New energy automobile battery health state assessment method and system

Abstract

The invention relates to the field of battery state evaluation, in particular to a new energy automobile battery health state evaluation method and system. The method comprises the following steps: acquiring a vehicle battery multi-mode state monitoring parameter and a historical vehicle endurance mileage log; performing battery state dynamic evolution on the vehicle battery multi-mode state monitoring parameters to obtain battery multi-mode state evolution characteristics; carrying out multi-time-point maximum endurance calculation on the historical vehicle endurance mileage log, carrying out time sequence trend evolution, and constructing a vehicle endurance evolution trend chart; performing nonlinear attenuation trend analysis on the vehicle endurance evolution trend graph, and performing battery aging mechanism evolution analysis so as to construct a battery state aging evolution knowledge graph; according to the battery state aging evolution knowledge graph, environment change influence analysis is carried out on the vehicle endurance evolution trend graph, dynamic vehicle state modeling is carried out, and a dynamic vehicle twinborn model is constructed. According to the invention, efficient and accurate battery health state evaluation is realized.

Description

Technical Field

[0001] The present invention relates to the field of battery state evaluation, and particularly to a method and system for evaluating the health state of new energy vehicle batteries. Background Art

[0002] With the increasingly serious global environmental problems and the transformation of the energy structure, new energy vehicles (NEVs), as a green and environmentally friendly means of transportation, have gradually become an important direction for future transportation development. As the core component of new energy vehicles, the performance of the battery directly affects the cruising range, safety, and reliability of the whole vehicle. Therefore, the evaluation of the health state of the battery has become one of the key technologies in the research, development, production, and maintenance of new energy vehicles. The evaluation method of the state of health (SOH) of new energy vehicle batteries is not only of great significance for battery life management, optimization of charge and discharge strategies, and improvement of vehicle operation efficiency, but also provides necessary support for battery fault prediction, risk assessment, and safety guarantee.

[0003] New energy vehicle batteries usually use lithium-ion batteries or other types of chemical batteries. These batteries will be affected by various factors during long-term use, including the depth of charge and discharge, the operating temperature of the battery, the change of load, and the number of usage cycles. As the number of uses increases, the internal chemical reactions of the battery gradually age, and parameters such as the voltage, capacity, and internal resistance of the battery will change, ultimately leading to the decline of battery performance. The irreversibility of this process determines that the evaluation of the health state of the battery has important requirements for real-time and accuracy.

[0004] Traditional evaluation methods for the health state of new energy vehicle batteries usually rely on some simple battery parameters, such as voltage, current, temperature, and capacity, for preliminary judgment, and often based on empirical formulas and simple rule models. This evaluation method provides an incomplete judgment of the overall health state of the battery and has low accuracy, and cannot reflect the complex changes that occur in the battery during actual use in real time and dynamically. Therefore, a more intelligent and automated analysis and evaluation method for new energy vehicle batteries is needed. Summary of the Invention

[0005] To solve the above technical problems, the present invention proposes a method and system for evaluating the health state of new energy vehicle batteries to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a method for evaluating the health state of new energy vehicle batteries, including the following steps: Step S1: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle cruising range log; perform dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain the multi-modal state evolution characteristics of the battery; Step S2: Calculate the maximum cruising range at multiple time points for the historical vehicle cruising range log, and conduct a time-series trend evolution to construct a vehicle cruising range evolution trend graph; Step S3: Conduct a non-linear decay trend analysis on the vehicle cruising range evolution trend graph, and conduct an evolution analysis of the battery aging mechanism, so as to construct a battery state aging evolution knowledge graph; Step S4: According to the battery state aging evolution knowledge graph, conduct an environmental change impact analysis on the vehicle cruising range evolution trend graph, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; Step S5: Conduct multiple-period cruising range evolution simulations on the dynamic vehicle twin model, and conduct dynamic cruising range rolling predictions, so as to obtain long-term cruising range trend prediction data; Step S6: Conduct battery life state prediction fitting on the long-term cruising range trend prediction data, and conduct a quantitative evaluation of the comprehensive battery health state to generate a battery health state evaluation result.

[0007] The present invention obtains multi-modal state monitoring parameters of the vehicle battery and the historical vehicle cruising range log, conducts a dynamic evolution analysis on the battery state, reveals the evolution characteristics of the battery multi-modal state. The acquisition of the battery multi-modal state evolution characteristics helps to understand the state change law of the battery under different working conditions, and provides a basis for subsequent health state evaluation. By calculating the maximum cruising range at multiple time points and conducting time-series trend evolution analysis, a vehicle cruising range evolution trend graph is constructed to reveal the change trend of the vehicle cruising range ability. The construction of the vehicle cruising range evolution trend graph helps to understand the dynamic change of the vehicle cruising range performance and provides a reference for subsequent health state evaluation. Through non-linear decay trend analysis and battery aging mechanism evolution analysis, a battery state aging evolution knowledge graph is constructed to deeply understand the battery aging mechanism and state evolution law. The construction of the battery state aging evolution knowledge graph helps to reveal the key factors and laws in the battery aging process and provides theoretical support for subsequent battery health state evaluation. According to the battery state aging evolution knowledge graph, environmental change impact analysis and dynamic vehicle state modeling are conducted to construct a dynamic vehicle twin model to simulate the evolution of the vehicle state under different conditions. The construction of the dynamic vehicle twin model helps to simulate the state change of the vehicle in actual operation and provides a more accurate model support for health state evaluation. Through conducting multiple-period cruising range evolution simulations and dynamic cruising range rolling predictions on the dynamic vehicle twin model, long-term cruising range trend prediction data is obtained to predict the long-term change trend of the vehicle cruising range ability. The acquisition of the long-term cruising range trend prediction data helps to formulate reasonable maintenance and management strategies to ensure the long-term stable performance of the vehicle. Conduct battery life state prediction fitting and quantitative evaluation of the comprehensive battery health state on the long-term cruising range trend prediction data to generate a battery health state evaluation result to accurately evaluate the health state of the battery. The generation of the battery health state evaluation result helps to detect battery problems in advance and take corresponding measures to extend the battery life and improve the vehicle performance.

