New energy vehicle battery data analysis method and system

Abstract

The invention relates to the technical field of battery management, in particular to a new energy vehicle battery data analysis method and system, and the method comprises the following steps: merging a state feature set of a battery based on the charging and discharging period and use environment data of the new energy vehicle battery, extracting the voltage, current and temperature features of the state, and obtaining a state feature set; and carrying out feature combination on the battery degradation state, classifying and associating feature items and temperature features in the feature combination, and obtaining a state feature merging result. According to the method, through merging the new energy vehicle battery charging and discharging period and the use environment data, the battery state characteristics are accurately extracted, refined classification and correlation analysis are realized, the discrimination capability and adaptability of the state data are improved, and the state transition trend is flexibly captured based on the linkage trend of the state transition frequency and the current observation data; and dynamic evaluation is realized through correlation of periodic temperature and current, an accurate and reliable state change evaluation scheme is provided, and the dynamic response of battery management and the full-life periodic performance regulation and control quality are optimized.

Description

Technical Field

[0001] The present invention relates to the field of battery management technology, and in particular to a new energy vehicle battery data analysis method and system. Background Art

[0002] The field of battery management technology involves a series of technologies for monitoring, controlling and maintaining the performance and safety of battery systems. The core goal is to optimize the service life of the battery, improve its performance and ensure the safety of the use process. In the field of new energy vehicles, battery management monitors the battery's voltage, current, temperature and other parameters in real time to evaluate the battery's status, diagnose faults, and perform energy optimization, thermal management and balance control. Battery management technology also includes data analysis and processing technology to predict battery life and performance degradation, and support dynamic maintenance decisions, thereby improving energy efficiency while reducing maintenance costs and enhancing vehicle reliability.

[0003] Among them, the new energy vehicle battery data analysis method refers to the data processing technology used in new energy vehicle battery management, which aims to optimize the performance and life of the battery by collecting and analyzing battery operation data. The method includes data collection, processing and analysis to monitor battery status, predict performance degradation and life, evaluate usage efficiency, etc., which can achieve more accurate battery health management, optimize charging strategies, prevent overcharging or discharging, thereby improving the overall efficiency of the battery and extending its service life.

[0004] In practical applications, existing technologies mainly rely on real-time monitoring and fault warning of single voltage, current, and temperature parameters, which makes it difficult to achieve refined and systematic classification of multi-dimensional battery status, resulting in insufficient depth of status identification and trend analysis. Although existing technologies can perform basic status assessment and maintenance decisions through single parameter data, the capture of degradation status is mainly based on changes in a single indicator, and it is impossible to accurately identify slight differences in differentiated degradation states, making it difficult for the system to provide comprehensive warnings when dealing with complex working environments. Existing methods lack feature correlation capabilities and cross-parameter linkage analysis. This single-dimensional monitoring method easily leads to insufficient state transition monitoring, affecting the state prediction and performance optimization of the battery throughout its entire cycle. Existing periodic status assessment methods lack multi-dimensional data associations with charge and discharge cycles and usage environments, resulting in insufficient dynamic adjustment capabilities for battery management, making it difficult to achieve targeted assessments in special environments such as high temperature, cold, or high loads, ultimately affecting the overall reliability and predictability of battery management. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a new energy vehicle battery data analysis method and system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a new energy vehicle battery data analysis method, comprising the following steps: S1: Based on the charge and discharge cycle and usage environment data of new energy vehicle batteries, the battery state feature set is merged, the voltage, current and temperature features of the state are extracted, and the battery degradation state is feature combined, and the feature items and temperature features in the feature combination are classified and associated to obtain the state feature merging result; S2: Based on the state feature merging result, the state transition frequency of the battery under differentiated operating conditions is counted, the transition between states is calculated, and a battery state transition frequency table is established. According to the battery state transition frequency table, combined with the voltage, current, and temperature distribution relationship of the state observation value, the probability linkage relationship of the differentiated state is analyzed to generate a state and feature linkage table; S3: Based on the state and feature linkage table, the battery state is decoded by the features of the current observation data, the transfer trend of the current state is analyzed, and a battery state linkage trend combination is obtained. According to the battery state linkage trend combination, the relationship between the battery states is analyzed in combination with the voltage and current trends to generate a battery state dynamic analysis result; S4: Based on the dynamic analysis results of the battery state, combined with the newly added charge and discharge cycle data, a periodic evaluation is performed on the state transition and feature linkage relationship, and the temperature and current characteristics in the state transition are associated to generate a new energy vehicle battery state change evaluation plan.

[0007] As a further solution of the present invention, the step of obtaining the state feature merging result is specifically as follows: S111: based on the charge and discharge cycle and use environment data of the new energy vehicle battery, call the voltage, current and temperature data of the battery, segment the parameters, calculate the average values ​​of the voltage, current and temperature within the time period, and generate a battery feature sequence set; S112: Based on the battery feature sequence set, a normalization operation is performed on the voltage feature sequence and the temperature feature sequence, and a temperature weight coefficient, an adjustment coefficient and a gain coefficient are set, using the formula: ; Calculate the merged state eigenvalues ​​to obtain a preliminary feature combination sequence, where: represents the merged state eigenvalue, , , Respectively represent the normalized voltage value, current value, and temperature value, is the temperature characteristic weight coefficient, is the adjustment coefficient, is the temperature characteristic gain coefficient; S113: calling the preliminary feature combination sequence, analyzing the combination changes of voltage and temperature normalization features item by item, applying conditional judgment to analyze the degradation trend of feature items, selecting feature combination items that meet the merging conditions and storing them by category, and obtaining the state feature merging result.

