Energy storage battery life prediction method and related equipment

Abstract

A life prediction method for energy storage batteries and related equipment, relating to the field of electrical digital data processing technology. In this method, a current operating data group and a current environmental data group corresponding to the current operating data group are obtained, and a health index is calculated based on the operating data in the current operating data group to obtain a time series; the time series is input into a preset time series analysis model to obtain a preliminary prediction curve; the difference between the environmental change table and the preset ideal environmental change table is input into a preset influencing factor model to obtain a compensation curve; the preliminary prediction curve is compensated by the compensation curve to obtain a prediction curve; when it is determined that the target time has been received, the battery life is determined in the prediction curve according to the target time. It can significantly reduce the interference of environmental factors on the prediction results and avoid potential losses caused by prediction errors.

Description

Technical Field

[0001] The present application relates to the technical field of electrical digital data processing, and in particular to a life prediction method for energy storage batteries and related equipment. Background Art

[0002] In the field of energy storage, energy storage batteries need to operate in a variety of application environments, such as grid regulation, electric vehicles, renewable energy systems, etc. These application environments are affected by a variety of external conditions, including temperature fluctuations, load changes, grid status, etc. These factors affect the discharge and charging rates of the battery, and thus affect the prediction of the remaining service life of the energy storage battery.

[0003] At present, energy storage battery life prediction technology mainly relies on battery operation data collected in actual applications. Through statistical analysis of these data, indicators of battery health status can be obtained, and then the remaining life of the battery can be predicted based on these health indicators. This prediction method relies heavily on the battery's past usage data to predict its future life.

[0004] However, energy storage battery life prediction technology faces some challenges. First, the actual battery operation data is usually affected by external environmental factors, such as temperature, humidity, load fluctuations, etc. These factors may affect the discharge and charging speed of the battery. If the data affected by the environment is directly used for prediction, the prediction results may not be accurate enough. Summary of the invention

[0005] The present application provides a life prediction method and related equipment for energy storage batteries, which are used to reduce the interference of environmental factors on the prediction results and avoid potential losses caused by prediction errors.

[0006] In a first aspect, the present application provides a method for predicting the life of an energy storage battery, comprising:

[0007] Obtaining a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a number of operation data at an identified time, and the current environment data group is composed of a number of environment data at an identified time;

[0008] Calculate the health index according to the operation data in the current operation data group to obtain a time series, where the time series is the change entity of the health index over time;

[0009] Input the time series into a preset time series analysis model to obtain a preliminary prediction curve, the preliminary prediction curve including a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators in the future as the marked time progresses;

[0010] The difference between the environmental change table and the preset ideal environmental change table is input into the preset influencing factor model to obtain a compensation curve, which is a compensation curve segment in which the compensation value changes with the marked time from the historical time to the future time;

[0011] The prediction curve is obtained by compensating the preliminary prediction curve using the compensation curve;

[0012] In the case where it is determined that the target time is received, the battery life is determined in the prediction curve according to the target time.

[0013] In the above embodiment, a preliminary prediction curve and a compensation curve for environmental impact are generated by acquiring and analyzing the current operating data group and the current environmental data group. The compensation curve is used to compensate for the preliminary prediction curve. The resulting prediction curve will be a battery life prediction curve that is not affected by the environment. This method can significantly reduce the interference of environmental factors on the prediction results and avoid potential losses caused by prediction errors.

[0014] In combination with some embodiments of the first aspect, in some embodiments, before the step of inputting the difference between the environment change table and the preset ideal environment change table into the preset influencing factor model to obtain the compensation curve, the method further includes:

[0015] Scoring each historical operation data in the historical operation data group according to the scoring rule;

[0016] Classifying the historical operation data into a number of similar condition groups according to the scores of the historical operation data;

[0017] Determine the difference between the change amount of the health indicator corresponding to the first historical operation data and the change amount of the health indicator corresponding to the second historical operation data as the change amount difference, the change amount is the difference between the health indicator corresponding to the historical operation data at a previous time and the health indicator corresponding to the historical operation data at a later identified time, the first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group;

[0018] determining an environmental data difference between environmental data corresponding to the first historical operation data and environmental data corresponding to the second historical operation data as an environmental data difference;

[0019] Determine the difference of environmental data and the difference of transformation amount as the impact data;

[0020] Count all impact data to obtain the impact data set;

[0021] The influence data set is used as a training sample to obtain the preset influence factor model.

[0022] In the above embodiment, by scoring the historical operation data according to the scoring rules and classifying it into different similar condition groups according to the preset score range, it is ensured that the environmental deviation of all historical operation data in the similar condition group is kept within a certain balance range, thereby improving the stability and comparability of the data. On this basis, the difference in health indicator change and environmental data of different historical operation data in the same condition group are further compared, so that the impact of environmental changes on battery health indicators can be accurately understood, thereby improving the prediction accuracy of environmental impacts. All impact data are summarized to obtain an impact data set that comprehensively reflects the difference in health indicator change and environmental change, thereby helping the model to be trained.

