weight of the stability of each electrical index in the battery application scene, the three fluctuation coefficients are weighted and summed to obtain the electrical characteristic fluctuation degree.

[0048] For example, in the communication base station backup power supply scene, the voltage stability is crucial for equipment power supply, and the voltage fluctuation coefficient can be given a weight of 40%. The current fluctuation coefficient and the power fluctuation coefficient are given a weight of 30% respectively. If the voltage fluctuation coefficient of a certain base station battery pack is 0.02, the current fluctuation coefficient is 0.03, and the power fluctuation coefficient is 0.025, then the electrical characteristic fluctuation degree is 0.02x40%+0.03x30%+0.025x30%=0.0245. The size of this value directly reflects the operation fluctuation situation in the electrical layer.

[0049] Meanwhile, the load current sequence and the load power sequence in the load characteristic data sequence are respectively subjected to index fluctuation calculation. Specifically, the calculation method is consistent with the electrical characteristic index, and the fluctuation degree is reflected by the coefficient of variation, and the load type is combined to weight the sum of the two indexes to obtain the load characteristic fluctuation degree.

[0050] Exemplarily, in the industrial forklift battery scenario, the stability of the load current directly affects the power output of the forklift, and the load current fluctuation coefficient can be given a weight of 60%, and the load power fluctuation coefficient can be given a weight of 40%. If the load current fluctuation coefficient of a forklift battery is 0.05 and the load power fluctuation coefficient is 0.04, then the load characteristic fluctuation degree is 0.05x60%+0.04x40%=0.046, which quantifies the running fluctuation of the load level.

[0051] Finally, the current running stability coefficient is determined according to the weighted evaluation of the electrical characteristic fluctuation degree and the load characteristic fluctuation degree. Specifically, when performing weighted evaluation, the weights of the electrical characteristics and the load characteristics should be set according to the focus of the battery application scenario. For example, in the energy storage power station scenario, the stability of the electrical characteristics has a greater impact on the grid-connected quality, and the electrical characteristic fluctuation degree can be given a weight of 60% and the load characteristic fluctuation degree can be given a weight of 40%.

[0052] In addition, the current running stability coefficient is calculated by normalizing the reciprocal of the two and then performing weighted summation. Specifically, first, the reciprocal of the electrical characteristic fluctuation degree and the reciprocal of the load characteristic fluctuation degree are mapped to the [0, 1] interval according to "normalization value = reciprocal / historical maximum reciprocal of the same type of battery (value 100)", and then the final coefficient is calculated according to the scene weight.

[0053] Exemplarily, if the electrical characteristic fluctuation degree of a certain energy storage battery pack is 0.0245 and the load characteristic fluctuation degree is 0.03, then the reciprocal of the electrical characteristic fluctuation degree is about 40.82, and the normalized value is 40.82 / 100=0.4082; the reciprocal of the load characteristic fluctuation degree is about 33.33, and the normalized value is 33.33 / 100=0.3333. If the electrical characteristic normalization value is given a weight of 60% and the load characteristic normalization value is given a weight of 40% in the energy storage scenario, then the current running stability coefficient is 0.4082x60%+0.3333x40%≈0.245+0.133≈0.378.

[0054] Since the current running stability coefficient is negatively correlated with the electrical characteristic fluctuation degree and the load characteristic fluctuation degree, if the electrical characteristic fluctuation degree of another battery pack is 0.05 (reciprocal 20) and the load characteristic fluctuation degree is 0.06 (reciprocal 16.67), the reciprocal normalization of the electrical characteristic fluctuation degree is 20 / 100 = 0.2, the reciprocal normalization of the load characteristic fluctuation degree is 16.67 / 100 ≈ 0.1667, and the current running stability coefficient is 0.2*60% + 0.1667*40% ≈ 0.12 + 0.067 ≈ 0.187, which is obviously lower than the former (0.378), indicating that the running state of the battery pack is more unstable.

[0055] Through the above steps, the abstract running state of the battery is converted into the quantifiable current running stability coefficient, avoiding the limitation of single index judgment, and providing a clear numerical reference for subsequent adjustment of the prediction depth and prediction scheme based on the current running stability coefficient.

[0056] S130: compensating the preset standard prediction depth according to the current prediction error confidence interval and the current running stability coefficient to generate an adaptive prediction depth; In the embodiments of the present application, in order to avoid the insufficient accuracy caused by the unified use of the preset standard prediction depth, the standard prediction depth needs to be compensated based on the current prediction error confidence interval and the current running stability coefficient to generate a prediction depth adapted to the actual scene, so as to ensure that the subsequent prediction scheme can reasonably control the resource consumption while ensuring the accuracy.

[0057] Specifically, first, the interval width ratio of the current prediction error confidence interval to the preset standard prediction error confidence interval is taken as the first prediction depth compensation coefficient. The current prediction error confidence interval reflects the error range that the remaining capacity prediction may exist in the current scene, and the preset standard prediction error confidence interval is the benchmark error range set based on a large amount of historical data and general scene. Through the interval width ratio of the two, the deviation degree of the current error situation relative to the benchmark can be quantified.

[0058] Further, the ratio of the preset standard running stability coefficient to the current running stability coefficient is taken as the second prediction depth compensation coefficient. The preset standard running stability coefficient is the benchmark coefficient set for the stable running state of the battery, and the current running stability coefficient reflects the stability degree of the actual running of the battery. The ratio of the two can reflect the difference between the current running stability and the benchmark state.

[0059] Further, the overall prediction depth compensation coefficient is obtained by weighted fusion based on the first prediction depth compensation coefficient and the second prediction depth compensation coefficient. In the weighted fusion process, reasonable weights need to be set in combination with the focus of the battery application scene on error control and running stability, so as to ensure that the overall compensation coefficient can comprehensively reflect the influence of error and running state on the prediction depth.

[0060] Finally, the product of the overall prediction depth compensation coefficient and the preset standard prediction depth is taken as the adaptive prediction depth. The preset standard prediction depth is a basic prediction precision set based on a conventional scene, and through multiplication with the overall compensation coefficient, the final adaptive prediction depth can be dynamically adjusted according to the current error size and running stability.

[0061] This step compensates the prediction depth in two dimensions of error and running state, so that the generated adaptive prediction depth can match the current actual scene, providing a reasonable depth basis for subsequent development of an adaptive prediction scheme and residual capacity prediction.