[0008] As a further solution of the present invention, the step of obtaining the battery state conversion frequency table is specifically as follows: S211: Based on the state feature merging result, screening the characteristic changes of the battery under differentiated operating conditions, dividing the state features by time period, and calculating the state duration and condition parameters of frequent conversion by identifying the characteristic changes of voltage, current, and temperature in the differentiated time period, and establishing a state conversion statistics set; S212: Determine the frequency of state transitions according to the state transition statistics set, and according to the weight relationship between the number of transitions and the duration, by the formula: ; The state transition frequency value is calculated, where: Indicates the state transition frequency value, is the number of initial state transitions, is the influencing factor of differentiated operating conditions, is the state duration, and are the voltage and current characteristic parameters respectively, and is the weight coefficient of voltage, current and time characteristics, is the conversion adjustment coefficient, is the conversion determination parameter; S213: calling the state transition frequency value, distributing and classifying the state transition frequency according to differentiated operating conditions, counting and summarizing frequency changes between states, and establishing a battery state transition frequency table.

[0009] As a further solution of the present invention, the step of obtaining the state and feature linkage table is specifically as follows: S221: extracting conversion frequency data according to the battery state conversion frequency table and combining it with the state observation value, calling the observed distribution values ​​of voltage, current and temperature, performing segmented statistics on the feature changes corresponding to the battery state according to the observed data, merging the feature distribution mode under each state, and obtaining a feature association basic set; S222: According to the feature association basic set, the distribution relationship between the differentiated state transition frequency and the features is compared item by item, and the probability linkage value is calculated according to the change trend of voltage, current and temperature between differentiated states, using the formula: ; Frequency weight, distribution weight and adjustment coefficient are introduced to calculate the set of state and feature linkage values, where: Indicates the linkage value between state and feature, Represents the state transition frequency value, , are the observed voltage and current values, is the observed temperature value, , are the frequency and distribution weights, is the adjustment coefficient, is the determination parameter; S223: Based on the state and feature linkage value set, the feature linkage values ​​are classified and processed according to the differentiated states, and the probability association of the differentiated states is calculated according to the linkage conditions between the observed voltage, current, temperature characteristics and the states, and a state and feature linkage table is established.

[0010] As a further solution of the present invention, the steps of obtaining the battery status linkage trend combination are specifically as follows: S311: Based on the state and feature linkage table, the voltage, current and temperature features in the current observation data are captured, feature items that meet the linkage conditions are extracted, the voltage and current values ​​in the current state features are analyzed item by item, and the matching degree in the linkage table is compared, and the matching degree with the current state is determined by calculating the matching correlation degree of each feature, so as to form a state feature matching set; S312: According to the state feature matching set, according to the change rate of the observed feature, the formula is used: ; The state transition trend value set is calculated, where: Indicates the state transition trend value, is the voltage and current distribution of the state feature matching set, is the current observation state feature, is the characteristic change frequency, and are observation weight and linkage weight respectively, is the adjustment coefficient, is the trend value in the previous state transition; S313: Based on the state transition trend value set, analyze the linkage trend of the features, and gradually organize the trend items that meet the target state change pattern by associating the transition condition combinations between the states to generate a battery state linkage trend combination.

[0011] As a further solution of the present invention, the steps of obtaining the battery status dynamic analysis result are specifically as follows: S321: extracting characteristic voltage and characteristic current based on the battery status linkage trend combination, comparing the two characteristic values ​​according to time periods, calculating the voltage and current fluctuation amplitudes in each time period, and filtering and removing items with fluctuation amplitudes lower than a preset threshold, thereby generating a high fluctuation amplitude feature set of voltage and current; S322: Classify the trend data of voltage and current according to the high fluctuation amplitude feature set of voltage and current, and associate and match each group of high fluctuation amplitude feature items with the battery status, using the formula: ; Calculate the matching relationship between battery states and generate a dynamic matching table of battery states, where: Indicates the degree of match between battery states, is the trend period of battery state transition, is the influence coefficient corresponding to the battery status, represents the characteristic voltage, Indicates characteristic current; S323: Based on the battery state dynamic matching table, analyze the change trends of voltage and current item by item, perform segmented processing and calculation on the matching degree between the battery state and the characteristic value, and generate a battery state dynamic analysis result.

[0012] As a further solution of the present invention, the steps for obtaining the new energy vehicle battery state change evaluation scheme are specifically as follows: S411: Based on the battery state dynamic analysis result, extract the voltage, current and temperature characteristics in the newly added charge and discharge cycle, calculate the periodic voltage change rate and current change rate, combine the characteristic values, and associate with the temperature characteristics to perform preliminary characteristic analysis of state transition, and generate the battery characteristic combination result in the cycle; S412: Calculate the characteristic value difference and evaluate the influence of temperature on the state transition according to the battery characteristic combination result in the cycle, using the formula: ; Obtain the characteristic intensity of the charge and discharge cycle, where Indicates the characteristic intensity of the charge and discharge cycle, It represents the voltage change rate, represents the rate of change of current, is the temperature characteristic, is the environmental impact factor within the cycle; S413: According to the characteristic strength of the charge and discharge cycle, the linkage results in each charge and discharge cycle are integrated, the state changes of the temperature, current, and voltage characteristics in the cycle are analyzed, the stability and change trend of the differentiated cycle state transition are evaluated, and a new energy vehicle battery state change evaluation plan is generated.