[0023] In combination with some embodiments of the first aspect, in some embodiments, the step of using the influencing data set as a training sample to obtain a preset influencing factor model specifically includes:

[0024] Input the impact data set into the preset influence rule to solve and obtain the coefficient of each change amount. The preset influence rule is: regard each change amount of the difference of the environmental data in the impact data as a preset parameter; assign a coefficient to each preset parameter; multiply these changes by their corresponding coefficients and sum them, and the result obtained is the difference of the change amount in the current impact data;

[0025] The preset influencing factor model is obtained based on the coefficient of each change amount as a training sample.

[0026] In the above embodiment, a preset influence rule is constructed by treating each change in the difference of the environmental data as a preset parameter and assigning a coefficient to each parameter. Then, by solving this rule, the coefficient of each change can be obtained, and these coefficients actually reflect the influence of each change on the result. Therefore, this method realizes the quantification of the influence of environmental data on the result, so that the influence of environmental factors can be clearly reflected in the prediction model in a numerical way.

[0027] In conjunction with some embodiments of the first aspect, in some embodiments,

[0028] The scoring rule is: subtract each environmental data in the historical operation data from the corresponding threshold to obtain a sub-score;

[0029] The sub-scores are multiplied by their corresponding weights and the average is taken to obtain the corresponding score.

[0030] In the above embodiment, the difference between each environmental data and its threshold is calculated, and then these sub-scores are combined with corresponding weights to calculate a weighted average score. This score can be used to measure the deviation between historical operating data and the set environmental standard, and reflect the importance of different environmental data through weights, thereby providing a quantitative basis for data analysis.

[0031] In conjunction with some embodiments of the first aspect, in some embodiments, before the step of inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve, the method further includes:

[0032] Construct the autocorrelation function diagram and partial autocorrelation function diagram of the current running data group;

[0033] Determine the order of the autoregressive term and the order of the moving average term based on the autocorrelation function graph and the partial autocorrelation function graph;

[0034] The order of the autoregressive term and the order of the moving average term are used as parameters and the historical operation data group is used as training samples to train the time series analysis model.

[0035] In the above embodiment, by constructing the autocorrelation function graph and the partial autocorrelation function graph of the current operation data group, the order of the autoregressive term and the order of the moving average term are further determined, and then these orders are used as parameters to train the time series analysis model using the historical operation data group as a training sample. By using the historical operation data group as a training sample, the existing historical data can be fully utilized to improve the prediction ability of the model.

[0036] In conjunction with some embodiments of the first aspect, in some embodiments, the step of inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve specifically includes:

[0037] When it is determined that the variance of any segment of data in the time series is greater than the variance of another segment of data, the time series is subjected to difference processing to obtain a difference time series;

[0038] Input the difference time series into the preset time series analysis model to obtain the initial prediction curve after difference;

[0039] The initial prediction curve after differentiation is inversely differentiated to obtain the initial prediction curve.

[0040] In the above embodiment, the method of first performing a difference process on the time series, then inputting the difference time series into a preset time series analysis model, and finally performing an inverse difference process on the obtained prediction curve to obtain the final prediction curve is used. The non-stationary time series is converted into a stationary series, so that the subsequent time series analysis model is more accurate and effective. In addition, the prediction result can be converted back to the original data scale through the inverse difference process.

[0041] In conjunction with some embodiments of the first aspect, in some embodiments, the step of calculating the health index according to the operating data in the current operating data group to obtain the time series specifically includes:

[0042] Calculating the current capacity of the battery according to the operating data in the current operating data group;

[0043] Divide the current capacity by the initial capacity to get the health indicator.

[0044] In the above embodiment, the health status of the battery is measured in a simple and intuitive way, that is, the ratio of the current capacity of the battery to the initial capacity. The higher the value, the better the health status of the battery; otherwise, it means that the health status of the battery is poor.

[0045] In a second aspect, an embodiment of the present application provides a life prediction system for an energy storage battery, comprising:

[0046] A data acquisition module, used to acquire a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a number of operation data at an identified time, and the current environment data group is composed of a number of environment data at an identified time;

[0047] The time series module is used to calculate the health index according to the operation data in the current operation data group to obtain the time series, and the time series is the change entity of the health index over time;

[0048] A prediction module, used to input the time series into a preset time series analysis model to obtain a preliminary prediction curve, the preliminary prediction curve including a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators in the future as the marked time progresses;

[0049] A compensation curve module is used to input the difference between the environmental change table and the preset ideal environmental change table into a preset influencing factor model to obtain a compensation curve, which is a compensation curve segment in which the compensation value changes with the marked time from historical time to future time;

[0050] A compensation module, used for compensating the preliminary prediction curve using the compensation curve to obtain the prediction curve;

[0051] The life prediction module is used to determine the battery life in the prediction curve according to the target time when it is determined that the target time is received.