[0008] Preferably, step S1 includes the following steps: Step S11: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle endurance mileage log; Step S12: Filter out abnormal parameters from the multi-modal state monitoring parameters of the vehicle battery to obtain filtered and optimized battery state monitoring parameters; Step S13: Analyze the time-sequence fluctuation of the driving voltage of the filtered and optimized battery state monitoring parameters to generate the battery voltage fluctuation characteristics of the driving state; Step S14: Perform discrete fitting on the battery capacity attenuation of the filtered and optimized battery state monitoring parameters to construct a battery capacity attenuation curve; Step S15: Perform dynamic evolution of the battery state on the battery voltage fluctuation characteristics of the driving state and the battery capacity attenuation curve to obtain the multi-modal state evolution characteristics of the battery.

[0009] The present invention provides comprehensive and systematic data support for subsequent health assessment by obtaining multi-modal monitoring parameters of the battery (such as battery voltage, current, temperature, internal resistance, etc.) and historical driving range logs. The multi-dimensional data input can more accurately reflect the actual working state and performance of the battery. The historical vehicle driving range logs provide long-term operation data for the model, helping to evaluate the performance of the battery at different usage stages and revealing the actual impact of battery degradation on the driving range. These data provide data support for subsequent analysis of the battery's health status, degradation characteristics, and future driving range capabilities, ensuring the scientificity and accuracy of the model. In the multi-modal monitoring data, there are abnormal data caused by various external interference factors (such as sensor failures, abnormal data acquisition, etc.). By filtering abnormal parameters from the battery state monitoring parameters, these noises can be removed, ensuring the accuracy and reliability of the data. The optimized battery monitoring parameters can more truly reflect the actual operating conditions of the battery, reducing errors caused by abnormal data, thereby improving the stability and prediction accuracy of the health status assessment model. The filtered data is more consistent and available, providing high-quality data support for subsequent analysis and further improving the reliability of the entire assessment process. The fluctuation of the battery voltage reflects the dynamic state of the battery during driving. Through the analysis of the temporal fluctuation of the driving voltage, the performance fluctuation characteristics of the battery under different driving conditions (such as acceleration, braking, load changes, etc.) can be effectively captured. The driving state battery voltage fluctuation characteristics help analyze the load response of the battery during actual operation and reveal potential problems such as over-discharge and charging imbalance of the battery. This characteristic provides strong evidence for subsequent battery health assessment. By analyzing the voltage fluctuation characteristics, the system can issue a warning in a timely manner when the battery shows abnormal fluctuations, identifying the degradation trend of the battery in advance, which helps to maintain and optimize the battery management strategy in a timely manner. Combining the battery voltage fluctuation characteristics with the capacity decay curve, the multi-modal state evolution characteristics of the battery under different working states can be obtained. This multi-dimensional dynamic evolution analysis provides a more comprehensive and in-depth assessment of the battery health status. By modeling the dynamic evolution of the battery state, the battery degradation modes under different usage scenarios can be identified, and potential problems of the battery, such as capacity decay, efficiency decline, overheating, etc., can be discovered in advance, thereby avoiding greater performance losses. The dynamic evolution characteristics of the battery state make the change trend of the battery health status clearer, providing long-term and dynamic prediction data for the battery management system and providing a scientific basis for battery maintenance and management. This dynamic evolution method can better adapt to complex and changeable working environments, predicting the performance of the battery under different loads, temperatures, and charging conditions, thus achieving more accurate battery health status assessment and long-term driving range prediction.

[0010] Preferably, the specific steps of step S14 are as follows: Collect the charge and discharge battery states of the filtered and optimized battery state monitoring parameters to obtain multiple charge and discharge battery state parameters; Analyze the change of battery internal resistance for multiple charge-discharge battery state parameters to generate the characteristics of battery internal resistance change; Evaluate the real-time charge transfer performance based on the characteristics of battery internal resistance change, so as to obtain multiple real-time charge transfer performance evaluation values; Perform polarization voltage evolution on multiple charge-discharge battery state parameters to obtain polarization voltage time-series evolution parameters; Calculate the change of battery capacity in a sliding time window for the polarization voltage time-series evolution parameters and multiple real-time charge transfer performance evaluation values to generate battery capacity change data for multiple time windows; Conduct battery capacity decay analysis based on the battery capacity change data for multiple time windows to obtain battery capacity decay parameters for multiple time windows; Perform discrete time-series fitting on the battery capacity decay parameters for multiple time windows to construct a battery capacity decay curve.