[0013] A new energy vehicle battery data analysis system, the new energy vehicle battery data analysis system is used to execute the above-mentioned new energy vehicle battery data analysis method, the system comprises: The state merging module extracts the voltage, current and temperature data of the battery based on the charge and discharge cycle and usage environment data of the new energy vehicle battery, classifies and merges the feature items, integrates the battery data under differentiated operating environments, and obtains the state feature set merging result; The conversion frequency statistics module extracts the state conversion characteristics under the operating conditions based on the state feature set merging result, calculates the conversion frequency between the differentiated states, analyzes the frequency distribution and performs cross operation, determines the state linkage probability, and obtains the state conversion linkage frequency table; The state association analysis module analyzes the state association between the feature combinations based on the state transition linkage frequency table and in combination with the current voltage, current and temperature features, and obtains the state dynamic association features by analyzing the linkage relationship between the differentiated battery state features through state transition combination analysis; The trend analysis module extracts the state feature combination in the battery charge and discharge cycle data based on the state dynamic association feature, performs transfer analysis on the voltage and current change trends, calculates the transfer trend between the battery state features, and analyzes the linkage relationship between the features to generate a state trend decoding result; The cycle state assessment module is based on the state trend decoding result and combined with the charge and discharge cycle data, analyzes the state changes within the cycle, extracts the temperature and current characteristics in the cycle, calculates the cycle correlation, and obtains the new energy vehicle battery state change assessment plan.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by merging the charging and discharging cycle and usage environment data of new energy vehicle batteries, the state characteristics can be extracted more accurately, and the comprehensive analysis depth of the battery state can be improved. By using the combination of voltage, current, and temperature characteristics, not only can the feature associations be refined and grouped, but also the slight differences of the battery in various degradation states can be accurately captured to achieve refined classification and merging. This feature combination and grouping further improves the discrimination and adaptability of the state data. Through frequency statistics and analysis of transitions between states, a battery state transition frequency table reflecting the battery under differentiated operating conditions can be efficiently generated, thereby enhancing the correlation and mapping capabilities of the state data under multi-dimensional conditions. Based on the current The decoding of previous observation data and the combination of linkage trends can flexibly capture the transfer trend of the current battery state, so that the entire method can timely gain insight into the linkage relationship and change direction between battery states. The periodic evaluation realizes the dynamic evaluation of the battery state by correlating the temperature and current characteristics in the charging and discharging cycle, providing an efficient and reliable state change evaluation solution, improving the accuracy and dynamic response capability of battery management, and effectively integrating the merging of state feature data, conversion frequency statistics and linkage relationship analysis. It not only optimizes the identification and trend analysis of the battery state, but also can improve the dynamic performance regulation and quality of the battery's life cycle management through periodic temperature and current correlation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2 It is a flow chart of the state feature merging result in the present invention; Figure 3 A flow chart of a battery state conversion frequency table in the present invention; Figure 4 It is a flow chart of the state and feature linkage table in the present invention; Figure 5 A flow chart of the battery status linkage trend combination in the present invention; Figure 6 A flow chart showing the results of dynamic analysis of battery status in the present invention; Figure 7 It is a flow chart of the battery state change evaluation scheme for new energy vehicles in the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] 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.

[0018] Example 1: Please refer to Figure 1 The present invention provides a technical solution: a new energy vehicle battery data analysis method, comprising the following steps: S1: Based on the charge and discharge cycle and usage environment data of new energy vehicle batteries, the battery state feature set is merged, the voltage, current and temperature features of the state are extracted, and the battery degradation state is feature combined, and the feature items and temperature features in the feature combination are classified and associated to obtain the state feature merging result; S2: Based on the state feature merging results, the state transition frequency of the battery under differentiated operating conditions is counted, the transition between states is calculated, and a battery state transition frequency table is established. According to the battery state transition frequency table, combined with the voltage, current, and temperature distribution relationship of the state observation values, the probability linkage relationship of the differentiated states is analyzed to generate a state and feature linkage table; S3: Based on the state and feature linkage table, the battery state is decoded through the features of the current observation data, the transfer trend of the current state is analyzed, and the battery state linkage trend combination is obtained. According to the battery state linkage trend combination, the relationship between the battery states is analyzed in combination with the voltage and current trends to generate the battery state dynamic analysis result; S4: Based on the dynamic analysis results of the battery status, combined with the newly added charge and discharge cycle data, the state transition and feature linkage relationship are periodically evaluated, the temperature and current characteristics in the state transition are associated, and an evaluation plan for the battery status change of new energy vehicles is generated.

[0019] The state feature merging results include voltage feature set, current feature set, and temperature feature set. The battery state conversion frequency table includes conversion frequency, state distribution relationship, and state conversion matrix. The state and feature linkage table includes linkage probability, state feature combination, and temperature correlation relationship. The battery state dynamic analysis results include state transfer relationship, voltage trend characteristics, and temperature change characteristics. The battery state linkage trend combination includes current state characteristics, state transfer direction, and linkage trend parameters. The new energy vehicle battery state change evaluation plan includes conversion cycle evaluation, temperature linkage feature evaluation, and current change evaluation.

[0020] See also Figure 2 , the specific steps for obtaining the state feature merging results are: S111: based on the charge and discharge cycle and use environment data of the new energy vehicle battery, call the voltage, current and temperature data of the battery, segment the parameters, calculate the average values ​​of the voltage, current and temperature within the time period, and generate a battery feature sequence set; Extract the charge and discharge cycle data and usage environment parameters of new energy vehicle batteries, call the battery voltage, current and temperature data, perform preliminary data segmentation processing on various parameters based on the time period, and perform partition calculations according to daily measurement values ​​to ensure a fine data structure. For the voltage data of each time period, take the average and normalize it by hourly interval to obtain the characteristic data of each time period. The current and temperature data are also normalized to the relative standard range according to the same processing method. The characteristic mean set of voltage, current and temperature per hour is obtained by calculation, and the data is called to generate voltage characteristic sequence, current characteristic sequence and temperature characteristic sequence, and store them as a battery characteristic sequence set.