[0052] In conjunction with some embodiments of the second aspect, in some embodiments, the system further includes:

[0053] A scoring module, used for scoring each historical operation data in the historical operation data group according to a scoring rule;

[0054] A classification module, used for classifying the historical operation data into a plurality of similar condition groups according to the scores of the historical operation data;

[0055] A first difference module is used to determine the difference between the change amount of the health indicator corresponding to the first historical operation data and the change amount of the health indicator corresponding to the second historical operation data as the change amount difference, the change amount is the difference between the health indicator corresponding to the historical operation data at a previous time and the health indicator corresponding to the historical operation data at a later identification time, the first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group;

[0056] A second difference module, configured to determine an environmental data difference between environmental data corresponding to the first historical operation data and environmental data corresponding to the second historical operation data as an environmental data difference;

[0057] An impact data module, used for determining the difference of environmental data and the difference of transformation amount as impact data;

[0058] A statistical module is used to collect statistics of all impact data to obtain an impact data set;

[0059] The first training module is used to obtain a preset influencing factor model by using the influencing data set as a training sample.

[0060] In conjunction with some embodiments of the second aspect, in some embodiments, the first training module specifically includes:

[0061] The coefficient submodule is used to input the impact data set into the preset influence rule to solve and obtain the coefficient of each change amount. The preset influence rule is: each change amount of the difference of the environmental data in the impact data is regarded as a preset parameter; a coefficient is assigned to each preset parameter; these changes are multiplied by their corresponding coefficients and summed, and the result obtained is the difference of the transformation amount in the current impact data;

[0062] The first training submodule is used to obtain a preset influencing factor model based on the coefficient of each variation as a training sample.

[0063] In conjunction with some embodiments of the second aspect, in some embodiments, the scoring rule is: subtracting each environmental data in the historical operation data from the corresponding threshold value to obtain a sub-score;

[0064] The sub-scores are multiplied by their corresponding weights and the average is taken to obtain the corresponding score.

[0065] In conjunction with some embodiments of the second aspect, in some embodiments, the system further includes:

[0066] A construction module, used for constructing an autocorrelation function diagram and a partial autocorrelation function diagram of a current running data group;

[0067] An order module is used to determine the order of the autoregressive term and the order of the moving average term based on the autocorrelation function graph and the partial autocorrelation function graph;

[0068] The second training module is used to train the time series analysis model by taking the order of the autoregressive term and the order of the moving average term as parameters and the historical operation data group as training samples.

[0069] In conjunction with some embodiments of the second aspect, in some embodiments, the prediction module specifically includes:

[0070] A difference submodule is used to perform difference processing on the time series to obtain a differential time series when it is determined that the variance of any segment of data in the time series is greater than a preset variance threshold value with respect to another segment of data;

[0071] The prediction submodule is used to input the differential time series into the preset time series analysis model to obtain a preliminary prediction curve after differentiation;

[0072] The reverse difference numerator module is used to reverse the difference of the preliminary prediction curve to obtain the preliminary prediction curve.

[0073] In conjunction with some embodiments of the second aspect, in some embodiments, the time series module specifically includes:

[0074] A capacity calculation submodule, used to calculate the current capacity of the battery according to the operation data in the current operation data group;

[0075] The health indicator submodule is used to obtain the health indicator by dividing the current capacity by the initial capacity.

[0076] In a third aspect, an embodiment of the present application provides a life prediction system for an energy storage battery, the system comprising: one or more processors and a memory;

[0077] The memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions. The one or more processors call the computer instructions to enable the life prediction system of the energy storage battery to execute the method described in the first aspect and the first possible implementation method in the first aspect.

[0078] In a fourth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a server, enables the server to execute the method described in the first aspect and the first possible implementation manner in the first aspect.

[0079] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions. When the instructions are executed on a life prediction system for an energy storage battery, the life prediction system for the energy storage battery executes the method described in the first aspect and the first possible implementation method in the first aspect.

[0080] It can be understood that the energy storage battery life prediction system provided in the second aspect, the energy storage battery life prediction system provided in the third aspect, the computer program product provided in the fourth aspect, and the computer storage medium provided in the fifth aspect are all used to execute the energy storage battery life prediction method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0081] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0082] 1. The life prediction method of the energy storage battery provided in the present application generates a preliminary prediction curve and a compensation curve for environmental impact by acquiring and analyzing the current operation data group and the current environmental data group, and the compensation curve is used to compensate the preliminary prediction curve. The obtained prediction curve will be a battery life prediction curve that is not affected by the environment. This method can significantly reduce the interference of environmental factors on the prediction results and avoid potential losses caused by prediction errors.

[0083] 2. The life prediction method of energy storage batteries provided in this application scores historical operating data according to scoring rules, and classifies them into different similar condition groups according to preset scoring intervals, thereby ensuring that the environmental deviations of all historical operating data in similar condition groups remain within a certain balance range, thereby improving the stability and comparability of the data. On this basis, the difference in health indicator change and environmental data of different historical operating data in the same condition group are further compared, which can accurately understand the impact of environmental changes on battery health indicators, thereby improving the prediction accuracy of environmental impacts. All impact data are summarized to obtain an impact data set that comprehensively reflects the difference in health indicator change and environmental change, which helps the model to be trained.

