Low-speed electric vehicle battery life assessment prediction method and system based on big data

Abstract

The present invention relates to the technical field of battery life prediction, specifically to a method and system for evaluating and predicting the battery life of a low-speed electric vehicle based on big data, comprising the following steps: based on a big data platform, collecting the operating data of the battery of a low-speed electric vehicle, calculating the entropy values ​​of the number of charge and discharge times, current, voltage and temperature parameters of the battery, analyzing the influence of various factors on the battery performance, and generating the battery parameter weight distribution results. In the present invention, by collecting and analyzing various operating parameters of the battery of a low-speed electric vehicle and calculating their entropy values, their influence on the battery performance is evaluated. By summarizing and sorting out the weight distribution results of these parameters, the use status of the battery under complex working conditions can be carefully reflected. Based on the analysis of this weight distribution, key performance indicators are extracted, and through the correlation between these indicators and the internal resistance of the battery and material aging, the core factors affecting the health status of the battery are effectively captured, thereby improving the precise control of the battery aging and decay process.

Description

Technical Field

[0001] The present invention relates to the technical field of battery life prediction, and in particular to a method and system for evaluating and predicting the battery life of a low-speed electric vehicle based on big data. Background Art

[0002] The low-speed electric vehicle battery life assessment and prediction method refers to building a suitable assessment and prediction model for the specific battery type used in low-speed electric vehicles based on its operating conditions and usage environment to accurately predict its remaining service life, helping manufacturers understand the health status of the battery.

[0003] The deficiencies of existing technologies in battery life assessment are mainly reflected in the fact that it is difficult to effectively consider the combined effects of ambient temperature, charge and discharge cycles, etc. on battery performance in its monitoring and analysis of battery operating conditions. Most of them are based on historical data for analysis, and fail to dynamically monitor and evaluate the health of the battery in real time, resulting in delayed evaluation results and not conducive to timely discovery of potential failure risks. At the same time, in terms of dealing with the transition of battery health status, existing technologies are also difficult to accurately capture changes in battery status under complex working conditions. This can easily lead to untimely or excessive maintenance of batteries, increasing maintenance costs and shortening the service life of batteries. In addition, it is difficult to effectively predict the trend of battery life changes under different conditions, resulting in a lack of flexibility and accuracy in practical applications. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a method and system for evaluating and predicting the battery life of a low-speed electric vehicle based on big data.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data, comprising the following steps:

[0006] S1: Based on the big data platform, collect the operating data of low-speed electric vehicle batteries, calculate the entropy value of battery charge and discharge times, current, voltage and temperature parameters, analyze the influence of various factors on battery performance, and generate battery parameter weight distribution results;

[0007] S2: Based on the battery parameter weight distribution results, select the key performance indicators of the battery, analyze the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the battery materials, determine the factors affecting the health status of the battery, and generate the key health status factor set of the battery;

[0008] S3: Based on the key health status factor set of the battery, the battery's health status, warning status and fault status are classified, the possibility of the battery changing from a healthy state to other states is analyzed, and the battery's state change trend is judged by combining the current operating parameters with the historical status data to obtain the battery's current health status and potential risk information;

[0009] S4: Based on the current health status and potential risk information of the battery, analyze the battery failure rate and service life according to the current operating conditions. By quantitatively analyzing the failure risk and analyzing the battery life distribution under different conditions, generate the battery life and failure rate evaluation results.

[0010] As a further solution of the present invention, the steps of analyzing the influence of various factors on battery performance are specifically as follows:

[0011] S111: Based on the big data platform, the battery charge and discharge times, current, voltage and temperature are monitored, abnormal values ​​are removed through filtering, and archived to obtain the battery operation data records;

[0012] S112: Based on the battery operation data record, use the formula: Calculate battery parameters The entropy value of the parameter is obtained;

[0013] in, is the number of values ​​that the parameter can take. is a parameter No. The probability distribution of the values, is the logarithm of the probability value;

[0014] S113: Extract key influencing factors based on the entropy information of the parameters, perform normalization processing, integrate weight information, and generate a battery parameter weight distribution result.

[0015] As a further solution of the present invention, the steps of analyzing the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the battery materials are specifically as follows:

[0016] S211: Based on the battery parameter weight distribution result, analyze the battery charge and discharge times, current, voltage and temperature fluctuations, take the parameters with higher entropy values ​​as monitoring objects, and generate a battery key performance indicator set;

[0017] S212: Based on the key performance indicator set of the battery, monitor the change of the internal resistance of the battery and the material aging rate, compare the correlation between the parameters and the increase of the internal resistance and the aging of the material, and generate the analysis results of the internal resistance of the battery and the aging rate of the material.