[0021] S112: Based on the battery feature sequence set, normalization operation of the voltage feature sequence and the temperature feature sequence is performed, and the temperature weight coefficient, adjustment coefficient and gain coefficient are set, using the formula: ; Calculate the merged state eigenvalues ​​to obtain a preliminary feature combination sequence, where: represents the merged state eigenvalue, , , Respectively represent the normalized voltage value, current value, and temperature value, is the temperature characteristic weight coefficient, is the adjustment coefficient, is the temperature characteristic gain coefficient; The benefit of the formula is that it introduces multiple merge calculations through temperature gain parameters, weight coefficients and adjustment coefficients, realizes the control of the influence of temperature characteristics on the overall state characteristics, and enhances the correlation analysis between parameters; In this formula, select the voltage value is 12V, current value For 100A, the temperature value is 300K, weight coefficient Set to 0.5, the gain factor Set to 2, adjustment coefficient Set to 0.1; First calculate : ; Then multiply the result by : ; Then calculate : ; Finally, find the overall square root: ; The results show that The calculation result reflecting the current state feature combination sequence represents the merged value of the current battery state feature, which is further used as a basis for judging the degradation trend.

[0022] S113: calling the preliminary feature combination sequence, analyzing the combination changes of voltage and temperature normalization features item by item, applying conditional judgment to analyze the degradation trend of feature items, selecting feature combination items that meet the merging conditions and storing them by category, and obtaining the state feature merging result; The feature combination sequence is called to analyze the combination changes of each voltage and temperature normalized feature item by item. The voltage and temperature parameter values ​​are gradually divided according to the periodic change interval and normalized and compared to form a normalized feature difference sequence. Each data group is sorted in sequence, and the stability index of each feature item within the time period is calculated. A change trend set including differentiated feature combination items is constructed, and the degradation state is judged item by item according to the obtained change trend. By comparing the continuity differences of each feature combination item in the feature item, the degradation degree and trend change of the combination item are analyzed to form a complete state feature merging result.

[0023] See also Figure 3 , the steps for obtaining the battery state conversion frequency table are as follows: S211: Based on the state feature merging result, the characteristic changes of the battery under differentiated operating conditions are screened, the state features are divided by time period, and the state duration and condition parameters of frequent conversions are calculated by identifying the characteristic changes of voltage, current, and temperature in the differentiated time period, and a state conversion statistical set is established; Based on the state feature merging results, the characteristic changes of the battery under differentiated operating conditions are screened, the state feature sequence set generated previously is called, the data is subdivided according to time periods, and the voltage, current, and temperature values ​​of each section are compared according to the time interval distribution. The characteristic changes of the current time period are judged according to the change rate and cumulative value of the characteristic items, and a set containing the characteristic change rate and conversion accumulation parameters within the time period is formed. The state characteristic items of each time period are classified according to the change frequency, and the threshold of frequent conversion is set. The value of high-frequency conversion under specific conditions is calculated to generate the cumulative frequency set of the corresponding state. According to the operating state and change rate of each characteristic item, the change set of the state duration is further calculated to obtain the state characteristic distribution of the current time period, identify the conditions for high-frequency conversion, and generate a preliminary state statistics set containing the state conversion frequency and conversion conditions.

[0024] S212: Determine the frequency of state transitions according to the state transition statistics set, and according to the weight relationship between the number of transitions and the duration, use the formula: ; The state transition frequency value is calculated, where: Indicates the state transition frequency value, is the number of initial state transitions, is the influencing factor of differentiated operating conditions, is the state duration, and are the voltage and current characteristic parameters respectively, and is the weight coefficient of voltage, current and time characteristics, is the conversion adjustment coefficient, is the conversion determination parameter; The benefit of the formula is that it introduces multiple weighted adjustment coefficients through the combination of voltage, current and state duration characteristics, and accurately controls the judgment conditions of the switching frequency; Set the initial state transition times =20, differential operating condition factor is 1.5, the duration of the state For 10 hours, voltage parameters For 12V, current parameters 100A, time weight is 0.8, and the voltage and current characteristic weights is 1.2, the conversion adjustment coefficient is 0.3, conversion judgment parameter is 5; calculate : ; Recalculate : ; Find the final frequency value: ; This result shows that the state transition frequency value It is 1.51314, which reflects the state transition of the battery under the current conditions and provides basic data for establishing the state transition frequency table.

[0025] S213: calling the state transition frequency value, distributing and classifying the state transition frequency according to differentiated operating conditions, counting and summarizing frequency changes between states, and establishing a battery state transition frequency table; Call the state transition frequency value, classify and organize the state transition frequencies under differentiated conditions according to differentiated operating conditions, sort them according to the high and low transition frequencies between states and the transition frequencies between differentiated conditions, build a frequency distribution matrix, merge the frequency sequences between differentiated states based on this matrix, compare and count the high-frequency and low-frequency conversion values ​​under specific conditions, summarize the conversion frequencies under each differentiated condition, form a distribution structure of state frequencies, and gradually summarize the basic data of the conversion frequency matrix in the frequency group corresponding to each operating condition, classify and record various types of frequency conversion information, and generate a state transition frequency table with differentiated operating conditions.

[0026] See also Figure 4 , the specific steps for obtaining the state and feature linkage table are: S221: extracting conversion frequency data according to the battery state conversion frequency table and combining it with the state observation value, calling the observed distribution values ​​of voltage, current and temperature, performing segmented statistics on the feature changes corresponding to the battery state according to the observed data, merging the feature distribution mode under each state, and obtaining the feature association basic set; According to the battery state conversion frequency table, the conversion frequency data is extracted and combined with the state observation value, the observed distribution values ​​of voltage, current and temperature are called, and each observation data is grouped by time period. Based on the frequency characteristics of the voltage, current and temperature changes in a specific period of time, the characteristic distribution pattern of the battery in the differentiated state is calculated, and the distribution trend of voltage and current in each state is analyzed. The frequency fluctuation characteristics are determined according to the mean and variance of the distribution. The frequency fluctuations are recorded and compared to obtain the characteristic change trend between the differentiated states. The time distribution of temperature is combined with the frequency table data to pair the observation value with the state conversion frequency. According to the time distribution characteristics of the observed voltage and temperature, the frequency information in each state is associated with the specific temperature parameters to generate a distribution feature set of voltage, current and temperature. Then, the data in the conversion frequency table is associated and integrated to obtain the state feature association basic set.