[0011] The present invention collects charge and discharge states by using optimized battery state monitoring parameters, obtains the detailed performance of the battery during actual use, and comprehensively reflects the operation of the battery under different loads and charge and discharge states. This provides multi-dimensional data for the accurate assessment of the battery health state. The state changes during the charge and discharge process reflect the performance fluctuations of the battery in real time, capture the degradation patterns of the battery under extreme conditions such as high load and long-term charge and discharge, and further improve the accuracy of health assessment. By collecting multiple charge and discharge battery state parameters in real time, more real-time information is provided for the battery management system (BMS), enhancing the scientific nature and accuracy of battery management decisions, helping the system automatically adjust charging and discharging strategies. The internal resistance of the battery is one of the important parameters for measuring the battery health state, and its change is often an early signal of battery degradation. By analyzing the change of internal resistance of multiple charge and discharge battery state parameters, the internal degradation trend of the battery can be clearly identified, especially after long-term use, the impact of internal resistance change on battery capacity and efficiency. The change of internal resistance can not only reflect the battery health state, but also reveal the charge transfer efficiency of the battery. When the internal resistance is too high, the charge and discharge efficiency and endurance of the battery will decline. Therefore, analyzing the change characteristics of internal resistance helps to predict battery degradation in advance, optimize charging strategies and usage patterns. By evaluating the real-time charge transfer performance through the change characteristics of battery internal resistance, the efficiency performance of the battery under different working states can be measured more accurately. Especially when the battery is in different charge and discharge states, the charge transfer performance reflects the energy conversion efficiency of the battery and directly affects the endurance performance. This step dynamically evaluates the energy transfer efficiency of the battery during operation, revealing whether there is a low charge transfer efficiency under specific environments or loads. This provides data support for the subsequent quantification of the battery health state. By continuously evaluating the real-time charge transfer performance of the battery, the battery management system dynamically adjusts charging strategies, load distribution, etc. according to the evaluation results, further improving the battery performance and extending its service life. The polarization voltage is one of the internal resistance effects generated during the charge and discharge process of the battery, which can reveal the internal inhomogeneity that appears after long-term charge and discharge of the battery. By analyzing the evolution of the polarization voltage of the charge and discharge battery state parameters, an in-depth understanding of the electrochemical behavior and energy storage effect of the battery under different working conditions can be obtained. The time series evolution of the polarization voltage reflects the balance between the charging efficiency of the battery and the internal electrochemical reaction. When the polarization voltage of the battery is abnormal, it is often a precursor to internal degradation and capacity attenuation of the battery. Through this analysis, the degradation problem of the battery can be detected in time to avoid further performance decline. The time series evolution parameters of the polarization voltage, as a sensitive index, can effectively track the aging process of the battery and assist in the assessment of the battery health state, providing reliable data support. The calculation of the sliding time window battery capacity change can help analyze the capacity change of the battery in different time periods. By setting multiple time windows, the speed and trend of battery capacity degradation can be tracked more carefully, capturing small changes, so as to achieve more accurate degradation prediction. Through the sliding window technology,Implementing high-frequency battery capacity monitoring to promptly detect minor changes in battery capacity is of great significance for enhancing the real-time monitoring capabilities of battery management systems and optimizing battery usage and maintenance strategies. The sliding time window method is flexible and adaptable to different battery usage scenarios and requirements. For example, the length of the time window can be adjusted according to the charging frequency and load size, enabling the calculation of battery capacity changes to better fit the actual usage situation. By analyzing the battery capacity change data of multiple time windows, the capacity decay trend of the battery can be more comprehensively captured. This analysis method based on multiple time windows can reveal the decay rate and pattern of the battery under different usage conditions and load situations, further improving the accuracy of battery life prediction. Compared with the traditional single-time-point capacity decay analysis, the analysis based on multiple time windows can obtain more refined decay characteristics, thus achieving a more accurate capacity decay prediction, helping to extend the battery life and reduce sudden failures. The obtained battery capacity decay parameters can assist in optimizing charging, discharging, and maintenance strategies. For example, the battery charging frequency and load situation can be dynamically adjusted according to the decay trend to avoid accelerated decay. By discrete time series fitting, a more accurate battery capacity decay curve can be constructed. This curve can comprehensively reflect the capacity decay process of the battery, including the decay trend under different load conditions, helping to more accurately predict the remaining life of the battery. The capacity decay curve can clearly show the decay speed of the battery over time and provide quantitative decay data, providing decision-making support for battery replacement, maintenance, and charging strategies. Through this curve, the critical point of battery decay can be predicted in advance to avoid performance loss caused by excessive battery decay.,

[0012] Preferably, step S2 includes the following steps: Step S21: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage log to obtain the maximum endurance mileage at multiple time points; Step S22: Conduct successive adjacent endurance change analysis on the maximum endurance mileage at multiple time points to generate vehicle endurance change characteristics; Step S23: Perform periodic fluctuation evolution on the vehicle endurance change characteristics to generate vehicle endurance periodic fluctuation characteristics; Step S24: Perform time series trend evolution on the vehicle endurance periodic fluctuation characteristics to generate vehicle endurance evolution trend data; Step S25: Conduct in-depth feature mining on the vehicle endurance evolution trend data to construct a vehicle endurance evolution trend graph.

[0013] Through the calculation of the maximum cruising range at multiple time points for the historical vehicle cruising range log, the maximum cruising range of the vehicle at different time points can be accurately obtained, thereby helping to evaluate the maximum performance of the battery at different life cycle stages. This calculation can reveal the change of battery performance over time, especially when the battery ages or the usage frequency changes, and the change of the cruising ability. The calculation of the maximum cruising range at multiple time points provides a comparative analysis across time periods, enabling in-depth analysis of the law of the change of the vehicle battery health state over time. Comparing the cruising data at different time points helps to identify the trends and laws of battery performance decline. Through successive adjacent cruising change analysis, the change of each cruising can be captured, thereby more finely analyzing the decline trend of the battery. By identifying the change amplitude and speed of the cruising change, it is possible to more accurately judge whether the battery has experienced performance degradation or other potential problems. The change characteristics of the cruising are often closely related to the health state of the battery. For example, a sudden drop in the cruising range is an indication of serious decline or failure of the battery. Through this analysis, changes in the battery health state can be detected earlier for early warning. The battery cruising range is often affected by various factors such as temperature, driving habits, and road conditions. The cruising fluctuations caused by these factors will show periodic changes. By analyzing the periodic fluctuation evolution of the cruising change characteristics, these periodic fluctuations can be identified and quantified, providing more accurate time series data for subsequent health assessment. The periodic fluctuation characteristics reveal the repetitive decline pattern of the battery during long-term use, helping to predict the performance of the battery at a future time point. This provides a reliable basis for the battery management system to adjust the charging and discharging strategies in advance to avoid a sharp drop in the cruising range. Through time series trend evolution analysis, the cruising performance of the vehicle is comprehensively described by a time series model, thereby revealing the change trend of the cruising range over time. The analysis of the time series data helps to deeply understand the change of the battery health state in different time periods. The time series trend evolution can show the change pattern of the battery health state, revealing the potential law of the cruising range decline. Through this trend analysis, the future cruising ability of the battery can be predicted more accurately, providing a scientific basis for the adjustment of the battery management strategy. Deep feature mining extracts key features through multi-dimensional analysis of the vehicle cruising evolution trend data. These features help to more accurately understand the health status of the battery and capture the potential factors behind different cruising changes (such as battery aging, environmental temperature, load fluctuation, etc.).

[0014] Preferably, the specific steps of step S3 are as follows: Step S31: Perform a non-linear decay trend analysis on the vehicle cruising evolution trend graph to extract the non-linear decay trend of the cruising range; Step S32: Perform timestamp multi-point marking on the battery multi-modal state evolution characteristics and the non-linear decay trend of the cruising range to extract multiple timestamp nodes; Step S33: Perform parameter matching based on multiple timestamp nodes to obtain the parameter matching time points; Step S34: Use the parameter matching time points to perform time series cross - correlation mining on the multi - modal state evolution characteristics of the battery and the non - linear attenuation trend of the battery life, so as to obtain the battery state - battery life correlation characteristics; Step S35: Conduct an evolutionary analysis of the battery aging mechanism based on the battery state - battery life correlation characteristics, thereby constructing a knowledge graph of battery state aging evolution.