[0062] The step S130 in the method provided in the embodiments of the present application includes: The interval width ratio of the current prediction error confidence interval to the preset standard prediction error confidence interval is taken as a first prediction depth compensation coefficient; The ratio of the preset standard running stability coefficient to the current running stability coefficient is taken as a second prediction depth compensation coefficient; The overall prediction depth compensation coefficient is obtained by weighted fusion based on the first prediction depth compensation coefficient and the second prediction depth compensation coefficient; The product of the overall prediction depth compensation coefficient and the preset standard prediction depth is taken as the adaptive prediction depth.

[0063] In the embodiments of the present application, in order to make the precision of the battery residual capacity prediction fit the current actual running scene, the prediction depth needs to be adjusted by multi-dimensional compensation coefficient calculation based on the current prediction error and running stability, so as to generate an adaptive prediction depth, providing a basis for subsequent development of a precise and efficient prediction scheme.

[0064] Specifically, first, the interval width ratio of the current prediction error confidence interval to the preset standard prediction error confidence interval is taken as a first prediction depth compensation coefficient. The current prediction error confidence interval is an error fluctuation range calculated based on the current SOC region and aging state of the target battery pack, and the preset standard prediction error confidence interval is a reference error range set based on historical data of a large number of similar batteries under standard working conditions, and the interval width ratio of the two can intuitively reflect the difference between the current error risk and the reference state.

[0065] For example, if the width of the current prediction error confidence interval is 2.5% and the width of the preset standard prediction error confidence interval is 1.5%, the first prediction depth compensation coefficient is 2.5% / 1.5%≈1.67, which is greater than 1, indicating that the current error risk is higher than the reference, and the precision needs to be improved by adjusting the prediction depth to cover the error fluctuation.

[0066] Further, a ratio of the preset standard running stability coefficient and the current running stability coefficient is set as a second prediction depth compensation coefficient. The preset standard running stability coefficient is a reference coefficient set for the battery in a stable running state, and is usually 0.8 (the closer the coefficient is to 1, the more stable the running is). The current running stability coefficient reflects the stability of the actual running of the battery.

[0067] For example, if the current running stability coefficient of the target battery pack is 0.5, which is lower than the preset standard running stability coefficient 0.8, the second prediction depth compensation coefficient is 0.8 / 0.5 = 1.6. Since the second prediction depth compensation coefficient is greater than 1, it indicates that the current running state is unstable, and the prediction depth needs to be deepened to reduce the interference of unstable factors on the prediction result.

[0068] On the contrary, if the current running stability coefficient of the target battery pack is 0.9, which is higher than the preset standard running stability coefficient 0.8, the second prediction depth compensation coefficient is 0.8 / 0.9 ≈ 0.89. Since the coefficient is less than 1, it indicates that the running state is better than the reference, and the prediction depth can be appropriately reduced to save resources.

[0069] Further, after obtaining the first and second prediction depth compensation coefficients, the overall prediction depth compensation coefficient is obtained by weighted fusion based on the two, to comprehensively balance the influence of the current error risk and the running stability on the prediction depth.

[0070] Specifically, the weight setting in the weighted fusion needs to be combined with the focus of the battery application scene on error control and running stability. For example, in the new energy vehicle power battery scene, error control directly affects the judgment of the cruising range, and the first prediction depth compensation coefficient is given a weight of 60%, and the second prediction depth compensation coefficient is given a weight of 40%. In addition, if in the communication base station standby power supply scene, the running stability is more critical to power supply reliability, the first prediction depth compensation coefficient can be adjusted to a weight of 40%, and the second prediction depth compensation coefficient can be adjusted to a weight of 60%.

[0071] For example, assuming that in the new energy vehicle scene, the first prediction depth compensation coefficient is 1.67, and the second prediction depth compensation coefficient is 1.6, the overall prediction depth compensation coefficient is 1.67x60%+1.6x40%=1.642, which comprehensively reflects the common demand of error and running state on the prediction depth.

[0072] Finally, the product of the overall prediction depth compensation coefficient and the preset standard prediction depth is taken as the adaptive prediction depth. The preset standard prediction depth is the basic prediction accuracy set based on the conventional scene, which is usually measured by “data sampling frequency x prediction time window”.

[0073] Exemplarily, the preset standard prediction depth is "sampled once every 5 seconds x predict future 30 minutes". If the overall prediction depth compensation coefficient is 1.642, the adaptive prediction depth is "sampled once every 3 seconds x predict future 49.26 minutes" (the sampling frequency and the prediction window are enlarged by the coefficient to improve the accuracy); if the overall prediction depth compensation coefficient is 0.89, the adaptive prediction depth is "sampled once every 5.6 seconds x predict future 26.7 minutes" (the sampling frequency and the prediction window are reduced by the coefficient to reduce resource consumption).

[0074] Through the above steps, the generated adaptive prediction depth can accurately match the current error risk and running stability, avoiding both the problem of insufficient prediction accuracy when the error is too high and the problem of excessive resource consumption when the running is stable, and laying a reasonable accuracy foundation for subsequent development of an adaptive prediction scheme based on the depth and battery residual capacity prediction.

[0075] S140: Based on the preset prediction index space, an adaptive prediction scheme is formulated according to the current running stability coefficient and the adaptive prediction depth, adaptive multi-source prediction data is collected and acquired according to the adaptive prediction scheme, and a battery residual capacity prediction plug-in based on deep learning is called to perform battery residual capacity prediction on the target battery pack according to the adaptive multi-source prediction data, and output a battery residual capacity evaluation result.

[0076] In the embodiments of the present application, in order to avoid the problems of insufficient prediction accuracy or resource waste caused by using fixed indicators and single models, an adaptive scheme needs to be formulated first relying on preset index space and key parameters, and then data is collected and a deep learning plug-in is called to complete prediction according to the scheme, so as to ensure that the final evaluation result is both accurate and can meet the efficiency requirements of actual application.

[0077] Specifically, first, the preset prediction index space of the target battery pack is acquired, which covers all prediction indicators related to residual capacity prediction. Then, the historical running records of similar battery packs are combined, and the relevance of each prediction indicator and the prediction result in the space is evaluated one by one with the current SOC area as the constraint, the index relevance of each prediction indicator is obtained, and the indicators are sorted in descending order of relevance to generate a prediction index sequence.

[0078] Further, according to the adaptive prediction depth, the number of adaptive indicators is determined by referring to the preset prediction depth-indicator number mapping table, a corresponding number of indicators are selected from the prediction index sequence to form an adaptive prediction indicator set.