[0027] S222: According to the feature association basic set, the distribution relationship between the differentiated state transition frequency and the features is compared item by item, and the probability linkage value is calculated according to the change trend of voltage, current and temperature between differentiated states, using the formula: ; Frequency weight, distribution weight and adjustment coefficient are introduced to calculate the set of state and feature linkage values, where: Indicates the linkage value between state and feature, Represents the state transition frequency value, , are the observed voltage and current values, is the observed temperature value, , are the frequency and distribution weights, is the adjustment coefficient, is the determination parameter; The benefit of the formula is that, by introducing the frequency characteristics of differentiated voltage and current and combining the distribution weight of temperature, it can effectively reflect the linkage strength between differentiated states and is suitable for linkage relationship analysis under multi-dimensional conditions; In the example, the state transition frequency value is set 15 times / hour, observe the voltage value is 13V, observe the current value is 110A, observe the temperature value 310K, frequency weight is 0.7, distribution weight is 0.6, the adjustment coefficient is 0.8, the judgment parameter is 3; First calculate the absolute value of the difference between the voltage and current weights : ; Then calculate the frequency distribution of linkage strength : ; Finally, add the decision parameter and take the square root: ; This result shows that the linkage value It is 2.288, which reflects the strength of the feature linkage relationship in a specific state. It is the basic value of the state and feature linkage table, ensuring the rationality of the feature association between differentiated states.

[0028] S223: Based on the state and feature linkage value set, the feature linkage values ​​are classified and processed according to the differentiated states, and the probability association of the differentiated states is calculated according to the linkage conditions between the observed voltage, current, and temperature characteristics and the states, and a state and feature linkage table is established; The state and feature linkage value set is called, and each feature linkage value is gradually classified according to the state. First, the state conversion frequency and the change trend of the feature association value are sorted, and the conversion frequency of each state under the voltage, current and temperature characteristic distribution is extracted respectively. The temperature and voltage linkage data of each state are integrated one by one, and the probability association matrix of the linkage conditions is established. According to the fluctuation of temperature and current characteristics, the linkage strength between differentiated states is calculated item by item, and the conversion conditions between differentiated states are determined. Through weighted calculation of voltage and temperature characteristic frequencies, a linkage probability matrix containing frequency and distribution characteristics is established. The differentiated feature linkage conditions are combined into the column vector of the matrix, and the feature linkage relationship between states is classified and sorted. The state and feature linkage table is generated, which provides basic data support for further state analysis.

[0029] See also Figure 5 , the specific steps for obtaining the battery status linkage trend combination are: S311: Based on the state and feature linkage table, the voltage, current and temperature features in the current observation data are captured, the feature items that meet the linkage conditions are extracted, the voltage and current values ​​in the current state features are analyzed item by item, and the matching degree in the linkage table is compared. By calculating the matching correlation degree of each feature, the matching with the current state is determined to form a state feature matching set; Based on the state and feature linkage table, first collect the voltage, current and temperature characteristics of the battery in the current observation data, and screen them one by one to ensure the representativeness and accuracy of each feature in the observation data under the current operating conditions, extract the voltage and current distribution data, match the data with the preset distribution intervals in the linkage table one by one, ensure that the screened feature items meet the battery state association requirements in the linkage table, and obtain a preliminary feature item set. The feature item set is further compared. For the voltage and current data, the numerical range of each item is calculated with the reference data in the linkage table. By calculating the matching degree of each feature under differentiated states, the feature items with the best fitness are screened to form a feature set that adapts to the current battery state. The feature items in the adaptation set are rearranged to verify the consistency of the voltage, current and temperature features, and obtain a state feature matching set.

[0030] S312: According to the state feature matching set, according to the change rate of the observed feature, the formula is used: ; The state transition trend value set is calculated, where: Indicates the state transition trend value, is the voltage and current distribution of the state feature matching set, is the current observation state feature, is the characteristic change frequency, and are observation weight and linkage weight respectively, is the adjustment coefficient, is the trend value in the previous state transition; The benefit of the formula is that by introducing the observation weights , linkage weight And the adjustment coefficient , effectively balance the weight of the current feature item and the previous state transfer trend, and enhance the accurate description of the battery state linkage trend; Set feature matching set , the current state feature set , characteristic change frequency , the previous state transition trend value , weight , , ; First, calculate the weighted operation value of each parameter: ; ; ; Substitute the above results into the formula for calculation: ; The result shows that the calculated state transition trend value is 5.03, which provides a basis for further obtaining the battery state linkage trend combination.

[0031] S313: Analyze the linkage trend of the features based on the state transition trend value set, and gradually sort out the trend items that meet the target state change pattern by associating the transition condition combinations between the states to generate a battery state linkage trend combination; Based on the state transition trend value set and combined with the previously obtained state feature matching set, the feature items in the state feature matching set are screened again to extract the feature items that meet the transition trend value. First, the voltage, current, and temperature features are compared one by one to calculate their compatibility with the state feature matching set to obtain candidate feature items that meet the transition trend. The candidate feature items are sorted according to the degree of fitness, and the feature items with the highest fitness are screened to form a preliminary linkage feature set. The feature items in the set are sorted and classified, and the battery state feature combinations that meet the transfer trend are combined to construct a battery state linkage trend combination to fully reflect the feature association trend of the current battery state.