[0015] Through the multi-point maximum endurance calculation of the historical vehicle endurance mileage log, the maximum endurance mileage of the vehicle at different time points can be accurately obtained, thereby helping to evaluate the maximum performance of the battery at different life cycle stages. This calculation can reveal the change of battery performance over time, especially when the battery ages or the usage frequency changes, the change of the endurance ability. The multi-point maximum endurance calculation provides a comparative analysis across time periods, enabling in-depth analysis of the law of the change of the vehicle battery health state over time. Comparing the endurance data at different time points helps to identify the trends and laws of battery performance decline. Through successive adjacent endurance change analysis, the change of each endurance can be captured, so as to more finely analyze the decline trend of the battery. By identifying the change amplitude and speed of the endurance change, it is possible to more accurately judge whether the battery has performance degradation or other potential problems. The change characteristics of the endurance are often closely related to the health state of the battery. For example, a sudden drop in endurance is a sign of serious battery decline or failure. Through this analysis, the change of the battery health state can be detected earlier for early warning. The vehicle endurance is often affected by various factors, such as temperature, driving habits, road conditions, etc. The endurance fluctuations caused by these factors will show periodic changes. By analyzing the periodic fluctuation evolution of the endurance change characteristics, these periodic fluctuations are identified and quantified, providing more accurate time series data for subsequent health assessment. The periodic fluctuation characteristics reveal the repetitive decline mode of the battery during long-term use, helping to predict the performance of the battery at a certain future time point. This provides a reliable basis for the battery management system to adjust the charging and discharging strategies in advance to avoid a sharp drop in endurance. Through the time series trend evolution analysis, the endurance performance of the vehicle is comprehensively described by a time series model, thereby revealing the change trend of the endurance over time. The analysis of the time series data helps to deeply understand the change of the battery health state in different time periods. The time series trend evolution can show the change mode of the battery health state, revealing the potential law of the endurance decline. Through this trend analysis, the future endurance ability of the battery can be predicted more accurately, providing a scientific basis for the adjustment of the battery management strategy. Through in-depth feature mining, key features are extracted through multi-dimensional analysis of the vehicle endurance evolution trend data. These features help to more accurately understand the health status of the battery and capture the potential factors behind different endurance changes (such as battery aging, environmental temperature, load fluctuation, etc.). By constructing a vehicle endurance evolution trend diagram, the complex endurance change information is visualized, making the change trend of the battery health status clear at a glance. This provides an intuitive tool for battery managers to help them monitor the battery health status in real time and make corresponding decisions.

[0016] Preferably, the specific steps of step S4 are as follows: Step S41: Extract the historical vehicle driving environment parameters according to the historical vehicle endurance mileage log; Step S42: Calculate the environmental temperature change for the historical vehicle driving environment parameters to obtain the environmental temperature change curve; Step S43: Identify the driving scenarios for the historical vehicle driving environment parameters to obtain different driving scenario characteristics; Step S44: Analyze the impact of environmental changes on the vehicle endurance evolution trend graph based on the environmental temperature change curve and different driving scenario characteristics, so as to obtain the endurance trend - environmental change impact law; Step S45: Based on the endurance trend - environmental change impact law and the battery state aging evolution knowledge graph, perform dynamic vehicle state modeling on the vehicle battery multi-modal state monitoring parameters to construct a dynamic vehicle twin model.

[0017] The present invention provides data support for analyzing the impact of environmental factors on vehicle endurance by extracting historical vehicle driving environment parameters (such as temperature, humidity, road type, etc.). These parameters can reveal the performance of battery endurance in different environments, help identify the potential impact of different environmental conditions on battery performance. Different driving environments (such as cities, highways, mountains, etc.) have different impacts on the battery. By extracting these environmental parameters, accurately evaluate how environmental factors are related to the health status and endurance performance of the battery, and identify which external environments are the key factors affecting endurance changes. Environmental temperature is one of the key factors affecting battery performance. High temperature will accelerate battery degradation, while low temperature affects the charge and discharge efficiency of the battery. By calculating the environmental temperature change curve, understand the change law of battery endurance under different temperature conditions, and provide an important reference for subsequent health assessment. The temperature change curve provides a quantitative environmental impact model for vehicle endurance, enabling the comparison of the temperature change with the trend of endurance degradation, thus helping to more accurately evaluate the impact of different temperature environments on battery health. Driving behavior has a direct impact on battery consumption. By identifying different driving scenarios (such as urban driving conditions, highway driving conditions, mountain driving, etc.) in historical vehicle driving, and analyzing the performance of the battery under different working conditions according to the different characteristics of the scenarios. For example, the battery is more easily consumed in urban congestion, while high-speed driving makes the charge and discharge efficiency of the battery higher. Driving scenario recognition can provide more accurate upstream and downstream factors for endurance evolution. Each scenario corresponds to different battery loads and energy consumption patterns. Understanding these characteristics helps to more accurately predict the endurance ability of the battery under different driving conditions. By combining the environmental temperature change and driving scenario characteristics, deeply analyze the comprehensive impact of environmental factors on endurance. The change of environmental temperature and the difference in driving modes work together, resulting in different endurance abilities of the battery under the same remaining battery charge. This analysis helps to comprehensively understand the environmental factors behind the change of battery performance. Environmental changes include not only temperature, but also driving scenarios, terrain, climate and other factors. Through multi-dimensional impact analysis, more comprehensively grasp the impact law of environmental changes on battery endurance, and provide a more detailed and accurate analysis for future endurance prediction. By integrating the endurance trend and the impact law of environmental changes, plus the knowledge graph of battery state aging evolution, a dynamic vehicle twin model can be constructed. This model can reflect the state changes of the battery under various environmental conditions in real time, including the comprehensive impact of factors such as endurance, temperature, and driving behavior on battery health. The dynamic twin model provides a real-time monitoring framework for battery health status assessment. Through this model, the battery management system of the vehicle dynamically monitors the health status of the battery and predicts the future battery performance.