[0084] 3. The life prediction method of energy storage batteries provided in this application constructs a preset influence rule by treating each change in the difference of environmental data as a preset parameter and assigning a coefficient to each parameter. Then, by solving this rule, the coefficient of each change can be obtained, and these coefficients actually reflect the influence of each change on the result. Therefore, this method realizes the quantification of the impact of environmental data on the results, so that the impact of environmental factors can be clearly reflected in the prediction model in a numerical way. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 A schematic flow chart of the life prediction method of the energy storage battery provided in this application.

[0086] Figure 2 A second flow chart of the life prediction method of the energy storage battery provided in this application.

[0087] Figure 3 A schematic diagram of a modular virtual device for the life prediction system of energy storage batteries provided in this application.

[0088] Figure 4 A schematic diagram of a physical device of the energy storage battery life prediction system provided in this application. DETAILED DESCRIPTION

[0089] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.

[0090] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0091] The following is a description of the life prediction method of the energy storage battery in this embodiment:

[0092] like Figure 1 As shown, Figure 1 A schematic flow chart of the life prediction method of the energy storage battery provided in this application.

[0093] S101. Obtain a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a plurality of operation data with an identified time, and the current environment data group is composed of a plurality of environment data with an identified time.

[0094] The current operation data group contains various battery data collected by the energy storage battery during operation, which are marked as operation data. For example, each operation data may include key parameters such as battery voltage, current, and remaining power. These data are obtained at a specific collection time point, so each operation data corresponds to a specific timestamp.

[0095] Similarly, the current environmental data group contains various environmental data collected by the energy storage battery during operation. Each environmental data may include factors such as temperature, humidity, load fluctuation, etc. Each environmental data corresponds to a specific timestamp, and these timestamps correspond one-to-one with the timestamps in the current operation data group. In other words, each time the operation data is collected, the corresponding environmental data will also be collected.

[0096] It should be noted that when matching the operation data and the environment data according to the timestamps, it is not necessary to require the timestamps to be strictly equal. In fact, in some cases, it is impossible to ensure that the collection time of the operation data and the environment data is completely consistent. Therefore, a time difference threshold can be set. As long as the difference between the timestamps of the two sets of data is within this threshold, they are considered to be corresponding.

[0097] S102. Calculate the health index according to the operation data in the current operation data group to obtain a time series, where the time series is an entity of changes in the health index over time.

[0098] In a specific embodiment, the current capacity of the battery is calculated based on the operating data in the current operating data group; the health index is obtained by dividing the current capacity by the initial capacity.

[0099] In a specific embodiment, the current of the battery is measured over a period of time to obtain the power consumption, which is then multiplied by the voltage to obtain the power, which is then divided by the percentage of the power used to obtain the current capacity. In some other embodiments, a discharge curve is made using the voltage and current, and the current capacity is determined using the discharge curve. In some other embodiments, the current capacity of the battery is directly measured by directly controlling the battery to discharge at a certain current or voltage, and then measuring how long the battery can continue to discharge, which is not limited here.

[0100] It can be seen that the health of the battery can be measured in a simple and intuitive way, that is, the ratio of the current capacity of the battery to the initial capacity. The higher the value, the better the health of the battery; otherwise, it means that the health of the battery is poor.

[0101] S103, inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve, wherein the preliminary prediction curve includes a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators with marked time in the future.

[0102] In some embodiments, the time series needs to be preprocessed first, because many of the above embodiments are predictions, so some data may be inaccurate. For the time series analysis model, inaccurate data may cause large errors in the model prediction results. In some embodiments, if the difference between two adjacent data points exceeds 2 times (or 3 times) the mean, these two data points are removed and the median is used for filling, which is not limited here.

[0103] The training process of the time series analysis model is as follows: construct the autocorrelation function graph and partial autocorrelation function graph of the current operating data group; determine the order of the autoregressive term and the order of the moving average term based on the autocorrelation function graph and the partial autocorrelation function graph; use the order of the autoregressive term and the order of the moving average term as parameters and the historical operating data group as training samples to train the time series analysis model.

[0104] It should be noted that the construction of autocorrelation function graphs and partial autocorrelation function graphs has been quite mature in related technologies. For example, the autocorrelation function graphs and partial autocorrelation function graphs can be directly generated through the statsmodels library, and no limitation is made here.

[0105] Observe the partial autocorrelation function graph and find out at which point the lag is first truncated (i.e., falls within the confidence interval or close to 0). This lag value is the value of the order of the autoregressive term. Observe the autocorrelation function graph and find out at which point the lag is first truncated. This lag value is the value of the order of the moving average term. The relevant technology for determining the order of the autoregressive term and the order of the moving average term is quite mature and is not limited here.

[0106] It can be seen that by constructing the autocorrelation function graph and partial autocorrelation function graph of the current operation data group, the order of the autoregressive term and the order of the moving average term are further determined, and then these orders are used as parameters to train the time series analysis model using the historical operation data group as a training sample. By using the historical operation data group as a training sample, the existing historical data can be fully utilized to improve the prediction ability of the model.