[0032] See also Figure 6 , the specific steps for obtaining the dynamic analysis results of the battery status are: S321: extracting characteristic voltage and characteristic current based on the battery status linkage trend combination, comparing the two characteristic values ​​according to time periods, calculating the voltage and current fluctuation amplitudes in each time period, and filtering to remove items with fluctuation amplitudes lower than a preset threshold, thereby generating a high fluctuation amplitude feature set of voltage and current; Based on the battery status linkage trend combination, the characteristic voltage is extracted from the voltage characteristic value set, and the characteristic current is extracted from the current characteristic value set. The two characteristic values ​​are compared according to the time period. First, the length of the time period is determined in order to unify the time series unit. By comparing the changes in voltage and current in each time period, the fluctuation amplitude of the voltage and current in each section is obtained. Then, each item in the fluctuation amplitude set is calculated, and the items with fluctuation amplitudes lower than the preset threshold are screened and eliminated to ensure the stability of high fluctuation items, forming a voltage and current feature set with high fluctuation amplitude. According to the obtained feature set, the time period of the high fluctuation amplitude feature is determined, and the voltage and current state when the fluctuation amplitude is the largest is taken as the representative feature in the section.

[0033] S322: Classify the trend data of voltage and current according to the high fluctuation amplitude feature set of voltage and current, and associate and match each group of high fluctuation amplitude feature items with the battery status, using the formula: ; Calculate the matching relationship between battery states and generate a dynamic matching table of battery states, where: Indicates the degree of match between battery states, is the trend period of battery state transition, is the influence coefficient corresponding to the battery status, represents the characteristic voltage, Indicates characteristic current; The benefit of the formula is that by combining the voltage and current characteristics and introducing the normalization of time and influence coefficients, the characteristic fluctuations within the differentiated time periods are made more comparable, effectively enhancing the accuracy of trend identification; Set characteristic voltage Volts and Characteristic Current Ampere, get trend period Hours and Impact Factor , substitute into the formula: ; Calculate the numerator first: ; Then calculate the denominator: ; Calculated : ; This result shows that the dynamic matching degree between battery states is 3.45, which can be used to further analyze the correlation between battery states.

[0034] S323: Based on the battery state dynamic matching table, analyze the change trends of voltage and current item by item, perform segmented processing and calculation on the matching degree between the battery state and the characteristic value, and generate a battery state dynamic analysis result; Based on the dynamic matching table of battery status, the high fluctuation amplitude feature set and the battery status linkage trend are combined for correlation analysis. Combined with the changing trend of voltage and current characteristic values, the battery status in differentiated time periods is matched and analyzed item by item, and the parts with higher state transition frequency are extracted. The voltage and current data are processed in segments, and the data are divided into multiple sets in the time series. The state transfer frequency in each segment is detected, and the voltage and current volatility of differentiated state characteristics is clustered. The battery status dynamic analysis results obtained after clustering characterize the change pattern of each battery status.

[0035] See also Figure 7 , the specific steps for obtaining the new energy vehicle battery state change assessment plan are: S411: Based on the results of the dynamic analysis of the battery state, extract the voltage, current and temperature characteristics in the newly added charge and discharge cycle, calculate the periodic voltage change rate and current change rate, combine the characteristic values, and associate them with the temperature characteristics to perform preliminary characteristic analysis of the state transition, and generate the battery characteristic combination result in the cycle; Based on the results of dynamic analysis of battery status, the voltage, current and temperature characteristic data of the newly added charge and discharge cycle are collected. First, the start and end time of each charge and discharge cycle are extracted from the data recording device. All voltage, current and temperature data of the time period are used as the core data set of the cycle. The voltage value in the data set is differentially calculated. The voltage change rate within the cycle is calculated by dividing the voltage difference between the start and end points of the cycle by the length of the time period. At the same time, the current change rate is obtained by performing the same operation on the current value. Then, the temperature characteristic data is called, and the average temperature change amplitude within the cycle is calculated in combination with the continuous reading records of the temperature sensor within the cycle. It is used as the temperature characteristic indicator. In the data processing process, the voltage and current change rates within the charge and discharge cycle are used to perform correlation analysis with the temperature change characteristics. The characteristic data set of each cycle is grouped into a cycle feature set, and the feature set within the cycle is summarized into a feature combination data. Based on this feature combination data, a preliminary feature analysis of state transition is performed to identify the linkage trend of battery characteristics within the cycle and generate the battery feature combination result within the cycle.

[0036] S412: Based on the battery feature combination results within the cycle, calculate the feature value difference and evaluate the impact of temperature on state transition, using the formula: ; Obtain the characteristic intensity of the charge and discharge cycle, where Indicates the characteristic intensity of the charge and discharge cycle, It represents the voltage change rate, represents the rate of change of current, is the temperature characteristic, is the environmental impact factor within the cycle; The benefit of the formula is that it effectively associates the voltage, current and temperature characteristics through multi-parameter interactive calculations, and further enhances the correlation accuracy of the battery characteristic strength through square root and absolute value operations, thereby improving the accuracy and correlation of the characteristic strength calculation within the cycle; set up (The rate of change of voltage within a cycle is obtained by dividing the voltage value increment by the length of the time period), (Current change rate within a cycle, time-weighted average of current change within a cycle), (temperature characteristics within the cycle, average value recorded by the temperature sensor within the cycle), (Environmental impact factors are estimated by monitoring environmental condition sensor parameters combined with historical data) Substitute each parameter into the formula: ; ; ; ; The result shows that the characteristic strength value of the periodic feature linkage relationship is 0.000235, indicating the linkage effect between the characteristic parameters within the period, which is used to further generate the periodic feature linkage relationship under the state transition condition.