[0018] As a further solution of the present invention, the step of determining the factors affecting the battery health state is specifically:

[0019] S221: Based on the battery key performance indicator set, the formula is used: Calculate battery key performance indicators The weighted probability distribution of , obtain performance indicator information that affects the battery health status;

[0020] in, Is a key performance indicator The number of occurrences in the sample, is the total number of samples, is a weighting factor; S222: Based on the performance indicator information that affects the battery health status, regularly monitor the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value, record its change trend in different time periods, summarize the data, and generate a set of key battery health status factors.

[0021] As a further solution of the present invention, the steps of analyzing the possibility of the battery changing from a healthy state to other states are specifically as follows:

[0022] S311: Based on the key health status factor set of the battery, analyze the relationship between the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value and the battery operating parameters, set thresholds through historical data and compare with actual values, perform status classification, and generate battery status classification results;

[0023] S312: According to the battery status classification result, the formula is used: Calculate the battery status Transfer to state The weighted transition probability , and obtain the prediction result of battery state transition;

[0024] in, Is the battery from the state Transfer to state The number of times It is a state transfer arrive The weight of Is the battery in state The total number of times, Yes Status correction factor.

[0025] As a further solution of the present invention, the step of determining the state change trend of the battery is specifically as follows:

[0026] S321: Based on the prediction result of the battery state transition, combined with the historical state data and real-time operation parameters of the battery, a trend analysis is performed on the battery operation parameters to obtain a battery state change trend analysis result;

[0027] S322: Based on the battery status change trend analysis results, identify regular changes in the battery operation data, integrate all relevant data, and obtain the battery's current health status and potential risk information.

[0028] As a further solution of the present invention, the steps of analyzing the life distribution of batteries under different conditions are specifically as follows:

[0029] S411: Based on the current health status of the battery and potential risk information, use the formula: Calculation under specific operating conditions and time The conditional probability of battery failure is , obtain the failure probability information of the battery under specific conditions and time;

[0030] in, is the specific operating conditions in which the battery has failed The conditional probability of occurrence, is the overall probability of battery failure, is the time decay factor, is the time decay coefficient, Indicates the battery operating time, is a specific operating condition Probability of occurrence;

[0031] S412: Based on the failure probability information of the battery under specific conditions and time, the battery status risk level is analyzed in combination with the failure probability, real-time data is monitored, and battery life distribution analysis is performed to generate battery life and failure rate assessment results.

[0032] A low-speed electric vehicle battery life assessment and prediction system based on big data, the low-speed electric vehicle battery life assessment and prediction system based on big data is used to execute the above-mentioned low-speed electric vehicle battery life assessment and prediction method based on big data, and the system includes.

[0033] The data acquisition and processing module is based on a big data platform to collect the operating data of low-speed electric vehicle batteries, calculate the entropy value of each parameter, combine the historical operating data and entropy value information of the battery, process the weight ratio of each parameter, and generate the battery parameter weight distribution result;

[0034] The battery parameter weight analysis module selects the key performance indicators of the battery from the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth and temperature drift value based on the battery parameter weight distribution results, analyzes the relationship between the key performance indicators and the battery internal resistance and battery material aging rate, and obtains the battery key performance indicator analysis results;

[0035] The battery key performance indicator extraction module is based on the battery key performance indicator analysis results. Through the statistical distribution of the battery key performance indicators, the factors affecting the battery health status are determined, its performance under different states is analyzed, and a battery key health status factor set is generated;

[0036] The battery health status assessment module is based on the battery key health status factor set. By analyzing the current operating parameters and historical status data of the battery, it determines the battery's transition trend from a healthy state to other states and obtains the battery's current health status and potential risk information.

[0037] The battery life and failure rate assessment module quantitatively analyzes the battery failure risk and generates battery life and failure rate assessment results based on the battery's current health status and potential risk information, combined with current operating conditions, including ambient temperature and frequency of use.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are:

[0039] In the present invention, by collecting and analyzing various operating parameters of the low-speed electric vehicle battery and calculating its entropy value, its influence on the battery performance is evaluated. By summarizing the weight distribution results of these parameters, more accurate parameter analysis and processing are achieved, which can carefully reflect the use status of the battery under complex working conditions. Based on the analysis of this weight distribution, key performance indicators are extracted. Through the correlation between these indicators and the internal resistance of the battery and material aging, the core factors affecting the health status of the battery are effectively captured, thereby improving the precise control of the battery aging and decay process. In terms of the dynamic prediction of the battery health status, by combining historical data and the current state, the transition probability between different states is calculated, potential faults can be warned in advance, and the accuracy of battery status monitoring is improved. By analyzing the failure rate and life distribution under different operating conditions, the risk factors of the battery are quantified, and a comprehensive evaluation of the battery life is achieved. The overall processing logic comprehensively considers multiple influencing factors through dynamic analysis of data and status warning, accurately evaluates the health status and life of the battery, and optimizes the battery maintenance and use strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the workflow of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0042] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0043] See also Figure 1 The present invention provides a technical solution: a method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data, comprising the following steps:

[0044] S1: Based on the big data platform, collect the operating data of low-speed electric vehicle batteries, extract the battery charge and discharge times, current, voltage and temperature, calculate the entropy value of the parameters, process the weight ratio of each parameter according to the historical operation status and entropy value information of the battery, analyze the influence of various factors on battery performance, summarize the overall parameter weight information, and generate the battery parameter weight distribution result;

[0045] S2: Based on the battery parameter weight distribution results, select the battery key performance indicators from the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value, analyze the relationship between the battery key performance indicators and the battery internal resistance and battery material aging rate, determine the factors affecting the battery health status according to the statistical distribution of the battery key performance indicators, and generate the battery key health status factor set;

[0046] S3: Based on the key health status factor set of the battery, the battery's health status, warning status and fault status are classified, the transition probability under different states is calculated, the possibility of the battery changing from a healthy state to other states is analyzed, and the battery's state change trend is judged by combining the current operating parameters with the historical status data, and the current health status and potential risk information of the battery are obtained;

[0047] S4: Based on the current health status and potential risk information of the battery, analyze the battery failure rate and service life according to the current operating conditions. The current operating conditions include ambient temperature and usage frequency information. By quantitatively analyzing the failure risk and calculating the failure probability of the battery, the life distribution of the battery under different conditions is analyzed to generate the battery life and failure rate evaluation results.

[0048] The battery parameter weight allocation results include the weight distribution of charge and discharge times, current, voltage, and temperature and their impact on battery performance. The key battery health status factor set includes capacity loss, internal resistance increase, change in discharge depth, and energy conversion efficiency. The battery's current health status and potential risk information include the probability of the battery switching between healthy status, warning status, and fault status. The battery life and failure rate assessment results include the failure rate and life prediction under different ambient temperatures and usage intensities.

[0049] The specific steps for analyzing the impact of various factors on battery performance are:

[0050] S111: Based on the big data platform, the battery charge and discharge times, current, voltage and temperature are monitored, abnormal values ​​are removed through filtering, and archived to obtain the battery operation data records;

[0051] Based on the big data platform, the operating information of low-speed electric vehicle batteries can be obtained through distributed data collection. The acquisition of data mainly depends on the sensors in the battery management system (BMS). First, the sensors need to be calibrated and debugged to ensure that they can accurately monitor the number of charge and discharge cycles, current, voltage and temperature of the battery. The data are transmitted from the BMS to the data acquisition platform through the CAN bus. The collected data will be transmitted to the big data platform in real time and recorded according to the set time interval. The system will automatically perform a preliminary screening of the data to eliminate abnormal values, and then use the Kalman filtering method to extract the stable values ​​of the battery current, voltage and temperature. Kalman filtering is a technology that suppresses the noise of the signal through a recursive algorithm, which can effectively filter out high-frequency noise data. By analyzing the filtered data, the number of battery charge and discharge cycles is accumulated and counted by the software, and the current and voltage are processed by sampling and averaging multiple data points to obtain the final value. The temperature data is processed by the same sampling method and archived.

[0052] S112: Based on the battery operation data record, use the formula: Calculate battery parameters (such as charge and discharge times, current, voltage or temperature) to obtain the entropy value information of the parameters;

[0053] in, The entropy value is used to measure the degree of disorder of the parameter in historical data. A higher entropy value indicates that the data distribution is more random, while a lower entropy value indicates that the data is more concentrated or has a strong regularity. is the number of values ​​that the parameter can take (i.e. the number of partitions), is a parameter No. The probability distribution of a value represents the frequency of the parameter value in the interval. It is usually obtained by statistically analyzing historical data and normalizing it. The probability distribution can be obtained by frequency statistics. For example, historical data can be collected, the data can be grouped into intervals, and the frequency of each interval can be counted. Then, the frequency can be divided by the total data volume to obtain the probability distribution of each interval. The value of is the logarithm of the probability value.

[0054] Assume that the current data of a battery can be divided into three intervals, and the probability distribution of each interval is , , , then the entropy value of the battery current parameter is calculated as follows:

[0055] The results show that the entropy value of the current parameter is 0.473, which reflects the disorder of the battery current data during operation. A higher entropy value indicates that the current fluctuates greatly. A similar method can be used to calculate the entropy values ​​of the charge and discharge times, voltage, and temperature, which can be further used for weight analysis and life assessment.

[0056] S113: extracting key influencing factors according to the entropy information of the parameters, performing normalization processing, integrating weight information, and generating a battery parameter weight distribution result;

[0057] In the analysis of battery performance, multivariate regression analysis methods can be used to determine the correlation between various parameters and battery performance for the collected charge and discharge times, current, voltage and temperature data. First, the principal component analysis method is used to reduce the dimension of these data and extract the key parameters that mainly affect battery performance. By performing linear regression analysis on the reduced-dimensional data, the specific degree of influence of each parameter on the attenuation of battery performance can be obtained. For example, through regression analysis, it is found that when the temperature is too high, the discharge efficiency of the battery will be greatly reduced. The importance of the temperature parameter will be particularly prominent in this case. By normalizing the regression coefficient of each parameter, the weight of each parameter in the performance evaluation can be further clarified. Finally, all the weight information obtained from the analysis will be integrated.