[0018] Preferably, the specific steps of step S5 are as follows: Step S51: Calculate the driving scenario frequency of the historical vehicle endurance mileage log to obtain different driving scenario frequency parameters; Step S52: Calculate the peak value of driving scenario frequency based on different driving scenario frequency parameters, and extract the normal driving scenarios; Step S53: Simulate the endurance evolution of the dynamic vehicle twin model for multiple periods based on the normal driving scenarios to extract the endurance evolution simulation data for multiple periods; Step S54: Analyze the endurance situation of the current scenario for the endurance evolution simulation data of multiple periods, so as to obtain the endurance situation characteristics of the current scenario; Step S55: Conduct dynamic endurance rolling prediction based on the endurance situation characteristics of the current scenario, so as to obtain the long-term endurance trend prediction data.

[0019] The present invention calculates the frequencies of different driving scenarios to deeply understand the distribution of common driving conditions during the actual use of the vehicle. This helps to identify the typical driving scenarios (such as urban driving, highway driving, mixed driving conditions, etc.) in which the vehicle is in most of the time. The calculated driving scenario frequency parameters provide data support for subsequent driving scenario feature analysis, making the modeling and analysis process more in line with the actual driving mode and avoiding errors caused by assuming unrealistic driving conditions. By calculating the frequency peaks of driving scenarios, the most frequently occurring driving scenarios are extracted to determine which scenarios dominate in daily use. These normalized driving scenarios have the most significant impact on the battery, so it is crucial to understand their characteristics. After extracting the normalized driving scenarios, targeted analysis is carried out on these scenarios to evaluate the performance of the battery in these scenarios, thereby adjusting the battery management strategy. For example, different battery charge and discharge schemes are formulated for the differences between urban driving conditions (low speed, high-frequency acceleration and deceleration) and highway driving (constant speed, high-power output) to improve the working efficiency of the battery and extend its service life. Through multiple-time-period endurance evolution simulations of the dynamic twin model based on the normalized driving scenarios, the endurance change trends of the vehicle in various scenarios at different time periods are obtained. These simulation data provide sequential and more refined inputs for endurance prediction. The simulation data of different time periods help to provide a clear map for the long-term and short-term endurance trends of the vehicle. By considering factors such as driving conditions and climate change, the change in the health state of the battery at different time periods is more accurately reflected. Through endurance trend analysis of the multiple-time-period endurance evolution simulation data, the endurance performance in the current driving scenario is dynamically understood. This analysis can reveal the immediate performance of the battery in different driving scenarios, such as whether the battery has endurance fluctuations due to excessive or insufficient load. The endurance trend characteristics provide accurate characteristic data for evaluating the endurance fluctuations, decay rate and health state in the current scenario. For example, by analyzing the influence of battery load, temperature and other environmental factors during driving, it is possible to identify in real time whether the battery is in a healthy state or at risk of decline. Through dynamic endurance rolling prediction based on the endurance trend characteristics of the current scenario, a long-term endurance trend prediction is provided for the vehicle. This prediction can not only reflect the future endurance performance of the battery but also anticipate in advance the endurance decline faced by the battery, facilitating timely maintenance and optimization measures. Through long-term endurance trend prediction, the battery management system realizes more intelligent resource scheduling and energy management. For example, when it is predicted that the battery endurance capacity will decline, the system takes strategies in advance (such as adjusting the charging cycle, reducing energy loss) to extend the effective service life of the battery.

[0020] Preferably, the specific steps of step S6 are as follows: Step S61: Calculate the endurance decay rate for the long-term endurance trend prediction data to obtain the endurance decay rate; Step S62: Based on the battery life decay rate, perform prediction fitting for the battery life status to obtain a battery life status prediction curve; Step S63: Quantitatively evaluate the battery comprehensive health status for the battery life status prediction data to generate a battery health status evaluation result.

[0021] Through calculating the decay rate for the long-term battery life trend prediction data, the present invention can clearly reveal the decay speed of the battery, which provides an important quantitative index for the battery health management, helps to evaluate whether the battery is in a normal usage state or has started to decay. By continuously tracking the decay rate and capturing the change trend of the battery health in real time, the calculated battery life decay rate is used for prediction fitting of the battery life status, and a prediction curve of the remaining battery life status is obtained. This curve provides an intuitive battery health warning for vehicle users, helping them understand the remaining available years or cycles of the battery. The battery life prediction curve can help evaluate the current aging state of the battery and predict the future health decay trend according to the battery performance in different usage cycles, which provides a scientific basis for the subsequent management and replacement of the battery. Through the quantitative evaluation of the battery life status prediction data, the current health status of the battery can be comprehensively and integrally reflected. The evaluation result not only considers factors such as decay rate and remaining life, but also combines external factors such as environmental changes, driving patterns, and temperature to give a comprehensive health score, providing an all-round view of the battery health status. The quantitative evaluation result can be presented in the form of charts, reports, etc., enabling users to intuitively understand the battery health status and identify potential risks or problem points. Such a clear evaluation report helps vehicle owners or managers make reasonable decisions, such as whether to replace the battery in advance or whether to adjust the battery management strategy.

[0022] In this specification, a battery health status evaluation system for new energy vehicles is provided, which is used to execute the above-mentioned battery health status evaluation method for new energy vehicles, and includes: A state evolution module, configured to obtain multi-modal state monitoring parameters of the vehicle battery and historical vehicle driving range logs; perform dynamic evolution of the battery state for the multi-modal state monitoring parameters of the vehicle battery to obtain multi-modal state evolution characteristics of the battery; A driving range evolution trend module, configured to perform multi-point maximum driving range calculation on the historical vehicle driving range logs and perform time-series trend evolution to construct a vehicle driving range evolution trend graph; A non-linear decay trend module, configured to perform non-linear decay trend analysis on the vehicle driving range evolution trend graph and perform battery aging mechanism evolution analysis, so as to construct a battery state aging evolution knowledge graph; A twin model module, configured to perform environmental change impact analysis on the vehicle driving range evolution trend graph according to the battery state aging evolution knowledge graph, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; The endurance trend prediction module is used to simulate the endurance evolution of the dynamic vehicle twin model for multiple periods and conduct dynamic endurance rolling prediction, so as to obtain long-term endurance trend prediction data; The health status assessment module is used to predict and fit the battery life status of the long-term endurance trend prediction data and conduct a quantitative assessment of the comprehensive health status of the battery to generate a battery health status assessment result.