[0079] Meanwhile, the number of selected models is calculated: the ratio of the current running smooth coefficient to the average historical running smooth coefficient of the target battery group in the historical time range is multiplied by the standard number of selected models, and the result is rounded off, and it is ensured that the number is not less than 2 and not greater than the total number of models in the battery residual capacity prediction plug-in. The adaptive prediction index set and the number of selected models are used as an adaptive prediction scheme.

[0080] Further, after determining the adaptive prediction scheme, the adaptive multi-source prediction data corresponding to the target battery group is collected according to the adaptive prediction index set in the adaptive prediction scheme, so as to ensure that the data can completely cover the information of the selected index.

[0081] Further, according to the adaptive prediction index set, the adaptive battery capacity prediction plug-in is matched and called, and the adaptive battery capacity prediction plug-in contains not less than 10 battery capacity prediction models. The models are randomly selected from the adaptive battery capacity prediction plug-in according to the number of selected models, and the collected adaptive multi-source prediction data is input into these models for residual capacity prediction. Finally, the prediction results output by the multiple models are fitted by mean value, and the battery residual capacity evaluation result is obtained and output.

[0082] This step adjusts the prediction index and the number of models dynamically, combines the multi-model prediction ability of the deep learning plug-in, ensures the adaptability of the prediction process to the actual scene, improves the reliability of the evaluation result through the mean value fitting of multiple results, and provides a complete and efficient implementation path for the accurate judgment of the battery residual capacity.

[0083] The step S140 in the method provided by the embodiment of the application includes: obtaining a preset prediction index space of the target battery group, wherein the preset prediction index space includes a plurality of prediction indexes; based on the historical operation records of the same type battery group of the target battery group, evaluating the relevance of the plurality of prediction indexes and the battery residual capacity prediction result respectively with the current SOC region as a constraint, to obtain a plurality of index correlation degrees; arranging the plurality of prediction indexes in descending order of the index correlation degrees to generate a prediction index sequence; determining the number of adaptive indexes according to the adaptive prediction depth based on a preset prediction depth-index number mapping table, and selecting the prediction indexes of the number of adaptive indexes in the prediction index sequence as an adaptive prediction index set; the ratio of the current running smooth coefficient to the average historical running smooth coefficient of the target battery group in the historical time range is multiplied by the standard number of selected models and rounded off to obtain the number of selected models, wherein the standard number of selected models is 5, the number of selected models is greater than or equal to 2 and less than or equal to the total number of models in the battery residual capacity prediction plug-in. The adaptive prediction index set and the adaptive selection model number are taken as an adaptive prediction scheme.

[0084] Adaptive multi-source prediction data of the target battery pack are collected according to the adaptive prediction index set; An adaptive battery capacity prediction plug-in is matched and called according to the adaptive prediction index set, wherein the adaptive battery capacity prediction plug-in includes N battery capacity prediction models, and N is greater than or equal to 10; The same number of prediction models are randomly selected from the N battery capacity prediction models of the adaptive battery capacity prediction plug-in according to the adaptive selection model number, battery residual capacity prediction is respectively performed according to the adaptive multi-source prediction data, and mean fitting is performed on a plurality of prediction results to obtain a battery residual capacity evaluation result.

[0085] In the embodiments of the present application, in order to accurately match the index selection and model use of battery residual capacity prediction with the current running state and prediction demand, the adaptive prediction index is screened through a preset prediction index space, the model number is determined in combination with a running smoothness coefficient, and then data is collected and an adaptive battery capacity prediction plug-in is called for prediction according to an adaptive prediction scheme, so as to ensure that the evaluation result is accurate and efficient.

[0086] Specifically, first, a preset prediction index space of a target battery pack is obtained. The preset prediction index space sets prediction indexes from three dimensions of the battery pack as a whole, individual cells, and external environment, which can comprehensively cover key factors affecting residual capacity prediction.

[0087] In the method provided in the embodiments of the present application, the plurality of prediction indexes include group-level parameters, individual cell parameters, and environmental parameters, wherein the target battery pack includes a plurality of individual cells, the group-level parameters at least include terminal voltage and total current of the battery pack, the individual cell parameters at least include cell voltage, pole temperature, and internal resistance of the individual cell, and the environmental parameters at least include environmental temperature and environmental humidity.

[0088] The group-level parameters are from the macroscopic level of the battery pack, the terminal voltage of the battery pack is directly related to the overall energy reserve state, the total current reflects the energy output intensity, and the combination of the two can quickly judge the basic running performance of the battery pack.

[0089] In addition, the individual cell parameters focus on the characteristics of individual cells, the voltage difference of individual cells can reflect the consistency of the battery cell, the pole temperature can warn the local overheating risk, and the internal resistance of the battery cell reflects the aging degree of the battery cell.

[0090] In addition, the environmental parameters focus on external influencing factors. The ambient temperature can significantly change the chemical reaction rate of the battery, and the ambient humidity can affect the insulation performance and service life of the battery. Incorporating these two types of parameters can make the prediction model more suitable for actual use scenarios. The three types of parameters complement each other from different dimensions to form a comprehensive and accurate prediction index basis.

[0091] Further, based on the historical operation records of the same type of battery pack as the target battery pack, and with the current SOC region as a constraint, the relevance of a plurality of prediction indicators and the battery remaining capacity prediction results is evaluated respectively to obtain a plurality of indicator relevance degrees.

[0092] Among them, the same type of battery pack refers to a battery pack with the same model number as the target battery pack and a similar use scenario. Its historical operation records include indicator data and corresponding real remaining capacity values at different time nodes. In addition, the current SOC region is used as a constraint because the influence of the same indicator on the remaining capacity varies at different SOC regions. For example, in the low SOC region (<20%), the single battery internal resistance has a more significant impact on the remaining capacity, while in the high SOC region (≥80%), the relevance of the battery pack terminal voltage is stronger.

[0093] Specifically, the relevance degree is determined by calculating the correlation coefficient (i.e., the Pearson correlation coefficient) of each indicator and the real remaining capacity during the evaluation. The closer the absolute value of the correlation coefficient is to 1, the stronger the relevance of the indicator to the prediction result. If the evaluation in the intermediate SOC region (20%-80%) shows that the correlation coefficient of the battery pack terminal voltage and the remaining capacity is 0.91, the correlation coefficient of the single battery voltage is 0.87, and the correlation coefficient of the ambient temperature is 0.73, the corresponding indicator relevance degrees are 0.91, 0.87, and 0.73, respectively.