[0058] The specific steps for analyzing the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the battery materials are as follows:

[0059] S211: Based on the battery parameter weight distribution result, analyze the battery charge and discharge times, current, voltage and temperature fluctuations, take the parameters with higher entropy values ​​as monitoring objects, and generate a battery key performance indicator set;

[0060] Based on the calculated entropy value of battery parameters The entropy value reflects the fluctuation and uncertainty of various battery parameters (such as charge and discharge times, current, voltage and temperature). By analyzing the entropy values ​​of these parameters, the operating stability of the battery can be evaluated.

[0061] When selecting key performance indicators of batteries, it is necessary to screen them according to the entropy value of each parameter.

[0062] The specific operation is to first select parameters with higher entropy values ​​as key monitoring objects according to the size of the entropy values, because these parameters have greater volatility and may have a greater impact on the performance of the battery;

[0063] Then, based on past experience, key performance indicators that have a significant impact on the health status of the battery are selected. These indicators include capacity decay rate, internal resistance change rate, energy efficiency, discharge depth and temperature drift value.

[0064] Among them, capacity decay rate: The capacity decay rate of a battery is the rate at which the battery's stored power decreases over time.

[0065] Entropy value of charge and discharge times It directly reflects the number and stability of battery fluctuations during the charging and discharging process.

[0066] If the entropy value of the charge and discharge times is large, it indicates that the charge and discharge cycles of the battery are unevenly distributed, which may be accompanied by faster capacity decay.

[0067] Therefore, the high entropy value of the charge and discharge times It can be used as a basis for selecting the key performance indicator of capacity attenuation rate.

[0068] Internal resistance change rate: The internal resistance change rate is the rate at which the internal resistance of the battery increases over time.

[0069] Entropy of current and voltage and This has a direct impact on this indicator.

[0070] If the entropy value of current or voltage is high, it means that the current and voltage fluctuate greatly during the charging and discharging process of the battery, which is usually related to the instability of the chemical reaction inside the battery and may accelerate the increase of the internal resistance of the battery.

[0071] Therefore, if the entropy of current or voltage is high, the rate of change of internal resistance may be a key performance indicator worth monitoring.

[0072] Energy efficiency: Energy efficiency is the ratio of energy stored to energy released by a battery during charging and discharging.

[0073] If the entropy of temperature A higher value means that the battery has large temperature fluctuations when operating under different temperature conditions, which may cause uneven reactions inside the battery and thus affect energy efficiency.

[0074] A high temperature entropy value means that the performance of the battery fluctuates greatly under different temperature environments, so energy efficiency can be used as an important performance indicator.

[0075] Depth of discharge: Depth of discharge is the percentage of total capacity released by the battery during each discharge cycle.

[0076] The depth of discharge is usually related to the number of charge and discharge cycles and current fluctuations of the battery.

[0077] If the entropy value of the charge and discharge times and current entropy A higher value indicates that the battery has greater uncertainty during these operations, which may lead to inconsistent depth of discharge.

[0078] At this point, the depth of discharge as an important performance indicator can help evaluate the actual working efficiency and life of the battery.

[0079] Temperature drift value: Temperature drift value is the performance fluctuation of the battery under different temperature conditions.

[0080] Entropy of Temperature The larger the value, the more unstable the battery is in an environment with large temperature fluctuations, and the impact of temperature on battery performance is more significant.

[0081] Therefore, the temperature drift value is very critical as a performance indicator, especially in extreme temperature environments.

[0082] By analyzing the entropy value of battery parameters , parameters with higher entropy values ​​can be prioritized because these parameters have higher uncertainties and may have a greater impact on the overall performance of the battery.

[0083] Then, the key performance indicators corresponding to each parameter are combined to finally select the appropriate performance indicators.

[0084] For example, if the entropy of temperature If the entropy value of the charge and discharge times is higher, energy efficiency and temperature drift value are preferred as key performance indicators; If it is higher, the capacity attenuation rate and discharge depth should be monitored.