[0037] S413: According to the intensity of the charge and discharge cycle characteristics, the linkage results in each charge and discharge cycle are integrated, the state changes of the temperature, current, and voltage characteristics in the cycle are analyzed, the stability and change trend of the differentiated cycle state conversion are evaluated, and a new energy vehicle battery state change evaluation plan is generated; Based on the linkage relationship of periodic characteristics, the linkage results in each charging and discharging cycle are integrated. First, the changing trends of voltage, current and temperature characteristics in each cycle are defined and a trend feature set is established. The trend features in each cycle are compared, and the periodic features with large differences are extracted as core indicators. After the trend feature set is integrated, the similarity analysis of the trends of linkage changes between cycles is performed, and the periodic features that conform to the linkage law are selected. The periodic linkage trend chart is established by marking and matching the feature changes of each cycle. This trend chart is used to describe the changing laws between differentiated cycles, and through the changes in the current, voltage and temperature characteristic states between each cycle, the periodic linkage trend is further converted into an evaluation of the battery state changes of new energy vehicles, forming an evaluation plan for the battery state changes of new energy vehicles.

[0038] The new energy vehicle battery data analysis system is used to execute the above-mentioned new energy vehicle battery data analysis method, and the system includes: The state merging module extracts the voltage, current and temperature data of the battery based on the charge and discharge cycle and usage environment data of the new energy vehicle battery, classifies and merges the feature items, integrates the battery data under differentiated operating environments, and obtains the state feature set merging result; The conversion frequency statistics module extracts the state conversion characteristics under the operating conditions based on the state feature set merging results, calculates the conversion frequency between differentiated states, analyzes the frequency distribution and performs cross operations, determines the state linkage probability, and obtains the state conversion linkage frequency table; The state association analysis module analyzes the state association between feature combinations based on the state transition linkage frequency table and the current voltage, current and temperature characteristics. By analyzing the state transfer combination, the linkage relationship between differentiated battery state features is analyzed to obtain the state dynamic association characteristics. The trend analysis module extracts the state feature combination in the battery charge and discharge cycle data based on the state dynamic correlation feature, performs transfer analysis on the voltage and current change trends, calculates the transfer trend between battery state features, analyzes the linkage relationship between features, and generates state trend decoding results; The cycle state assessment module is based on the state trend decoding results and combined with the charge and discharge cycle data to analyze the state changes within the cycle, extract the temperature and current characteristics in the cycle, calculate the cycle correlation, and obtain the new energy vehicle battery state change assessment plan.

[0039] 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 new energy vehicle battery data analysis method, characterized in that: The following steps are involved: Based on the charge and discharge cycle and usage environment data of new energy vehicle batteries, the battery state feature set is merged, the voltage, current and temperature features of the state are extracted, and the battery degradation state is feature combined, and the feature items and temperature features in the feature combination are classified and associated to obtain the state feature merging result; Based on the state feature merging result, the state transition frequency of the battery under differentiated operating conditions is counted, the transition between states is calculated, and a battery state transition frequency table is established. According to the battery state transition frequency table, combined with the voltage, current, and temperature distribution relationship of the state observation value, the probability linkage relationship of the differentiated state is analyzed to generate a state and feature linkage table; Based on the state and feature linkage table, the battery state is decoded through the features of the current observation data, the transfer trend of the current state is analyzed, and a battery state linkage trend combination is obtained. According to the battery state linkage trend combination, the relationship between the battery states is analyzed in combination with the voltage and current trends to generate a battery state dynamic analysis result; Based on the dynamic analysis results of the battery status, combined with the newly added charge and discharge cycle data, the state transition and feature linkage relationship are periodically evaluated, the temperature and current characteristics in the state transition are associated, and an evaluation plan for the battery status change of new energy vehicles is generated.

2. The new energy vehicle battery data analysis method according to claim 1, characterized in that: The steps for obtaining the state feature merging result are specifically as follows: Based on the charge and discharge cycle and usage environment data of new energy vehicle batteries, the voltage, current and temperature data of the batteries are called, the parameters are segmented, the average values ​​of voltage, current and temperature within the time period are calculated, and the battery feature sequence set is generated; Based on the battery feature sequence set, the voltage feature sequence and the temperature feature sequence are normalized, and the temperature weight coefficient, the adjustment coefficient and the gain coefficient are set, using the formula: ； Calculate the merged state eigenvalues ​​to obtain a preliminary feature combination sequence, where: represents the merged state eigenvalue, , , Respectively represent the normalized voltage value, current value, and temperature value, is the temperature characteristic weight coefficient, is the adjustment coefficient, is the temperature characteristic gain coefficient; The preliminary feature combination sequence is called, the combination changes of voltage and temperature normalization features are analyzed item by item, the degradation trend of feature items is analyzed by conditional judgment, feature combination items that meet the merging conditions are selected and stored by category, and the state feature merging result is obtained.

3. The new energy vehicle battery data analysis method according to claim 2, characterized in that: The steps for obtaining the battery state conversion frequency table are specifically as follows: Based on the state feature merging result, the characteristic changes of the battery under differentiated operating conditions are screened, the state features are divided by time period, and by identifying the characteristic changes of voltage, current, and temperature in the differentiated time period, the state duration and condition parameters of frequent conversions are calculated, and a state conversion statistics set is established; According to the state transition statistics set, the frequency between state transitions is determined, and according to the weight relationship between the number of transitions and the duration, the formula is used: ； The state transition frequency value is calculated, where: Indicates the state transition frequency value, is the number of initial state transitions, is the influencing factor of differentiated operating conditions, is the state duration, and are the voltage and current characteristic parameters respectively, and is the weight coefficient of voltage, current and time characteristics, is the conversion adjustment coefficient, is the conversion determination parameter; The state transition frequency value is called, the state transition frequency is distributed and classified according to differentiated operating conditions, the frequency changes between states are counted and summarized and recorded, and a battery state transition frequency table is established.