[0023] Through the real-time collection of multi-modal state parameters, the present invention can comprehensively monitor the performance of the battery under different working conditions, provide data support for subsequent health assessment, analyze the dynamic evolution of the battery state, accurately identify the change trend of battery health, and provide an important basis for subsequent battery maintenance and management. Through the time series analysis of historical vehicle endurance mileage, comprehensively understand the performance of the battery in the past period of time, and help identify which factors have a greater impact on endurance. The endurance evolution trend chart can provide a scientific historical basis for subsequent endurance prediction, help vehicle owners understand the endurance change trend of the battery in advance. The non-linear decay trend analysis can help identify the complex patterns of battery degradation, so as to provide more accurate degradation prediction, which is crucial for accurately evaluating the battery health status. By constructing a knowledge graph of battery state aging evolution, it is possible to deeply understand the internal mechanism of battery aging and provide a theoretical basis for battery maintenance and optimization. By analyzing the impact of environmental changes (such as temperature, humidity, etc.) on endurance, it is possible to more accurately predict the performance of the battery under different driving conditions, which helps to maximize the battery performance. By simulating the dynamic performance of the battery under various environmental changes through the twin model, it is possible to provide real-time and accurate battery health status prediction for the battery management system, helping users better cope with the impact of complex environments on battery performance. Through the simulation of endurance evolution for multiple periods, long-term endurance prediction results are obtained, helping vehicle owners make travel plans in advance. The long-term endurance prediction data provides an important decision-making basis for the battery management system. The system optimizes the charging, discharging, and battery temperature control strategies according to the prediction results, improves the battery performance and extends the service life. The health status assessment module can comprehensively and accurately evaluate the current health status of the battery by combining multiple factors such as endurance trend prediction and battery decay rate. The health status assessment result provides a scientific basis for the battery management system. The system makes adjustments based on these assessment results, such as optimizing the charge and discharge cycles and adjusting the battery maintenance frequency, so as to effectively extend the battery service life and improve the overall performance. Description of the Drawings

[0024] Figure 1 It is a schematic flow chart of the steps of a method for evaluating the health status of a new energy vehicle battery according to the present invention; Figure 2 It is a schematic detailed implementation step flow chart of step S1; Figure 3It is a schematic diagram of the detailed implementation steps of step S2; Figure 4 It is a schematic diagram of the detailed implementation steps of step S3. Specific implementation manner

[0025] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0026] The embodiments of the present application provide a method and system for evaluating the health state of a new energy vehicle battery. The execution subjects of the method and system include, but are not limited to, the following general computing nodes that are equipped with the system: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. The data processing platform includes, but is not limited to, at least one of an audio and image management system, an information management system, and a cloud data management system.

[0027] Please refer to Figures 1 to 4 , the present invention provides a method for evaluating the health state of a new energy vehicle battery, and the method for evaluating the health state of a new energy vehicle battery includes the following steps: Step S1: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle endurance mileage log; perform dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain the multi-modal state evolution characteristics of the battery; Step S2: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage log, and perform time-series trend evolution to construct a vehicle endurance evolution trend graph; Step S3: Perform non-linear attenuation trend analysis on the vehicle endurance evolution trend graph, and perform battery aging mechanism evolution analysis, so as to construct a battery state aging evolution knowledge graph; Step S4: Perform environmental change impact analysis on the vehicle endurance evolution trend graph according to the battery state aging evolution knowledge graph, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; Step S5: Perform endurance evolution simulation for multiple periods on the dynamic vehicle twin model, and perform dynamic endurance rolling prediction to obtain long-term endurance trend prediction data; Step S6: Perform battery life state prediction fitting on the long-term endurance trend prediction data, and perform quantitative evaluation of the comprehensive health state of the battery to generate a battery health state evaluation result.

[0028] The present invention obtains multi-modal state monitoring parameters of a vehicle battery and historical vehicle endurance mileage logs, conducts dynamic evolution analysis on the battery state to reveal the evolution characteristics of the battery's multi-modal state. The acquisition of the battery's multi-modal state evolution characteristics helps to understand the state change law of the battery under different working conditions, providing a basis for subsequent health state assessment. Through multi-point maximum endurance calculation and time-series trend evolution analysis, a vehicle endurance evolution trend graph is constructed to reveal the change trend of the vehicle's endurance ability. The construction of the vehicle endurance evolution trend graph helps to understand the dynamic change of the vehicle's endurance performance, providing a reference for subsequent health state assessment. Through non-linear decay trend analysis and battery aging mechanism evolution analysis, a battery state aging evolution knowledge graph is constructed to deeply understand the battery aging mechanism and state evolution law. The construction of the battery state aging evolution knowledge graph helps to reveal the key factors and laws in the battery aging process, providing theoretical support for subsequent battery health state assessment. According to the battery state aging evolution knowledge graph, environmental change impact analysis and dynamic vehicle state modeling are carried out to construct a dynamic vehicle twin model to simulate the evolution of the vehicle state under different conditions. The construction of the dynamic vehicle twin model helps to simulate the state change of the vehicle in actual operation, providing more accurate model support for health state assessment. Through multiple-period endurance evolution simulation and dynamic endurance rolling prediction of the dynamic vehicle twin model, long-term endurance trend prediction data is obtained to predict the long-term change trend of the vehicle's endurance ability. The acquisition of the long-term endurance trend prediction data helps to formulate reasonable maintenance and management strategies to ensure the long-term stable performance of the vehicle. The long-term endurance trend prediction data is used for battery life state prediction fitting and battery comprehensive health state quantitative assessment to generate battery health state assessment results, accurately assessing the health state of the battery. The generation of the battery health state assessment results helps to detect battery problems in advance and take corresponding measures to extend the battery life and improve the vehicle performance.