[0085] S212: Based on the battery key performance indicator set, monitor the battery internal resistance change and material aging rate, compare the correlation between the parameters and the internal resistance increase and material aging, and generate the battery internal resistance and material aging rate analysis results;

[0086] When analyzing the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the materials, it is first necessary to monitor the changing trend of the internal resistance in detail. By regularly measuring the internal resistance value of the battery and recording the changing data, the increase in internal resistance usually means that the internal materials of the battery are aging, thus affecting the overall performance of the battery. Key performance indicators such as capacity decay rate, energy efficiency and temperature drift value also reflect the degree of material aging. By comparing these key performance indicators with the internal resistance data, the correlation between them can be found. For example, under the same usage time, the battery with increased internal resistance often has a higher capacity decay rate, a lower discharge depth, and a more obvious decrease in energy efficiency. In addition, the relationship between the temperature drift value and material aging is also very close. Under normal circumstances, abnormal temperature fluctuations will lead to accelerated aging of battery materials. Through comprehensive analysis of these data, the specific manifestation of the aging rate of battery materials can be obtained. In the analysis process, first record the changing trend of each key performance indicator, compare it with the internal resistance and material aging values, find the correlation between these parameters, and finally generate the analysis results of the battery internal resistance and material aging rate.

[0087] The steps to determine the factors that affect the battery health state are as follows:

[0088] S221: Based on the battery key performance indicator set, the formula is used: Calculate battery key performance indicators The weighted probability distribution of , obtain performance indicator information that affects the battery health status;

[0089] in, It indicates the proportion of a specific performance indicator in the overall sample, reflecting the importance of the indicator in the overall performance evaluation. Is a key performance indicator The number of occurrences in the sample indicates the frequency of the indicator appearing in the sample, which is usually obtained by statistically analyzing the collected battery performance data. For example, the battery status in the sample can be tested regularly, and the capacity attenuation rate, internal resistance change rate and other data in each test can be recorded. is the total number of samples, which indicates the total number of battery samples tested, usually obtained by testing the performance of the battery multiple times. is a weighting factor that indicates the degree of influence of a specific performance indicator under different environments and conditions. This factor can be dynamically adjusted based on different battery operating environments, load conditions, and life test results. For example, if the battery is used in a high temperature environment, the temperature drift value Higher values ​​will be assigned to reflect the significant effect that temperature fluctuations have on battery performance.

[0090] Assume that the number of occurrences of capacity decay rate from 100 battery samples is 40, the internal resistance change rate is 25, the energy efficiency is 20, the discharge depth is 10, and the temperature drift value is 5. The weighting factor of capacity decay rate is Set to 0.1, the weighting factor of the internal resistance change rate Set to 0.15, the weighting factor of energy efficiency Set to 0.2, the weighting factor of the discharge depth Set to 0.05, the weighting factor of the temperature drift value Set to 0.1.

[0091] Substituting the above data into the formula, we can get the weighted distribution of each key performance indicator:

[0092] Weighted distribution of capacity decay rate:

[0093]

[0094] Weighted distribution of internal resistance change rate:

[0095]

[0096] Weighted distribution of energy efficiency:

[0097]

[0098] Weighted distribution of discharge depth:

[0099]

[0100] Weighted distribution of temperature drift values:

[0101]

[0102] The results show that among the 100 samples, the weighted distribution of capacity attenuation rate is about 36.4%, the internal resistance change rate is about 21.8%, the energy efficiency is about 16.7%, the discharge depth is about 9.5%, and the temperature drift value is about 4.5%. Through these data, it is possible to identify which key performance indicators play a major role in battery life and performance evaluation.

[0103] S222: Based on the performance indicator information that affects the battery health status, regularly monitor the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value, record their change trends in different time periods, summarize the data, and generate a set of key battery health status factors;

[0104] Based on the weighted distribution of the key performance indicators of the battery, the key performance indicators with the largest distribution proportion are selected as the main factors affecting the health status of the battery. First, the key performance indicators are sorted. The weighted distribution of capacity decay rate is 36.4%, the internal resistance change rate is 21.8%, the energy efficiency is 16.7%, the discharge depth is 9.5%, and the temperature drift value is 4.5%. Therefore, the three indicators of capacity decay rate, internal resistance change rate and energy efficiency are selected as the main factors because of their largest proportion. These factors are the core data affecting the health status of the battery.

[0105] Next, it is necessary to monitor these three main factors in detail. The health status of the battery can be analyzed by regularly measuring the battery's capacity decay rate, internal resistance change and energy efficiency. The monitoring system can be used to record the changing trends of these factors over time in real time, and historical data can be obtained for analysis and comparison. In particular, it is necessary to observe how these factors change during the battery aging process, so as to further evaluate the battery life. Finally, these data are aggregated to generate a set of key battery health status factors to make a more accurate prediction and evaluation of the battery's health status.

[0106] The steps to analyze the possibility of a battery changing from a healthy state to another state are as follows:

[0107] S311: Based on the key health status factor set of the battery, analyze the relationship between the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value and the battery operating parameters, set thresholds through historical data and compare with actual values, perform status classification, and generate battery status classification results;

[0108] To classify the health status, warning status and fault status of the battery, it is first necessary to correlate and analyze the key health factors such as capacity decay rate, internal resistance change rate, energy efficiency, etc. with the various operating parameters of the battery. Then, set the threshold of each factor and determine the reasonable range of these thresholds based on the historical data collected previously. For example, the capacity decay rate can be determined by long-term monitoring of the battery discharge data to determine its change rate, and set the threshold of the capacity decay rate to a certain percentage. The internal resistance change rate determines a maximum change rate threshold based on the impedance change curve of the battery during the use cycle. The energy efficiency can be set according to the input-output ratio of the power during the charge and discharge cycle. By comparing the actual values ​​of these factors with the thresholds, the battery status is gradually classified. All key factors of the battery in a healthy state are kept within the set threshold range. The warning state is when one or more factors approach or reach the set threshold. If the values ​​of certain key factors exceed the set limit threshold, the battery is classified as a fault state.