[0107] It should be noted that a time series is a set of discrete time series data points. The preliminary prediction curve contains all discrete points, that is, the data point density of the prediction curve should be greater than the original time series, which may mean inserting additional prediction data points between the original data points; and the time span of the prediction curve should be greater than the original time series, that is, it has predictions for future time points.

[0108] In actual use, the battery prediction curve is actually not stationary, which means that the statistical characteristics of its data (such as mean and variance) change over time. This non-stationarity is a problem, and the above time series analysis models are all based on the assumption that the data is stationary. Therefore, if the model is used to directly analyze the battery prediction curve, a large prediction error may be obtained.

[0109] Therefore, a specific solution is: when it is determined that the variance of any segment of data in the time series is greater than the variance of another segment of data than the preset variance threshold, the time series is differentiated to obtain a differential time series; the differential time series is input into the preset time series analysis model to obtain a preliminary prediction curve after differentiation; the preliminary prediction curve after differentiation is inversely differentiated to obtain a preliminary prediction curve.

[0110] The battery prediction curve is differentiated to make it a stationary series. The processed data is then input into the time series analysis model, which may result in a more accurate prediction result. The initial prediction curve after differentiation is then reverse-differentiated to obtain the initial prediction curve.

[0111] It can be seen that the method of first performing differential processing on the time series, then inputting the differential time series into the preset time series analysis model, and finally performing inverse differential processing on the obtained prediction curve to obtain the final prediction curve makes the non-stationary time series converted into a stationary series, thereby making the subsequent time series analysis model more accurate and effective. In addition, through inverse differential processing, the prediction results can be converted back to the original data scale.

[0112] S104, inputting the difference between the environment change table and the preset ideal environment change table into the preset influencing factor model to obtain a compensation curve, where the compensation curve is a compensation curve segment in which the compensation value changes with the marked time from the historical time to the future time.

[0113] S105. Compensate the preliminary prediction curve using the compensation curve to obtain a prediction curve.

[0114] In a specific embodiment, the initial prediction curve and the compensation curve are simply subtracted to obtain the result.

[0115] S106: When it is determined that the target time is received, determine the battery life in the prediction curve according to the target time.

[0116] It can be seen that by acquiring and analyzing the current operating data group and the current environmental data group to generate a preliminary prediction curve and a compensation curve for environmental impact, and using the compensation curve to compensate for the preliminary prediction curve, the resulting prediction curve will be a battery life prediction curve that is not affected by the environment. This method can significantly reduce the interference of environmental factors on the prediction results and avoid potential losses caused by prediction errors.

[0117] In the above embodiments, the interference of environmental factors on the prediction results can be significantly reduced, and potential losses caused by prediction errors can be avoided. In practical applications, how to accurately understand the impact of environmental changes on battery health indicators. Figure 2 The second flow chart of the life prediction method of the energy storage battery shown in FIG. 1 specifically describes the life prediction method of the energy storage battery in the embodiment of the present application:

[0118] like Figure 2 As shown, Figure 2 A second flow chart of the life prediction method of the energy storage battery provided in this application.

[0119] S201. Score each historical operation data in the historical operation data group according to a scoring rule.

[0120] The scoring rule is: subtract each environmental data in the historical operation data from the corresponding threshold to obtain a sub-score;

[0121] The sub-scores are multiplied by their corresponding weights and the average is taken to obtain the corresponding score.

[0122] In actual use, before step S201, the thresholds of each environmental parameter are determined. These thresholds are predefined standards used to evaluate whether each environmental parameter is within the normal operating range. And a weight is assigned to each environmental parameter. The purpose of the weight is to balance the impact of each parameter and ensure that no single environmental factor has a disproportionate impact on the final score due to a large fluctuation. For example, temperature, humidity, and load fluctuations may have different ranges of values ​​and degrees of impact on system health. The weight can adjust their contribution to the score so that they maintain corresponding proportionality and importance during the evaluation process.

[0123] It can be seen that by calculating the difference between each environmental data and its threshold, and then combining these sub-scores with the corresponding weights to calculate the weighted average score. This score can be used to measure the deviation between historical operating data and the set environmental standards, and reflect the importance of different environmental data through weights, thus providing a quantitative basis for data analysis.

[0124] S202: Classify the historical operation data into a plurality of similar condition groups according to the scores of the historical operation data.

[0125] Continuing with the above example, if the score is between 0 and 100, it can be divided into 10 intervals, namely [0, 10], [10, 20], ..., [90, 100]. Each historical running data is classified into the corresponding interval according to its propensity score.

[0126] S203. Determine the difference in the transformation amounts of the corresponding health indicators between the first historical operation data and the second historical operation data as the transformation amount difference. The transformation amount is the difference between the health indicator corresponding to the historical operation data at the previous time and the health indicator corresponding to the historical operation data at the next identification time. The first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group.