[0094] Further, the plurality of prediction indicators are arranged in descending order of indicator relevance degree to generate a prediction indicator sequence. Continuing the evaluation results in the above-mentioned intermediate SOC region, if the relevance degrees of all indicators from high to low are "battery pack terminal voltage (0.91) → single battery voltage (0.87) → ambient temperature (0.73) → pole temperature (0.68) → battery pack total current (0.65) → single battery internal resistance (0.62) → ambient humidity (0.58)", the prediction indicator sequence is the above-mentioned order to ensure that the indicators with stronger relevance and greater impact on prediction are placed in the front of the sequence.

[0095] Further, based on a pre-set prediction depth-indicator number mapping table, the number of adaptive indicators is determined according to the adaptive prediction depth matching, and the first adaptive number of prediction indicators in the prediction indicator sequence are selected as the adaptive prediction indicator set.

[0096] The preset prediction depth-index quantity mapping table is formulated in combination with a large number of prediction experiments, and the deeper the adaptive prediction depth, the higher the prediction accuracy required, and the more the matching index quantity. For example, when the adaptive prediction depth is marked as depth 1 (the highest accuracy requirement), 6 indexes are matched; when the depth is 2, 4 indexes are matched; and when the depth is 3 (the basic accuracy requirement), 2 indexes are matched.

[0097] For example, if the current adaptive prediction depth is depth 2, the first 4 indexes in the above sequence, i.e., the battery pack terminal voltage, the single battery voltage, the ambient temperature, and the pole temperature, are selected to form the adaptive prediction index set, which not only ensures that the indexes cover the key influencing factors, but also avoids excessive redundant indexes to increase the data acquisition cost.

[0098] Meanwhile, the ratio of the current running smoothness coefficient to the historical running smoothness coefficient average of the target battery pack in the historical time range is multiplied by the standard selected model quantity and rounded to obtain the adaptive selected model quantity.

[0099] The historical time range can be set to the past 30 days, and the average of the daily running smoothness coefficients in this period is calculated as the historical running smoothness coefficient average. The standard selected model quantity is fixed at 5, and the adaptive selected model quantity needs to satisfy ≥ 2, ≤ the total number of models in the battery remaining capacity prediction plug-in (N ≥ 10).

[0100] For example, if the current running smoothness coefficient is 0.5 and the historical running smoothness coefficient average is 0.4, the ratio of the two is 1.25, which is multiplied by 5 to obtain 6.25, and rounded to 6. If the total number of models in the plug-in is 12 (satisfying N ≥ 10), 6 is within the range of 2-12, so the adaptive selected model quantity is 6. If the current running smoothness coefficient is 0.2 and the historical average is 0.5, the ratio is 0.4, which is multiplied by 5 to obtain 2, and rounded to 2, i.e., the adaptive selected model quantity is 2, to ensure that the model quantity can support multiple result fitting to improve accuracy, and will not waste computing resources due to too many models.

[0101] Further, the adaptive prediction index set and the adaptive selected model quantity are used as the adaptive prediction scheme, which clearly defines the range of subsequent data collection and the number of prediction models used, providing clear guidance for the prediction process.

[0102] Further, after determining the adaptive prediction scheme, adaptive multi-source prediction data of the target battery pack is collected according to the adaptive prediction index set. Taking the adaptive prediction index set of "battery pack terminal voltage, single battery voltage, ambient temperature, and pole temperature" as an example, the terminal voltage data of the battery pack (recorded every 5 seconds), the voltage data of each single battery (recorded every 10 seconds), the pole temperature (recorded every 30 seconds) and the ambient temperature (recorded every 1 minute) are collected in real time through the battery management system (BMS) and the temperature sensor, so as to ensure that the collected data correspond to the adaptive prediction index one by one, and the data sampling frequency can meet the prediction accuracy requirement.

[0103] Further, the adaptive battery capacity prediction plug-in is matched and called according to the adaptive prediction index set, and the data features are accurately extracted through the model adapted to the adaptive prediction index set in the plug-in, so as to ensure that the subsequent prediction result can meet the actual demand.

[0104] In the method provided by the embodiment of the application, the construction method of the adaptive battery capacity prediction plug-in includes: The historical running records of the same type of battery pack are searched as a constraint according to the adaptive prediction index set, the sample multi-source data set is collected, and the historical battery residual capacity corresponding to different sample multi-source data at the same time is taken as a sample battery residual capacity to obtain a sample battery residual capacity set; The sample multi-source data set and the sample battery residual capacity set are taken as training data, the training data is equally divided into N parts, and is randomly selected with replacement to obtain N sample training sets; The N sample training sets are used to train the deep learning model to convergence respectively to obtain N battery capacity prediction models, and the adaptive battery capacity prediction plug-in is combined and constructed.

[0105] Specifically, first, the historical running records of the same type of battery pack are searched as a constraint according to the adaptive prediction index set, the sample multi-source data set is collected, and the sample battery residual capacity set is obtained. The same type of battery pack is consistent with the type of the target battery pack and has the same use scene, and the historical running records contain long-term accumulated index data and real residual capacity values at corresponding time nodes.

[0106] In addition, the constraint of the adaptive prediction index set means that only data related to the adaptive prediction index is collected. For example, if the adaptive prediction index set is "battery pack terminal voltage, single battery voltage, and ambient temperature", only the time series data of these three types of indexes is extracted from the historical running records, and irrelevant indexes such as pole temperature and ambient humidity are excluded to avoid redundant data interference with model training.

[0107] Specifically, when collecting, the time continuity and correspondence of the sample data should be ensured. The "terminal voltage 12.5V, single battery voltage 3.1V, and environment temperature 25℃" collected at a certain time are taken as a set of sample multi-source data, and the real residual capacity 85% detected by professional equipment at the same time is recorded as the corresponding sample battery residual capacity. In this way, thousands or even tens of thousands of sets of data are accumulated to form a sample multi-source data set and a sample battery residual capacity set, so as to ensure that the amount of training data can support the association between the model learning index and the residual capacity.

[0108] Further, the sample multi-source data set and the sample battery residual capacity set are taken as training data, the training data is equally divided into N parts, and is randomly selected with replacement to obtain N sample training sets.

[0109] Wherein, the value of N should satisfy ≥10, for example, set N=12, first divide all the training data (assuming a total of 12000 groups) into 12 parts, each containing 1000 groups of "sample multi-source data-residual capacity" corresponding data.