[0109] S312: According to the battery status classification result, the formula is used: Calculate the battery status Transfer to state The weighted transition probability , and obtain the prediction result of battery state transition;

[0110] in, Is the battery from the state Transfer to state The number of times is obtained by monitoring and recording battery state conversion events through the battery management system (BMS). It is a state transfer arrive The weights are set based on the battery operating conditions and the severity of the transfer observed in past data, for example, by analyzing the failure data and the impact on operating efficiency. Is the battery in state The total number of times is similar to , by long-term monitoring of battery status and recording in the status The duration of Yes Status The correction factor, considering the state The specificity of the conditions and their impact on battery life can be set through historical data analysis and predictive models, such as using statistical analysis to determine which conditions have higher risk factors.

[0111] Records the number of times the battery changes from a healthy state to a warning state , the number of times you stay in a healthy state . Set the weight for transitioning to the warning state , the weight of warning to failure , correction factor for fault conditions .

[0112] Use the formula to calculate the transition probability of each state:

[0113] Health to early warning:

[0114]

[0115] Warning of failure:

[0116]

[0117] The calculation results show that the weighted transition probability of the battery from a healthy state to a warning state is about 54.5%, while the weighted transition probability from a warning state to a fault state is about 41.7%. These data are compared with historical probability benchmarks. If the previous probability is lower, it indicates that the battery health may be deteriorating.

[0118] The specific steps for determining the battery status change trend are:

[0119] S321: Based on the prediction result of the battery state transition, combined with the historical state data and real-time operation parameters of the battery, a trend analysis is performed on the battery operation parameters to obtain a battery state change trend analysis result;

[0120] Combine the current operating parameters with historical status data to analyze the possibility of the battery changing from a healthy state to other states. Use historical data and real-time monitoring data to establish a baseline model that uses machine learning technology to predict the health trend of the battery. Through the continuous collection and analysis of real-time data, such as battery temperature, voltage, and charge and discharge cycles, these parameters are cleaned and preprocessed in real time, such as removing noise and outliers. Then, statistical analysis methods such as time series analysis and regression analysis are used to perform trend analysis on the battery's operating parameters. Through these analyses, the current operating parameters are compared with the historical model to identify data points that deviate from the norm. These data points may indicate an imminent change in the battery state, such as an increased probability of transition. Showing that batteries tend to enter warning states more frequently, this comprehensive analysis helps predict the future performance of the battery and provide decision support for maintenance.

[0121] S322: According to the battery status change trend analysis result, identify the regular changes in the battery operation data, integrate all relevant data, and obtain the current health status and potential risk information of the battery;

[0122] First, the data collected from the battery management system (BMS) is aggregated, including but not limited to battery charge and discharge records, temperature logs, voltage changes, etc., and regular changes in the battery's operating data are identified through data mining techniques such as cluster analysis used in the data integration process. Usually refers to the behavior or state of the battery that recurs under normal or specific conditions, for example, the battery's charge and discharge performance within a specific temperature range, periodic fluctuations in voltage changes, stability of energy output, etc. The clustering results are compared with historical fault data to identify factors that may cause battery performance degradation. Integrate all relevant information and use predictive models to estimate the future state of the battery to generate information on the battery's current health status and potential risks, describing the battery's current health status and possible future risks. This information is critical for formulating battery maintenance plans and operating strategies.

[0123] The specific steps for analyzing the life distribution of batteries under different conditions are:

[0124] S411: Based on the current health status of the battery and potential risk information, use the formula: Calculation under specific operating conditions and time The conditional probability of battery failure is , obtain the failure probability information of the battery under specific conditions and time;

[0125] in, Used to assess the health of the battery under current operating conditions, is the specific operating conditions in which the battery has failed The conditional probability of occurrence is based on historical data statistics. For example, by analyzing the distribution of battery failures at different temperatures and usage frequencies, the relationship between operating conditions and failure occurrence is statistically analyzed. It is the overall probability of battery failure. This value is obtained by statistically analyzing the frequency of failures during the long-term monitoring of the battery life. It is a global probability reflection of historical data. is the time decay factor. As the battery is used for a longer time, the health of the battery will naturally decline, so the probability of failure increases with time. It is the time decay coefficient. This value represents the growth rate of the battery failure rate over time. It is estimated through the historical performance data of the battery in different operating time periods. It reflects the decay rate of the battery during aging. The value varies according to the battery type and working environment. Indicates the battery's operating time, which can be in hours, days, etc. The battery management system obtains this information by recording the battery's power-on time. It directly affects the trend of the failure probability over time. is a specific operating condition The probability of occurrence is determined by monitoring the frequency of different operating conditions during battery operation (such as different temperature ranges, load frequency, etc.) to determine the probability of occurrence of the operating conditions in the overall environment.