[0127] For each piece of historical operation data, first, the change amount of the health indicator needs to be calculated. This change amount is calculated as the difference between the health indicator corresponding to the historical operation data at the previous time point and the health indicator corresponding to the historical operation data at the next time point. That is to say, if there are two pieces of historical operation data at time t1 and t2 (where t1 < t2), then the change amount of the health indicator at time t2 is the health indicator value at time t2 minus the health indicator value at time t1. Select two pieces of historical operation data within the same similar condition group, calculate their respective change amounts of the health indicators, and then calculate the difference between these two change amounts. This difference is called the transformation amount difference.

[0128] S204. Determine the environmental data difference between the environmental data corresponding to the first historical operation data and the environmental data corresponding to the second historical operation data as the environmental data difference.

[0129] It should be noted that the environmental data difference needs to correspond to the transformation amount difference. Therefore, the two previously selected data (the first historical operation data and the second historical operation data) are used, on the one hand, to calculate the environmental data difference, and on the other hand, to calculate the transformation amount difference. In this way, we can ensure that the environmental data difference corresponds to the transformation amount difference.

[0130] S205. Determine the environmental data difference and the transformation amount difference as the influencing data.

[0131] It should be noted that the above process is just an example with one set of data. In fact, other data also need to be processed in the same way.

[0132] S206. Statistically analyze all the influencing data to obtain an influencing data set.

[0133] Continuing with the above example, select a similar condition group, and then within this group, process each set of historical operation data according to the above method to obtain their respective influencing data. Select other similar condition groups and process the historical operation data within each group in the same way to obtain their respective influencing data.

[0134] S207. Input the influencing data set into the preset influence rule to solve and obtain the coefficient of each change amount. The preset influence rule is: regard each change amount of the environmental data difference in the influencing data as a preset parameter; assign a coefficient to each preset parameter; multiply these changes by their corresponding coefficients and sum them, and the result obtained is the transformation amount difference in the current influencing data.

[0135] Continuing with the above example, in the example described in steps S203 to S205, the transformation amount difference D1 (obtained in step 203) and the environmental data difference T (obtained in step 204) are obtained. According to the previous embodiment, the environmental data may include factors such as temperature, humidity, and load fluctuation. Then the environmental data difference T includes: temperature difference T1 (temperature environmental data difference between the temperature environmental data corresponding to the first historical operating data and the temperature environmental data corresponding to the second historical operating data), T2 (humidity environmental data difference between the humidity environmental data corresponding to the first historical operating data and the humidity environmental data corresponding to the second historical operating data), and T3 (load fluctuation environmental data difference between the load fluctuation environmental data corresponding to the first historical operating data and the load fluctuation environmental data corresponding to the second historical operating data).

[0136] Then according to the preset influence rule, the mathematical formula can be obtained as follows:

[0137] D1=C1T1+C2T2+C3T3

[0138] In the formula, C1, C2, and C3 are coefficients for temperature, humidity, and load fluctuation distribution.

[0139] As can be seen from step S206, other influencing data are also processed according to this example, and finally a formula group is obtained:

[0140] D1=C1T1+C2T2+C3T3

[0141] D2=C1T4+C2T5+C3T6

[0142] ……

[0143] D n =C1T n+1 +C2T n+2 +C3T n+3

[0144] Solving this parameter yields the coefficient for each variable.

[0145] S208. Obtain a preset influencing factor model based on the coefficient of each variation as a training sample.

[0146] It can be seen that by scoring the historical operating data according to the scoring rules and classifying them into different similar condition groups according to the preset scoring range, it is ensured that the environmental deviation of all historical operating data in the similar condition group is kept within a certain balance range, thereby improving the stability and comparability of the data. On this basis, the difference in health indicator change and environmental data of different historical operating data in the same condition group are further compared, which can accurately understand the impact of environmental changes on battery health indicators, thereby improving the prediction accuracy of environmental impacts. All impact data are summarized to obtain an impact data set that comprehensively reflects the difference in health indicator change and environmental change, which helps the model to be trained.

[0147] It can be seen that by treating each change in the difference of environmental data as a preset parameter and assigning a coefficient to each parameter, a preset influence rule is constructed. Then, by solving this rule, the coefficient of each change can be obtained, which actually reflects the influence of each change on the result. Therefore, this method realizes the quantification of the impact of environmental data on the result, so that the impact of environmental factors can be clearly reflected in the prediction model in a numerical way.

[0148] The following are device embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0149] refer to Figure 3 The embodiment of the present application provides a life prediction system for an energy storage battery. The life prediction system for an energy storage battery includes:

[0150] The data acquisition module 301 is used to acquire a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a plurality of operation data at an identified time, and the current environment data group is composed of a plurality of environment data at an identified time;

[0151] A time series module 302 is used to calculate the health index according to the operation data in the current operation data group to obtain a time series, where the time series is an entity of changes in the health index over time;

[0152] Prediction module 303, used to input the time series into a preset time series analysis model to obtain a preliminary prediction curve, the preliminary prediction curve including a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators in the future with the marked time;

[0153] The compensation curve module 304 is used to input the difference between the environment change table and the preset ideal environment change table into the preset influencing factor model to obtain a compensation curve, which is a compensation curve segment in which the compensation value changes with the marked time from the historical time to the future time;

[0154] A compensation module 305, used to compensate the preliminary prediction curve using the compensation curve to obtain a prediction curve;

[0155] The life prediction module 306 is used to determine the battery life in the prediction curve according to the target time when it is determined that the target time is received.