[0110] Further, 12 sample training sets are generated by random selection with replacement (i.e. Bootstrap sampling), wherein each training set still randomly selects 12000 groups of data from the original 12000 groups of data (allowing repeated selection), so as to ensure that each training set not only retains the overall distribution characteristics of the original data, but also has certain differences. The models trained based on different training sets can cover more data scenarios and improve the generalization ability of the plug-in as a whole.

[0111] For example, the first training set may contain more sample data in the low SOC area, and the second training set contains more sample data in the high SOC area, so as to avoid the model bias caused by uneven data distribution in a single training set.

[0112] Finally, the deep learning model is trained based on the N sample training sets respectively until convergence, and N battery capacity prediction models (N is greater than or equal to 10) are obtained, which are combined to build an adaptive battery capacity prediction plug-in to accurately capture the dynamic association between the adaptive prediction index and the residual capacity.

[0113] Specifically, the long short-term memory network (LSTM) is used as the basic framework of the deep learning model. Through the synergistic effect of the input gate, the forgetting gate, and the output gate, the time sequence data characteristics of the adaptive prediction index can be effectively processed, and the gradient disappearance problem of the traditional recurrent neural network (RNN) can be avoided.

[0114] In the network architecture building, the sample data characteristics and the prediction accuracy requirements need to be combined to reasonably set the network layer number, hidden layer neuron number, input and output dimension and other parameters. For example, if the prediction index set contains 3 types of indexes (terminal voltage, single cell voltage, and environmental temperature), the input dimension of the LSTM network is set to 3, corresponding to receiving 3 index data at a single time step; the hidden layer is set to 2 layers, and the number of neurons in each layer is set to 64, which not only ensures that the network can fully learn the characteristics, but also avoids the complexity of the architecture leading to low training efficiency; the output dimension is set to 1, directly outputting the residual capacity prediction value at the corresponding time.

[0115] Meanwhile, a fully connected layer and a Dropout layer (Dropout probability set to 0.2) are added after the LSTM layer. The fully connected layer is used to integrate the feature vector output by the LSTM layer, and the Dropout layer is used to prevent overfitting during training, further improving the model generalization ability.

[0116] In the model training phase, for N sample training sets, each training set uses an independent training process. First, the sample data is preprocessed, i.e., the sample multi-source data is mapped to the [0, 1] interval according to the Min-Max normalization method, eliminating the influence of different index dimension differences on training; the preprocessed time series data is divided into input sequences according to the "time window = 20", i.e., each input contains 20 consecutive time steps of index data, corresponding to the residual capacity value at the end of the time window, so that the model learns the mapping relationship from the index time series change to the residual capacity.

[0117] In addition, during specific training, the root mean square error (RMSE) is selected as the loss function to quantify the difference between the model prediction value and the true value of the sample battery residual capacity. The Adam optimizer (initial learning rate set to 0.001) is used to update the network parameters, the batch size is set to 32, the training rounds are set to 100 rounds, and the early stopping mechanism (Patience = 10) is introduced. If the RMSE on the validation set does not decrease for 10 consecutive rounds and the prediction accuracy of the model on the test set does not improve, the training is stopped and the model is determined to be converged.

[0118] Further, after completing the independent training of N sample training sets according to the above process, N converged LSTM battery capacity prediction models are obtained. Each model has different prediction performance in different working conditions due to the subtle differences in training data.

[0119] Finally, the N LSTM models obtained are integrated and packaged to build a battery capacity prediction plug-in. The plug-in sets a model calling interface, which can select the number of models according to subsequent adaptation, and randomly selects the corresponding number of models to participate in prediction. At the same time, the built-in data preprocessing module can automatically standardize and time window divide the input adaptive multi-source prediction data, without additional data processing operations, ensuring the convenience of plug-in calling.

[0120] Through the above construction process, the adaptive battery capacity prediction plug-in can not only accurately match the characteristics of the adaptive prediction index set, but also improve the reliability and generalization ability of prediction by relying on multi-model design, to ensure that the plug-in can output high-quality remaining capacity prediction results based on adaptive multi-source prediction data when called subsequently.

[0121] Further, the same number of prediction models are randomly selected from the N battery capacity prediction models of the adaptive battery capacity prediction plug-in according to the number of adaptive selected models, and battery remaining capacity prediction is performed according to the adaptive multi-source prediction data, and the mean fitting of multiple prediction results is performed to obtain the battery remaining capacity evaluation result.

[0122] Specifically, when selecting battery capacity prediction models, the corresponding number of models is extracted from the N (for example, 12) LSTM models through the random selection algorithm built-in the adaptive battery capacity prediction plug-in based on the number of adaptive selected models (for example, 6 calculated in the foregoing), and the probability of each model being selected is equal each time to avoid distortion of the prediction results due to model selection bias.

[0123] For example, if the number of adaptive selected models is 6, the adaptive battery capacity prediction plug-in will randomly extract the models numbered 2, 5, 7, 9, 10, and 11 from the 12 LSTM models, which are trained based on different sample training sets.

[0124] Among them, some models have better prediction accuracy in the high SOC area (≥80%) (error <1.8%), some models are sensitive to environmental temperature fluctuations (error <2.2% in low temperature environment), and some models are good at capturing capacity changes caused by subtle differences in single cell voltage. The combination of multiple models can cover the current working condition scenarios that the target battery pack may face.

[0125] Further, after obtaining the selected battery capacity prediction models, the preprocessed adaptive multi-source prediction data is input into each model for independent prediction to avoid bias caused by single model perspective limitations.

[0126] Specifically, first, the data preprocessing module built-in the adaptive battery capacity prediction plug-in will perform Min-Max standardization on the collected adaptive multi-source prediction data, which is consistent with the model training phase, to map the data to the [0, 1] interval.

[0127] Further, the standardized time series data is divided into input sequences according to the "time window = 20", for example, if 40 consecutive time steps of index data are collected, 2 groups of input sequences (1-20 steps, 21-40 steps) are generated, each sequence corresponds to an output of a remaining capacity prediction value, and the average of the two prediction values is taken as the final prediction result of the model.

[0128] Taking the battery capacity prediction model numbered 5 as an example, after inputting two sets of time series data, the output prediction values are 82.3% and 81.9% respectively, and the final prediction result of the model is (82.3%+81.9%) / 2=82.1%; the battery capacity prediction model numbered 7 outputs prediction values of 81.8% and 82.4% respectively, and the final prediction result is 82.1%; the battery capacity prediction model numbered 9 outputs prediction values of 82.5% and 82.0% respectively, and the final prediction result is 82.25%, and so on. The six models will output their respective final prediction results, which are assumed to be 82.1%, 81.9%, 82.1%, 82.25%, 81.8%, and 82.0% respectively.