[0126] Assume that according to the operating conditions The statistical data and historical failure data are used to obtain the conditional probability , the overall probability of failure , the probability of a specific operating condition , and the time attenuation coefficient is obtained through historical aging data , and record the current running time Hour.

[0127] Substitute the formula to calculate the failure probability:

[0128]

[0129] Compute the time decay term:

[0130]

[0131] Then the overall calculation is performed:

[0132]

[0133] Calculation results It shows that under current operating conditions (e.g., ambient temperature and frequency of use) and after 100 hours of battery operation, the probability of battery failure is 0.14%.

[0134] S412: Analyze the battery status risk level based on the failure probability information of the battery under specific conditions and time, combine the failure probability, monitor the real-time data, perform battery life distribution analysis, and generate battery life and failure rate evaluation results;

[0135] First, according to the transition of the battery between the healthy state, warning state and fault state , combined with the probability of failure under operating conditions , conduct a comprehensive analysis of the battery life under different usage conditions and time. In specific implementation, the risk level of each state is determined by comparing the transition probability and conditional failure probability of the current battery state. For example, when The probability of the battery moving from a healthy state to a warning state is high, while the probability of failure is When it is lower than a certain threshold, it can be inferred that the battery status tends to the warning point but has not yet reached the critical point of failure. By monitoring the real-time data of operating conditions such as ambient temperature and frequency of use, and combining it with the historical state transition records, the battery life is distributed and analyzed, and the life curve under each usage condition is refined. In this process, data clustering analysis is used to identify the operating mode of the battery under different operating conditions. Based on the transition probability and failure risk of each state, the life distribution of each operating condition is quantified, and finally a battery life distribution diagram under different conditions is generated, which clearly shows the change process of the battery from normal to failure under specific operating conditions and time, and provides a reliable basis for subsequent battery maintenance decisions.

[0136] A low-speed electric vehicle battery life evaluation and prediction system based on big data, the low-speed electric vehicle battery life evaluation and prediction system based on big data is used to execute the above-mentioned low-speed electric vehicle battery life evaluation and prediction method based on big data, the system includes:

[0137] The data acquisition and processing module is based on a big data platform to collect the operating data of low-speed electric vehicle batteries, calculate the entropy value of each parameter, combine the historical operating data and entropy value information of the battery, process the weight ratio of each parameter, and generate the battery parameter weight distribution result;

[0138] The battery parameter weight analysis module selects the key performance indicators of the battery from the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth and temperature drift value based on the battery parameter weight distribution results, analyzes the relationship between the key performance indicators and the battery internal resistance and battery material aging rate, and obtains the battery key performance indicator analysis results;

[0139] The battery key performance indicator extraction module is based on the battery key performance indicator analysis results. Through the statistical distribution of the battery key performance indicators, the factors affecting the battery health status are determined, its performance under different states is analyzed, and a battery key health status factor set is generated;

[0140] The battery health status assessment module is based on the battery key health status factor set. By analyzing the current operating parameters and historical status data of the battery, it determines the battery's transition trend from a healthy state to other states and obtains the battery's current health status and potential risk information.

[0141] The battery life and failure rate assessment module quantitatively analyzes the battery failure risk and generates battery life and failure rate assessment results based on the battery's current health status and potential risk information, combined with current operating conditions, including ambient temperature and frequency of use.