[0156] In some embodiments, the system further comprises:

[0157] A scoring module, used for scoring each historical operation data in the historical operation data group according to a scoring rule;

[0158] A classification module, used for classifying the historical operation data into a plurality of similar condition groups according to the scores of the historical operation data;

[0159] A first difference module is used to determine the difference between the change amount of the health indicator corresponding to the first historical operation data and the change amount of the health indicator corresponding to the second historical operation data as the change amount difference, the change amount is the difference between the health indicator corresponding to the historical operation data at a previous time and the health indicator corresponding to the historical operation data at a later identification time, the first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group;

[0160] A second difference module, configured to determine an environmental data difference between environmental data corresponding to the first historical operation data and environmental data corresponding to the second historical operation data as an environmental data difference;

[0161] An impact data module, used for determining the difference of environmental data and the difference of transformation amount as impact data;

[0162] A statistical module is used to collect statistics of all impact data to obtain an impact data set;

[0163] The first training module is used to obtain a preset influencing factor model by using the influencing data set as a training sample.

[0164] In some embodiments, the first training module specifically includes:

[0165] The coefficient submodule is used to input the impact data set into the preset influence rule to solve and obtain the coefficient of each change amount. The preset influence rule is: each change amount of the difference of the environmental data in the impact data is regarded as a preset parameter; a coefficient is assigned to each preset parameter; these changes are multiplied by their corresponding coefficients and summed, and the result obtained is the difference of the transformation amount in the current impact data;

[0166] The first training submodule is used to obtain a preset influencing factor model based on the coefficient of each variation as a training sample.

[0167] In some embodiments, the scoring rule is: subtracting each environmental data in the historical operation data from the corresponding threshold value to obtain a sub-score;

[0168] The sub-scores are multiplied by their corresponding weights and the average is taken to obtain the corresponding score.

[0169] In some embodiments, the system further comprises:

[0170] A construction module, used for constructing an autocorrelation function diagram and a partial autocorrelation function diagram of a current running data group;

[0171] An order module is used to determine the order of the autoregressive term and the order of the moving average term based on the autocorrelation function graph and the partial autocorrelation function graph;

[0172] The second training module is used to train the time series analysis model by taking the order of the autoregressive term and the order of the moving average term as parameters and the historical operation data group as training samples.

[0173] In some embodiments, the prediction module specifically includes:

[0174] A difference submodule is used to perform difference processing on the time series to obtain a differential time series when it is determined that the variance of any segment of data in the time series is greater than a preset variance threshold value with respect to another segment of data;

[0175] The prediction submodule is used to input the differential time series into the preset time series analysis model to obtain a preliminary prediction curve after differentiation;

[0176] The reverse difference numerator module is used to reverse the difference of the preliminary prediction curve to obtain the preliminary prediction curve.

[0177] In some embodiments, the time series module specifically includes:

[0178] A capacity calculation submodule, used to calculate the current capacity of the battery according to the operation data in the current operation data group;

[0179] The health indicator submodule is used to obtain the health indicator by dividing the current capacity by the initial capacity.

[0180] The present application also discloses a life prediction system for energy storage batteries. Figure 4 , which is a schematic diagram of a physical device of the life prediction system of the energy storage battery provided in the present application. The computer 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0181] The communication bus 402 is used to realize the connection and communication between these components.

[0182] The user interface 403 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0183] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0184] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0185] Among them, the memory 405 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 405 may optionally be at least one storage device located away from the aforementioned processor 401. Refer to Figure 4, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a life prediction method of an energy storage battery.

[0186] exist Figure 4 In the computer 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call the application program of the life prediction method of the energy storage battery stored in the memory 405. When executed by one or more processors 401, the computer 400 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0187] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0188] In the several implementation modes provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into a second system, or some features can be ignored or not executed. Second, the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0189] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0190] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0191] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0192] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0193] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for predicting the life of an energy storage battery, characterized in that: include: Acquire a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a plurality of operation data at an identified time, and the current environment data group is composed of a plurality of environment data at an identified time; Calculate the health index according to the operation data in the current operation data group to obtain a time series, wherein the time series is an entity of changes in the health index over time; Inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve, wherein the preliminary prediction curve includes a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators in the future as the marked time progresses; Scoring each historical operation data in the historical operation data group according to the scoring rule; Classifying the historical operation data into a plurality of similar condition groups according to the scores of the historical operation data; Determine the difference between the change amount of the health indicator corresponding to the first historical operation data and the change amount of the health indicator corresponding to the second historical operation data as the change amount difference, wherein the change amount is the difference between the health indicator corresponding to the historical operation data at a previous time and the health indicator corresponding to the historical operation data at a later identified time, wherein the first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group; determining an environmental data difference between environmental data corresponding to the first historical operation data and environmental data corresponding to the second historical operation data as an environmental data difference; Determine the environmental data difference and the transformation amount difference as impact data; Count all impact data to obtain the impact data set; Using the influencing data set as a training sample to obtain a preset influencing factor model; The difference between the environmental change table and the preset ideal environmental change table is input into the preset influencing factor model to obtain a compensation curve, wherein the compensation curve is a compensation curve segment in which the compensation value changes with the marked time from the historical time to the future time; The prediction curve is obtained by compensating the preliminary prediction curve using the compensation curve; In the case where it is determined that the target time is received, the battery life is determined in the prediction curve according to the target time.