[0129] Further, after obtaining the prediction results of all selected models, the results are calculated by mean fitting. Specifically, before fitting, possible outliers are first removed, and then the effective prediction results are summed and divided by the number of models to obtain the final battery remaining capacity evaluation result. If the prediction result of a model deviates from other results by more than 5% (based on the statistical threshold of prediction error of similar batteries), it is determined to be an outlier and is removed, and if there is no outlier, all are involved in the calculation.

[0130] Taking the prediction results of the above six models as an example, the sum is: 82.1%+81.9%+82.1%+82.25%+81.8%+82.0%=492.15%, divided by 6 to get 82.025%, rounded to two decimal places is 82.03%, which is the remaining capacity evaluation result of the target battery pack.

[0131] Through the above steps, the differentiated advantages of multiple models can cover the prediction needs under different working conditions, and the random errors of a single model can be offset by mean fitting, so that the remaining capacity evaluation result is closer to the true value. At the same time, a model is randomly selected for each prediction to avoid long-term deviation of a fixed model combination, ensuring that stable and accurate remaining capacity evaluation results can be output regardless of whether the target battery pack is in a high / low SOC area or a high / low temperature environment, providing reliable data support for battery operation and maintenance, range judgment and other practical applications.

[0132] The above specific embodiments achieve the following technical effects: The application provides a deep learning driven battery residual capacity dynamic evaluation method. First, the current SOC region of a target battery pack is read, a corresponding current prediction error confidence interval is matched and acquired, and the fluctuation range of the prediction error in the current scenario is determined. Then, the multi-source operation data sequence in the historical time zone is monitored and acquired, the fluctuation degrees of electrical characteristics and load characteristics are analyzed, and the current operation stability coefficient is output. Then, the current prediction error confidence interval and the current operation stability coefficient are combined to compensate the preset standard prediction depth, and the prediction depth adapted to the current scenario is generated. Then, based on the preset prediction index space, the adaptive prediction index set is screened and the adaptive model number is determined to form an adaptive prediction scheme, and an adaptive battery capacity prediction plug-in including N LSTM models is constructed. Finally, the adaptive multi-source prediction data is collected according to the adaptive prediction scheme, the model randomly selected from the adaptive battery capacity prediction plug-in is called for prediction, the mean value of the prediction result is fitted, and the battery residual capacity evaluation result is output, so that the accurate and dynamic evaluation of the battery residual capacity of the target battery pack in different SOC regions and different operation stability scenarios is realized.

[0133] The method provided by the application solves the problems in the traditional battery residual capacity evaluation, such as information redundancy or loss caused by fixed indexes, difficulty of a single model in adapting to complex working conditions, and mismatch between prediction depth and actual scenario, and improves the accuracy and dynamic adaptability of battery residual capacity evaluation, thereby providing technical support for battery operation and maintenance management and endurance judgment in new energy vehicle and energy storage power station scenarios.

[0134] Embodiment two, as shown in the accompanying Figure 2 Based on the inventive concept of the deep learning driven battery residual capacity dynamic evaluation method provided in embodiment one, the application further provides a deep learning driven battery residual capacity dynamic evaluation system, which specifically comprises: An SOC and error reading module 01 is configured to read the current SOC region of a target battery pack, and match and acquire a current prediction error confidence interval for battery residual capacity prediction. A multi-source data stability analysis module 02 is configured to monitor and acquire a multi-source operation data sequence of the target battery pack in a historical time zone, perform operation state stability analysis according to the multi-source operation data sequence, and output a current operation stability coefficient. A prediction depth compensation module 03 is configured to compensate a preset standard prediction depth according to the current prediction error confidence interval and the current operation stability coefficient, and generate an adaptive prediction depth. The prediction scheme execution module 04 is configured to formulate an adaptive prediction scheme based on a preset prediction index space according to the current running stationary coefficient and an adaptive prediction depth, collect adaptive multi-source prediction data according to the adaptive prediction scheme, and call a battery residual capacity prediction plug-in constructed based on deep learning to perform battery residual capacity prediction on the target battery pack according to the adaptive multi-source prediction data, and output a battery residual capacity evaluation result.

[0135] In one embodiment, the SOC and error reading module 01 is further configured to: The SOC region includes a high SOC region, a middle SOC region, and a low SOC region, wherein the high SOC region is SOC greater than or equal to 80%, the middle SOC region is SOC greater than or equal to 20% and less than 80%, and the low SOC region is SOC less than 20%, and each SOC region is identified with a prediction error confidence interval.

[0136] Further, the SOC and error reading module 01 further includes: In the high SOC region, the middle SOC region, and the low SOC region, any SOC region is randomly selected as a first SOC region; the current aging index of the target battery pack is obtained, and the current aging index is expanded according to a preset tolerance interval to obtain an aging index interval, wherein the current aging index is determined based on historical charge and discharge cycle times and historical deep discharge times; the historical running records of the target battery pack and similar battery packs are retrieved as constraints to obtain a sample predicted SOC value set and a sample true SOC value set under different running conditions; based on the sample predicted SOC value set and the sample true SOC value set, the difference between each sample predicted SOC value and the corresponding sample true SOC value is taken as a sample error to obtain a sample error set, and the error average and error standard deviation of the sample error set are calculated; a corresponding standard score is determined according to a preset confidence level, the lower limit of the confidence interval is obtained by subtracting the product of the standard score and the error standard deviation from the error average, and the upper limit of the confidence interval is obtained by adding the product of the standard score and the error standard deviation to the error average, to obtain a first prediction error confidence interval of the first SOC region.

[0137] In one embodiment, the multi-source data stationary analysis module 02 is further configured to: monitoring a plurality of source operation data sequences of the target battery pack in a historical time zone, wherein the plurality of source operation data sequences include electrical characteristic data sequences and load characteristic data sequences, the electrical characteristic data sequences include voltage sequences, current sequences and power sequences, and the load characteristic data sequences include load current sequences and load power sequences; performing index fluctuation calculation on the voltage sequences, current sequences and power sequences respectively to obtain voltage fluctuation coefficients, current fluctuation coefficients and power fluctuation coefficients, and performing weighted summation to obtain an electrical characteristic fluctuation degree; performing index fluctuation calculation on the load current sequences and load power sequences respectively, and performing weighted summation to obtain a load characteristic fluctuation degree; and performing weighted evaluation on the electrical characteristic fluctuation degree and the load characteristic fluctuation degree to determine a current operation stability coefficient, wherein the current operation stability coefficient is negatively correlated with the electrical characteristic fluctuation degree and the load characteristic fluctuation degree.