[0029] In an embodiment of the present invention, refer to Figure 1 , which is a schematic diagram of the step flow of a method for evaluating the health state of a new energy vehicle battery according to the present invention. In this example, the steps of the method for evaluating the health state of the new energy vehicle battery include: Step S1: Obtain multi-modal state monitoring parameters of the vehicle battery and historical vehicle endurance mileage logs; conduct dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain battery multi-modal state evolution characteristics; In this embodiment, first, it is necessary to identify and confirm the data sources, including the battery multimodal status monitoring parameters of the vehicle and the historical vehicle endurance mileage logs. The battery multimodal status monitoring parameters usually include battery voltage, current, temperature, internal resistance, etc., while the endurance mileage logs record the driving time, mileage, and endurance status of the vehicle. Appropriate tools and techniques are used to extract the battery status parameters from the vehicle's monitoring system or data recording device, which involves using APIs, database queries, or data export functions. For example, if the vehicle is equipped with OBD-II (On-Board Diagnostic System), real-time battery monitoring data can be obtained through the OBD-II interface. The collected data is organized into a unified format to ensure the structuring and consistency of the data. This usually involves merging data from different sources into a single table and ensuring that each parameter has a clear timestamp for subsequent analysis. After the data is merged, data cleaning is performed to remove duplicate, missing, or outlier values. This step is crucial because the quality of the data directly affects the accuracy of subsequent analysis. Statistical methods (such as Z-score, IQR) are used to identify outliers and handle them. The handling methods include filling in missing values or deleting abnormal records. Ensure that the timestamp formats of all data are consistent and convert numerical data to the corresponding floating-point or integer formats, which involves parsing dates and times, as well as unit conversion of battery status parameters (such as converting the voltage unit from millivolts to volts). After data cleaning and formatting, data validation is performed to check whether the data is complete and meets expectations. Data visualization tools are used to generate preliminary charts to help identify potential problems in the data. Define the concept of the dynamic evolution characteristics of the battery status and clarify the characteristics that need to be analyzed, such as voltage, temperature change rate, internal resistance change, etc. These characteristics will help identify the change patterns of the battery during use. Select appropriate feature extraction methods. Commonly used methods include time series analysis, sliding window techniques, etc. The sliding window technique is used to calculate the changes in the battery status over different time periods. For example, the average voltage of the battery and its change rate are calculated using data from a past period of time. Data analysis tools (such as the pandas library in Python) are used to process the cleaned data, set the sliding window, calculate the battery status characteristics for each time period, and store the results in a new dataset. Calculate the following characteristics: the fluctuation range of the average voltage and current, the temperature change rate, and the change in internal resistance. Visualize the extracted battery multimodal status evolution characteristics to generate a time series chart showing the change in the battery status over time. This will help understand the health status and usage efficiency of the battery and provide a basis for subsequent analysis.

[0030] Step S2: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage logs, and conduct time-series trend evolution to construct a vehicle endurance evolution trend chart; In this embodiment, the data sources of the historical vehicle cruising range logs are confirmed. These data usually include information such as the cruising range after each charge, charging time, driving mileage, etc. Ensure that the data contains timestamps for subsequent analysis. Use appropriate methods (such as API calls, database queries, or manual exports) to extract the historical vehicle cruising range logs from the vehicle data management system, ensuring that the data is complete and the timestamps correspond accurately to the cruising range. Organize the collected data into a unified format, ensuring that each record contains a timestamp, cruising range, and other relevant information (such as vehicle status, environmental conditions, etc.). Use Excel or data analysis software (such as Python or R) for data organization. Select a suitable time series analysis method for analyzing the maximum cruising range change trend at multiple time points. Adopt methods such as linear regression and moving average to help identify the long-term trend of the cruising range. Use data analysis tools to perform time series analysis on the maximum cruising range data. The steps include: calculating the change rate at each time point to evaluate the growth or decline trend of the cruising range, calculating the moving average to reduce the volatility of the data and make the trend smoother. Use visualization tools (such as matplotlib or seaborn) to generate a graph of the vehicle cruising range evolution trend. The graph should show the relationship between the time series and the maximum cruising range, clearly identifying the cruising range data and trend line at each time point. Analyze the generated graph of the cruising range evolution trend to identify the patterns and rules of the cruising range change, and analyze the influencing factors, such as seasonal changes, driving habits, traffic conditions, etc. Organize the graph of the cruising range evolution trend and the relevant analysis results into a report for subsequent discussion and decision-making. Ensure that the report includes graphs, data analysis, and conclusions for easy understanding and reference.

[0031] Step S3: Conduct a non-linear decay trend analysis on the graph of the vehicle cruising range evolution trend and perform an analysis on the evolution mechanism of battery aging, so as to construct a knowledge graph of the battery state aging evolution. In this embodiment, to ensure that the data in the vehicle endurance evolution trend chart is complete and easy to process, the data should include time series and corresponding endurance mileage information. When necessary, clean the data by removing outliers and missing data to ensure the accuracy of the analysis. Select a suitable non-linear regression model to describe the endurance decay trend. Commonly used models include polynomial regression, exponential decay model, and logarithmic regression, etc. The selection of the model should be based on the characteristics of the data and evaluate its fitting effect. Use data analysis tools (such as the scikit-learn library in Python) to perform non-linear regression analysis. The specific steps are as follows: Divide the endurance data into independent variables (time) and dependent variables (endurance mileage), and perform model fitting. Use the curve_fit function or similar functions to estimate the model parameters, evaluate the fitting effect, judge the accuracy of the model through indicators such as the R² value, and ensure that the model can reasonably reflect the change trend of endurance. Visualize the fitting results and the original data to generate scatter plots and fitting curve graphs to intuitively display the non-linear decay trend of endurance. Analyze the characteristics of the fitting curve, identify the changes in the decay rate and influencing factors, and determine the mechanism of battery aging, including chemical reactions, physical changes, and external environmental factors, etc. Clarify how these mechanisms affect battery performance and endurance decay. Integrate multi-modal battery state monitoring parameters (such as internal resistance, temperature, voltage, etc.) with endurance decay data, extract features related to battery aging, use correlation analysis methods to identify the most critical influencing factors, apply statistical analysis methods (such as principal component analysis PCA or factor analysis) to reduce the dimension of the extracted features, identify the factors that have the greatest impact on battery aging, combine literature research and expert knowledge, analyze the relationships between different factors, construct a preliminary model of the battery aging mechanism, record the analysis results, generate an analysis chart of the evolution of the battery aging mechanism, display the influence of each factor on the evolution of the battery state, use flowcharts or causal diagrams to display the relationships between different factors, determine the structure of the knowledge graph, including nodes (such as battery state, external environment, driving habits, etc.) and edges (such as influence relationships, causal relationships, etc.), integrate the battery aging mechanism and endurance decay trend obtained from the previous analysis into the knowledge graph, use a graph database (such as Neo4j) or a knowledge graph construction tool (such as GraphDB) to implement, model the relationships between each node and edge, ensure that the knowledge graph can reflect the battery aging process in the real world, input different battery states and external factors, generate a dynamically updated knowledge graph, output the constructed knowledge graph of the evolution of the battery state aging, and perform visual display to ensure that the graph structure is clear and easy for users to understand and use.