4. The new energy vehicle battery data analysis method according to claim 3, characterized in that: The steps for obtaining the state and feature linkage table are specifically as follows: According to the battery state conversion frequency table, the conversion frequency data is extracted and combined with the state observation value, the observed distribution values ​​of voltage, current and temperature are called, the feature changes corresponding to the battery state are statistically analyzed in sections according to the observed data, and the feature distribution mode under each state is merged to obtain the feature association basic set; According to the feature association basic set, the distribution relationship between the differentiated state transition frequency and the features is compared item by item, and the probability linkage value is calculated according to the change trend of voltage, current and temperature between differentiated states, using the formula: ； Frequency weight, distribution weight and adjustment coefficient are introduced to calculate the set of state and feature linkage values, where: Indicates the linkage value between state and feature, Represents the state transition frequency value, , are the observed voltage and current values, is the observed temperature value, , are the frequency and distribution weights, is the adjustment coefficient, is the determination parameter; Based on the state and feature linkage value set, the feature linkage values ​​are classified and processed according to the differentiated states, and the probability association of the differentiated states is calculated according to the linkage conditions between the observed voltage, current, temperature characteristics and the states, and a state and feature linkage table is established.

5. The new energy vehicle battery data analysis method according to claim 4, characterized in that: The steps for obtaining the battery status linkage trend combination are specifically as follows: Based on the state and feature linkage table, the voltage, current and temperature features in the current observation data are captured, feature items that meet the linkage conditions are extracted, the voltage and current values ​​in the current state features are analyzed item by item, and the matching degree in the linkage table is compared, and the matching degree with the current state is determined by calculating the matching correlation degree of each feature, so as to form a state feature matching set; According to the state feature matching set, according to the change rate of the observed feature, the formula is used: ； The state transition trend value set is calculated, where: Indicates the state transition trend value, is the voltage and current distribution of the state feature matching set, is the current observation state feature, is the characteristic change frequency, and are observation weight and linkage weight respectively, is the adjustment coefficient, is the trend value in the previous state transition; Based on the state transition trend value set, the linkage trend of the features is analyzed, and by combining the transition conditions between the associated states, the trend items that meet the target state change pattern are gradually sorted out to generate a battery state linkage trend combination.

6. The new energy vehicle battery data analysis method according to claim 5, characterized in that: The steps for obtaining the battery status dynamic analysis result are specifically as follows: Based on the battery status linkage trend combination, characteristic voltage and characteristic current are extracted, the two characteristic values ​​are compared according to time periods, the voltage and current fluctuation amplitudes in each time period are calculated, and the items with fluctuation amplitudes lower than a preset threshold are filtered out to generate a high fluctuation amplitude feature set of voltage and current; According to the high fluctuation amplitude feature set of the voltage and current, the trend data of the voltage and current are classified, and each group of high fluctuation amplitude feature items is associated and matched with the battery status, using the formula: ； Calculate the matching relationship between battery states and generate a dynamic matching table of battery states, where: Indicates the degree of match between battery states, is the trend period of battery state transition, is the influence coefficient corresponding to the battery status, represents the characteristic voltage, Indicates characteristic current; Based on the battery state dynamic matching table, the changing trends of voltage and current are analyzed item by item, the matching degree between the battery state and the characteristic value is processed and calculated in sections, and the battery state dynamic analysis result is generated.

7. The new energy vehicle battery data analysis method according to claim 6, characterized in that: The steps for obtaining the new energy vehicle battery state change assessment scheme are specifically as follows: Based on the battery state dynamic analysis result, extract the voltage, current and temperature characteristics in the newly added charge and discharge cycle, calculate the periodic voltage change rate and current change rate, combine the characteristic values, and associate the temperature characteristics to perform preliminary characteristic analysis of state transition, and generate the battery characteristic combination result in the cycle; Based on the combined results of the battery characteristics within the cycle, the characteristic value difference is calculated and the influence of temperature on the state transition is evaluated, using the formula: ； Obtain the characteristic intensity of the charge and discharge cycle, where Indicates the characteristic intensity of the charge and discharge cycle, It represents the voltage change rate, represents the rate of change of current, is the temperature characteristic, is the environmental impact factor within the cycle; According to the characteristic intensity of the charging and discharging cycle, the linkage results in each charging and discharging cycle are integrated, the state changes of the temperature, current, and voltage characteristics in the cycle are analyzed, the stability and change trend of the differentiated cycle state transition are evaluated, and a new energy vehicle battery state change evaluation plan is generated.

8. New energy vehicle battery data analysis system, characterized in that: According to any one of claims 1 to 7, the new energy vehicle battery data analysis method comprises: The state merging module extracts the voltage, current and temperature data of the battery based on the charge and discharge cycle and usage environment data of the new energy vehicle battery, classifies and merges the feature items, integrates the battery data under differentiated operating environments, and obtains the state feature set merging result; The conversion frequency statistics module extracts the state conversion characteristics under the operating conditions based on the state feature set merging result, calculates the conversion frequency between the differentiated states, analyzes the frequency distribution and performs cross operation, determines the state linkage probability, and obtains the state conversion linkage frequency table; The state association analysis module analyzes the state association between the feature combinations based on the state transition linkage frequency table and in combination with the current voltage, current and temperature features, and obtains the state dynamic association features by analyzing the linkage relationship between the differentiated battery state features through state transition combination analysis; The trend analysis module extracts the state feature combination in the battery charge and discharge cycle data based on the state dynamic association feature, performs transfer analysis on the voltage and current change trends, calculates the transfer trend between the battery state features, and analyzes the linkage relationship between the features to generate a state trend decoding result; The cycle state assessment module is based on the state trend decoding result and combined with the charge and discharge cycle data, analyzes the state changes within the cycle, extracts the temperature and current characteristics in the cycle, calculates the cycle correlation, and obtains the new energy vehicle battery state change assessment plan.

Images