[0138] In one embodiment, the prediction depth compensation module 03 is further configured to: determine a first prediction depth compensation coefficient based on a ratio of an interval width of the current prediction error confidence interval to a preset standard prediction error confidence interval, determine a second prediction depth compensation coefficient based on a ratio of a preset standard operation stability coefficient to the current operation stability coefficient, perform weighted fusion based on the first prediction depth compensation coefficient and the second prediction depth compensation coefficient to obtain an overall prediction depth compensation coefficient, and determine an adaptive prediction depth based on a product of the overall prediction depth compensation coefficient and a preset standard prediction depth.

[0139] In one embodiment, the prediction scheme execution module 04 is further configured to: obtaining a preset prediction index space of the target battery pack, wherein the preset prediction index space comprises a plurality of prediction indexes; based on historical operation records of the same type battery pack as the target battery pack, evaluating the relevance of the plurality of prediction indexes and the battery residual capacity prediction result respectively with the current SOC region as a constraint to obtain a plurality of index correlation degrees; arranging the plurality of prediction indexes in descending order of index correlation degree to generate a prediction index sequence; based on a preset prediction depth-index number mapping table, determining an adaptive index number according to the adaptive prediction depth matching, and selecting the first adaptive index number of prediction indexes in the prediction index sequence as an adaptive prediction index set; multiplying the ratio of the current operation smoothness coefficient and the average historical operation smoothness coefficient of the target battery pack within a historical time range by a standard selection model number and taking the integer part to obtain an adaptive selection model number, wherein the standard selection model number is 5, the adaptive selection model number is greater than or equal to 2 and less than or equal to the total number of models in the battery residual capacity prediction plug-in; taking the adaptive prediction index set and the adaptive selection model number as an adaptive prediction scheme. According to the adaptive prediction index set, the adaptive multi-source prediction data of the target battery pack is collected; according to the adaptive prediction index set, the adaptive battery capacity prediction plug-in is matched and called, wherein the adaptive battery capacity prediction plug-in comprises N battery capacity prediction models, and N is greater than or equal to 10; the same number of prediction models are randomly selected from the N battery capacity prediction models of the adaptive battery capacity prediction plug-in according to the adaptive selection model number, battery residual capacity prediction is performed according to the adaptive multi-source prediction data, and the mean value fitting is performed on the plurality of prediction results to obtain the battery residual capacity evaluation result.

[0140] Further, the prediction scheme execution module 04 further comprises: The plurality of prediction indexes comprise group-level parameters, single-cell parameters and environmental parameters, wherein the target battery pack comprises a plurality of single-cell batteries, the group-level parameters at least comprise terminal voltage and total current of the battery pack, the single-cell parameters at least comprise battery voltage, pole temperature and battery internal resistance of the single-cell battery, and the environmental parameters at least comprise environmental temperature and environmental humidity.

[0141] Further, the prediction scheme execution module 04 further comprises: Retrieving the historical operation records of the same type battery pack with the adaptive prediction index set as a constraint, collecting a sample multi-source data set, and taking the historical battery residual capacity corresponding to different sample multi-source data at the same time as a sample battery residual capacity to obtain a sample battery residual capacity set; taking the sample multi-source data set and the sample battery residual capacity set as training data, dividing the training data into N parts, and randomly selecting with replacement to obtain N sample training sets; The N battery capacity prediction models are obtained by training the deep learning model to convergence respectively using the N sample training sets, and a battery capacity prediction plug-in is constructed by combination.

[0142] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the above describes a specific embodiment of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0143] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0144] The present application and the drawings are only exemplary descriptions of the present application, and any and all modifications, changes, combinations or equivalents within the scope of the present application are considered to have been covered. Obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope of the present application. Thus, if these modifications and changes of the present application belong to the scope of the present application and its equivalent technology, the present application intends to include these modifications and changes.

Claims

1. A deep learning-driven method for dynamic evaluation of remaining battery capacity, characterized in that the method... include: Read the current SOC region of the target battery pack and match it to obtain the current prediction error confidence interval of the remaining battery capacity prediction; The target battery pack is monitored and acquired in a historical time zone using multi-source operating data sequences. Based on the multi-source operating data sequences, an operational stability analysis is performed, and the current operational stability coefficient is output. Based on the current prediction error confidence interval and the current operational stability coefficient, the preset standard prediction depth is compensated to generate an adapted prediction depth; Based on the preset prediction index space, an adaptation prediction scheme is formulated according to the current operating stability coefficient and the adaptation prediction depth. Adaptation multi-source prediction data is collected according to the adaptation prediction scheme, and a battery remaining capacity prediction plugin built based on deep learning is called. The remaining battery capacity of the target battery pack is predicted according to the adaptation multi-source prediction data, and the battery remaining capacity evaluation result is output.

2. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 1, characterized in that, The SOC region includes a high SOC region, an intermediate SOC region, and a low SOC region. The high SOC region is defined as having an SOC greater than or equal to 80%, the intermediate SOC region is defined as having an SOC greater than or equal to 20% and less than 80%, and the low SOC region is defined as having an SOC less than 20%. Each SOC region is also marked with a prediction error confidence interval.

3. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 2, characterized in that, The calculation process for the prediction error confidence interval includes: Randomly select any one of the high SOC region, intermediate SOC region and low SOC region as the first SOC region; The current aging index of the target battery pack is obtained, and the current aging index is expanded according to a preset tolerance range to obtain an aging index range, wherein the current aging index is determined based on the historical charge-discharge cycle count and the historical deep discharge count. Using the first SOC region and aging index range as constraints, retrieve historical operating records of similar battery packs of the target battery pack to obtain a set of predicted SOC values ​​and a set of actual SOC values ​​under different operating conditions. Based on the set of predicted SOC values ​​and the set of actual SOC values, the difference between the predicted SOC value and the corresponding actual SOC value of each sample is taken as the sample error, and the set of sample errors is obtained. The mean error and standard deviation of the set of sample errors are then calculated. The corresponding standard score is determined according to the preset confidence level. The lower limit of the confidence interval is obtained by subtracting the product of the standard score and the standard deviation of the error from the average error. The upper limit of the confidence interval is obtained by adding the product of the standard score and the standard deviation of the error to the average error. The first prediction error confidence interval of the first SOC region is obtained.

4. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 1, characterized in that, The system monitors and acquires multi-source operational data sequences of the target battery pack within a historical time zone. Based on these multi-source operational data sequences, it performs operational stability analysis and outputs the current operational stability coefficient, including: The target battery pack is monitored and acquired in a historical time zone using multi-source operating data sequences, wherein the multi-source operating data sequences include electrical characteristic data sequences and load characteristic data sequences. The electrical characteristic data sequences include voltage sequences, current sequences, and power sequences, and the load characteristic data sequences include load current sequences and load power sequences. The index volatility is calculated based on the voltage sequence, current sequence, and power sequence to obtain the voltage fluctuation coefficient, current fluctuation coefficient, and power fluctuation coefficient, and then the electrical characteristic volatility is obtained by weighted summation. The load characteristic fluctuation is obtained by calculating the index fluctuation based on the load current sequence and the load power sequence, and then summing them by weight. The current operational stability coefficient is determined based on a weighted evaluation of the electrical characteristic fluctuation and the load characteristic fluctuation, wherein the current operational stability coefficient is negatively correlated with the electrical characteristic fluctuation and the load characteristic fluctuation.

5. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 1, characterized in that, Based on the current prediction error confidence interval and the current operational stability coefficient, the preset standard prediction depth is compensated to generate an adapted prediction depth, including: The ratio of the width of the current prediction error confidence interval to the width of the preset standard prediction error confidence interval is used as the first prediction depth compensation coefficient. The ratio of the preset standard operating stability coefficient to the current operating stability coefficient is set as the second prediction depth compensation coefficient; The overall prediction depth compensation coefficient is obtained by weighted fusion of the first prediction depth compensation coefficient and the second prediction depth compensation coefficient. The product of the overall prediction depth compensation coefficient and the preset standard prediction depth is used as the adaptive prediction depth.

6. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 1, characterized in that, Based on a preset prediction index space, an adaptation prediction scheme is formulated according to the current operational stability coefficient and the adaptation prediction depth, including: Obtain a preset prediction index space for the target battery pack, wherein the preset prediction index space includes several prediction indices. Based on the historical operating records of similar battery packs of the target battery pack, and with the current SOC region as a constraint, the correlation between the prediction results of the several prediction indicators and the remaining battery capacity is evaluated to obtain the correlation degree of the several indicators. The prediction indicators are arranged in descending order of their correlation to generate a prediction indicator sequence. Based on the preset prediction depth-index quantity mapping table, the number of adaptive indicators is determined according to the adaptive prediction depth matching, and the predicted indicators of the previous number of adaptive indicators are selected from the prediction indicator sequence as the adaptive prediction indicator set. The ratio of the current operating stability coefficient to the average historical operating stability coefficient of the target battery pack within the historical time range is multiplied by the number of standard selected models and rounded to obtain the number of adaptive selected models. The number of standard selected models is 5, and the number of adaptive selected models is greater than or equal to 2 and less than or equal to the total number of models in the battery remaining capacity prediction plugin. The set of adaptation prediction indicators and the number of adaptation selection models are used as the adaptation prediction scheme.

7. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 6, characterized in that, The predicted parameters include group-level parameters, individual cell parameters, and environmental parameters. The target battery pack includes multiple individual cells. The group-level parameters include at least the battery pack's terminal voltage and total current. The individual cell parameters include at least the individual cell's battery voltage, terminal temperature, and internal resistance. The environmental parameters include at least the ambient temperature and ambient humidity.

8. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 6, characterized in that, According to the aforementioned adaptation prediction scheme, multi-source prediction data is collected, and a battery remaining capacity prediction plugin based on deep learning is invoked. Based on the multi-source prediction data, the remaining battery capacity of the target battery pack is predicted, and the battery remaining capacity assessment result is output, including: The multi-source prediction data for the target battery pack is collected and obtained according to the set of adaptation prediction indicators. The adaptive battery capacity prediction plugin is called according to the adaptive prediction index set, wherein the adaptive battery capacity prediction plugin includes N battery capacity prediction models, where N is greater than or equal to 10. According to the number of models selected for adaptation, the same number of prediction models are randomly selected from the N battery capacity prediction models of the adapted battery capacity prediction plugin. The remaining battery capacity is predicted according to the adapted multi-source prediction data, and the mean of multiple prediction results is fitted to obtain the remaining battery capacity assessment result.

9. The deep learning-driven dynamic evaluation method for remaining battery capacity according to claim 8, characterized in that, The method for constructing the adaptive battery capacity prediction plugin includes: Using the set of adaptation prediction indicators as constraints, retrieve historical operation records of similar battery packs, collect multi-source sample datasets, and use the historical battery remaining capacity corresponding to different multi-source sample data at the same time as the sample battery remaining capacity to obtain the sample battery remaining capacity set. The multi-source dataset and the remaining battery capacity dataset are used as training data. The training data is divided into N equal parts and randomly selected with replacement to obtain N sample training sets. The deep learning models are trained to convergence using the N sample training sets to obtain N battery capacity prediction models, which are then combined to construct an adaptive battery capacity prediction plugin.

10. A deep learning-driven dynamic evaluation system for remaining battery capacity, characterized in that, The system is used to execute the deep learning-driven dynamic evaluation method for battery remaining capacity according to any one of claims 1-9, and the system comprises: The SOC and error reading module is used to read the current SOC region of the target battery pack and match it to obtain the current prediction error confidence interval of the remaining battery capacity prediction. The multi-source data stability analysis module is used to monitor and acquire the multi-source operating data sequence of the target battery pack in the historical time zone, perform operating status stability analysis based on the multi-source operating data sequence, and output the current operating stability coefficient. The prediction depth compensation module is used to compensate the preset standard prediction depth based on the current prediction error confidence interval and the current operating stability coefficient, and generate an adapted prediction depth. The prediction scheme execution module is used to formulate an adaptation prediction scheme based on a preset prediction index space, the current operating stability coefficient, and the adaptation prediction depth; collect and acquire adaptation multi-source prediction data according to the adaptation prediction scheme; and call a battery remaining capacity prediction plugin built based on deep learning to predict the battery remaining capacity of the target battery pack according to the adaptation multi-source prediction data, and output the battery remaining capacity assessment result.

Images