2. The life prediction method of energy storage battery according to claim 1, characterized in that: The step of using the influencing data set as a training sample to obtain a preset influencing factor model specifically includes: Input the impact data set into the preset influence rule to solve and obtain the coefficient of each change amount, wherein the preset influence rule is: regard each change amount of the difference of the environmental data in the impact data as a preset parameter; assign a coefficient to each preset parameter; multiply these changes by their corresponding coefficients and sum them, and the result obtained is the difference of the transformation amount in the current impact data; The preset influencing factor model is obtained based on the coefficient of each change amount as a training sample.

3. The life prediction method of energy storage battery according to claim 1 is characterized in that: The scoring rule is: subtract each environmental data in the historical operation data from the corresponding threshold to obtain a sub-score; The sub-scores are multiplied by their corresponding weights and the average is taken to obtain the corresponding score.

4. The life prediction method of energy storage battery according to claim 1, characterized in that: Before the step of inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve, the method further includes: Construct the autocorrelation function diagram and partial autocorrelation function diagram of the current running data group; Determining the order of the autoregressive term and the order of the moving average term according to the autocorrelation function graph and the partial autocorrelation function graph; The order of the autoregressive term and the order of the moving average term are used as parameters, and the historical operation data group is used as a training sample to train a time series analysis model.

5. The life prediction method of energy storage battery according to claim 1, characterized in that: The step of inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve specifically includes: When it is determined that the variance of any one segment of data in the time series is greater than a preset variance threshold value, performing difference processing on the time series to obtain a differential time series; Inputting the differential time series into a preset time series analysis model to obtain a preliminary prediction curve after differentiation; The initial prediction curve after differentiation is inversely differentiated to obtain the initial prediction curve.

6. The life prediction method of energy storage battery according to claim 1, characterized in that: The step of calculating the health index according to the operation data in the current operation data group to obtain a time series specifically includes: Calculating a current capacity of the battery according to the operation data in the current operation data group; The health indicator is obtained by dividing the current capacity by the initial capacity.

7. A life prediction system for energy storage batteries, characterized in that: include: A data acquisition module, used to acquire a current operation data group and a current environment data group corresponding to the current operation data group, wherein the current operation data group is composed of a plurality of operation data at an identified time, and the current environment data group is composed of a plurality of environment data at an identified time; A time series module, used to calculate the health index according to the operation data in the current operation data group to obtain a time series, wherein the time series is an entity of changes in the health index over time; A prediction module, used for inputting the time series into a preset time series analysis model to obtain a preliminary prediction curve, wherein the preliminary prediction curve includes a current curve segment corresponding to the time series and a future curve segment predicting changes in health indicators in the future as the marked time progresses; A scoring module, used for scoring each historical operation data in the historical operation data group according to a scoring rule; A classification module, used for classifying the historical operation data into a plurality of similar condition groups according to the scores of the historical operation data; A first difference module is used to determine the difference between the change amount of the health indicator corresponding to the first historical operation data and the change amount of the health indicator corresponding to the second historical operation data as the change amount difference, the change amount is the difference between the health indicator corresponding to the historical operation data at a previous time and the health indicator corresponding to the historical operation data at a later identification time, the first historical operation data is any historical operation data in a similar condition group, and the second historical operation data is another historical operation data in the same similar condition group; A second difference module, configured to determine an environmental data difference between environmental data corresponding to the first historical operation data and environmental data corresponding to the second historical operation data as an environmental data difference; An impact data module, used for determining the difference of environmental data and the difference of transformation amount as impact data; A statistical module is used to collect statistics of all impact data to obtain an impact data set; The first training module is used to obtain a preset influencing factor model by using the influencing data set as a training sample; A compensation curve module, used to input the difference between the environment change table and the preset ideal environment change table into a preset influencing factor model to obtain a compensation curve, wherein the compensation curve is a compensation curve segment in which the compensation value changes with the marked time from the historical time to the future time; A compensation module, used for compensating the preliminary prediction curve using the compensation curve to obtain the prediction curve; The life prediction module is used to determine the battery life in the prediction curve according to the target time when it is determined that the target time is received.

8. A life prediction system for energy storage batteries, characterized in that: include: one or more processors and memory; The memory is coupled to the one or more processors, and the memory is used to store computer program code, wherein the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the life prediction system of the energy storage battery to execute the method as described in any one of claims 1-6.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the life prediction system of the energy storage battery, the life prediction system of the energy storage battery executes the method as described in any one of claims 1-6.

Images