[0142] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for evaluating and predicting the battery life of low-speed electric vehicles based on big data, characterized in that: The following steps are involved: S1: Based on the big data platform, collect the operating data of low-speed electric vehicle batteries, calculate the entropy value of battery charge and discharge times, current, voltage and temperature parameters, analyze the influence of various factors on battery performance, and generate battery parameter weight distribution results; S2: Based on the battery parameter weight distribution results, select the key performance indicators of the battery, analyze the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the battery materials, determine the factors affecting the health status of the battery, and generate the key health status factor set of the battery; The steps to determine the factors that affect the battery health state are as follows: S221: Based on the battery key performance indicator set, the formula is used: Calculate battery key performance indicators The weighted probability distribution of , obtain performance indicator information that affects the battery health status; in, Is a key performance indicator The number of occurrences in the sample, is the total number of samples, is the weighting factor; S222: Based on the performance indicator information that affects the battery health status, regularly monitor the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value, record their change trends in different time periods, summarize the data, and generate a set of key battery health status factors; S3: Based on the key health status factor set of the battery, the battery's health status, warning status and fault status are classified, the possibility of the battery changing from a healthy state to other states is analyzed, and the battery's state change trend is judged by combining the current operating parameters with the historical status data to obtain the battery's current health status and potential risk information; The steps to analyze the possibility of a battery changing from a healthy state to another state are as follows: S311: Based on the key health status factor set of the battery, analyze the relationship between the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth or temperature drift value and the battery operating parameters, set thresholds through historical data and compare with actual values, perform status classification, and generate battery status classification results; S312: According to the battery status classification result, the formula is used: Calculate the battery status Transfer to state The weighted transition probability , and obtain the prediction result of battery state transition; in, Is the battery from the state Transfer to state The number of times, It is a state transfer arrive The weight of Is the battery in state The total number of times, Yes Status Correction factor of S4: Analyze the battery failure rate and service life according to the current operating conditions based on the current health status and potential risk information of the battery. By quantitatively analyzing the failure risk and analyzing the battery life distribution under different conditions, generate the battery life and failure rate evaluation results. The specific steps for analyzing the life distribution of batteries under different conditions are: S411: Based on the current health status of the battery and potential risk information, use the formula: Calculation under operating conditions and time The conditional probability of battery failure is , obtain the failure probability information of the battery under the operating conditions and time; in, In the event that the battery has failed, the operating conditions The conditional probability of occurrence, is the overall probability of battery failure, is the time decay factor, is the time decay coefficient, Indicates the battery operating time, is the operating condition Probability of occurrence; S412: Analyze the battery status risk level based on the failure probability information of the battery under the operating conditions and time in combination with the failure probability, monitor real-time data, perform battery life distribution analysis, and generate battery life and failure rate assessment results.

2. The method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data according to claim 1 is characterized in that: The specific steps for analyzing the impact of various factors on battery performance are: S111: Based on the big data platform, the battery charge and discharge times, current, voltage and temperature are monitored, abnormal values ​​are removed through filtering, and archived to obtain the battery operation data records; S112: Based on the battery operation data record, use the formula: Calculate battery parameters The entropy value of the parameter is obtained; in, is the number of values ​​that the parameter can take. is a parameter No. The probability distribution of the values, is the logarithm of the probability value; S113: Extract key influencing factors based on the entropy information of the parameters, perform normalization processing, integrate weight information, and generate a battery parameter weight distribution result.

3. The method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data according to claim 2 is characterized in that: The specific steps for analyzing the relationship between the key performance indicators of the battery and the internal resistance of the battery and the aging rate of the battery materials are as follows: S211: Based on the battery parameter weight distribution result, analyze the battery charge and discharge times, current, voltage and temperature fluctuations, take the parameters with higher entropy values ​​as monitoring objects, and generate a battery key performance indicator set; S212: Based on the key performance indicator set of the battery, monitor the change of the internal resistance of the battery and the material aging rate, compare the correlation between the parameters and the increase of the internal resistance and the aging of the material, and generate the analysis results of the internal resistance of the battery and the aging rate of the material.

4. The method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data according to claim 1 is characterized in that: The specific steps for determining the battery status change trend are: S321: Based on the prediction result of the battery state transition, combined with the historical state data and real-time operation parameters of the battery, a trend analysis is performed on the battery operation parameters to obtain a battery state change trend analysis result; S322: Based on the battery status change trend analysis results, identify regular changes in the battery operation data, integrate all relevant data, and obtain the battery's current health status and potential risk information.

5. A low-speed electric vehicle battery life assessment and prediction system based on big data, characterized in that: According to any one of claims 1 to 4, the method for evaluating and predicting the battery life of a low-speed electric vehicle based on big data comprises: The data acquisition and processing module is based on a big data platform to collect the operating data of low-speed electric vehicle batteries, calculate the entropy value of each parameter, combine the historical operating data and entropy value information of the battery, process the weight ratio of each parameter, and generate the battery parameter weight distribution result; The battery parameter weight analysis module selects the key performance indicators of the battery from the capacity decay rate, internal resistance change rate, energy efficiency, discharge depth and temperature drift value based on the battery parameter weight distribution results, analyzes the relationship between the key performance indicators and the battery internal resistance and battery material aging rate, and obtains the battery key performance indicator analysis results; The battery key performance indicator extraction module is based on the battery key performance indicator analysis results. Through the statistical distribution of the battery key performance indicators, the factors affecting the battery health status are determined, its performance under different states is analyzed, and a battery key health status factor set is generated; The battery health status assessment module is based on the battery key health status factor set. By analyzing the current operating parameters and historical status data of the battery, it determines the battery's transition trend from a healthy state to other states and obtains the battery's current health status and potential risk information. The battery life and failure rate assessment module quantitatively analyzes the battery failure risk and generates battery life and failure rate assessment results based on the battery's current health status and potential risk information, combined with current operating conditions, including ambient temperature and frequency of use.

Images