[0032] Step S4: According to the knowledge graph of the evolution of the battery state aging, conduct an analysis of the impact of environmental changes on the vehicle endurance evolution trend chart, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; In this embodiment, data related to the vehicle's endurance evolution trend graph is collected, including environmental factors (such as temperature, humidity, air pressure) and vehicle states (such as battery health status, charging habits, etc.). Ensure that the data timestamps are consistent for accurate correlation analysis. Extract the environmental factors affecting endurance from the battery state aging evolution knowledge graph, and clarify how these factors are related to the endurance evolution trend. Use correlation analysis or regression analysis methods to quantify the impact degree of different environmental factors on endurance changes. Use data analysis tools (such as the statsmodels or pandas libraries in Python) for regression analysis and establish a model to describe the relationship between environmental factors and endurance. The specific steps include: setting the endurance evolution data as the dependent variable and the environmental factors as the independent variables for model fitting, evaluating the goodness of fit of the model, judging the significance of environmental factors through the R² value and P value, visualizing the analysis results, generating a relationship graph of environmental factors and endurance evolution trend, intuitively showing the changes in endurance under different environmental conditions, analyzing the rules shown in the graph, identifying the main influencing factors, providing a basis for subsequent dynamic modeling, determining the framework of dynamic vehicle state modeling, including input parameters (such as environmental factors, endurance status, battery health status) and output results (such as predicted endurance mileage, battery status, etc.), selecting a suitable dynamic modeling method. Commonly used ones are state space models, Kalman filters, machine learning models (such as LSTM, random forest, etc.). The selection should be based on the characteristics of the data and analysis requirements. Use programming tools (such as scikit-learn or TensorFlow in Python) to build a dynamic model. The specific steps include: dividing the dataset into a training set and a test set to ensure the generalization ability of the model, training the model, using historical data to predict future vehicle states and endurance performance, dynamically updating model parameters, validating the trained model, using the test set to evaluate the prediction accuracy of the model, and adjusting and optimizing according to the performance of the model to ensure that the model can maintain good prediction ability under different environmental conditions. Determine the composition of the dynamic vehicle twin model, including the combination of a physical model and a data model. The physical model describes the basic characteristics of the vehicle, while the data model is dynamically updated based on real-time monitoring data. Combine the results of dynamic modeling with the real-time monitoring data of the vehicle to form a complete dynamic twin model. This model can reflect the state changes and endurance performance of the vehicle in real time. Use a graphical modeling tool (such as MATLAB Simulink or Python libraries) to build a dynamic vehicle twin model. The specific steps include: designing the model structure to ensure that it can receive and process sensor data in real time, implementing a real-time update mechanism for the model, and dynamically adjusting the prediction results according to new data.

[0033] Step S5: Conduct endurance evolution simulations for multiple periods on the dynamic vehicle twin model and perform dynamic endurance rolling prediction to obtain long-term endurance trend prediction data; In this embodiment, determining the simulation targets for multiple time periods, including the environmental conditions, vehicle states, and battery parameters for each time period, setting the time span of the simulation, for example, conducting endurance evolution simulations by the hour, day, or week, in order to better capture the dynamic characteristics of endurance changes, collecting historical data, including environmental changes (temperature, humidity, etc.), vehicle operating states (driving modes, driving habits, etc.), and battery health conditions, ensuring that these data match the input parameters of the dynamic vehicle twin model, using programming tools (such as Python or MATLAB) to implement the endurance evolution simulation, the specific steps are as follows: Input the historical data into the dynamic vehicle twin model, run the model to predict the endurance performance at different time periods, according to the set time window, cyclically input the environmental and vehicle state data for each time period, record the endurance mileage at each time point, output the endurance evolution results for multiple time periods, generate a time series dataset, including the endurance mileage and corresponding state parameters at each time point, conduct a preliminary analysis to identify the trends and patterns of endurance changes, select a suitable dynamic prediction model, commonly used ones are time series analysis models (such as ARIMA, SARIMA) and machine learning models (such as LSTM, random forest, etc.), the selection of the model should be based on data characteristics and prediction requirements, extract features for prediction from the simulation results, including historical endurance mileage, environmental factors, driving behavior, etc., ensure the time series and continuity of the data for effective prediction, use data analysis tools (such as statsmodels or scikit - learn in Python) to conduct dynamic endurance rolling prediction, the specific steps include: Divide the data into a training set and a test set, use the training set to train the selected prediction model, define a rolling prediction mechanism, each time use the latest endurance data and state parameters to update the model, generate the prediction result for the next time point, verify the prediction result of the model, use the test set to evaluate the accuracy of the model, calculate the prediction error (such as MAE, RMSE) to quantify the model performance, and adjust and optimize the model parameters according to the results, integrate the results of the dynamic endurance rolling prediction to generate long - term endurance trend prediction data, these data should include the predicted endurance mileage and its uncertainty range at each time point, use visualization tools (such as matplotlib or seaborn) to generate long - term endurance trend prediction charts, display the comparison between the prediction results and historical data, and visually present the endurance trend changes, output the long - term endurance trend prediction data, generate a detailed report, including prediction results, analysis charts, and model evaluation, for facilitating relevant decision - making and subsequent applications.

[0034] Step S6: Perform a battery life state prediction fitting on the long - term endurance trend prediction data, and conduct a quantitative evaluation of the comprehensive battery health state to generate a battery health state evaluation result.

[0035] In this embodiment, long-term battery life trend prediction data is collected to ensure that the dataset contains information such as driving range, charging cycles, battery temperature, and usage frequency at each time point. These factors will affect the battery life status. Determine a suitable battery life status prediction model. Commonly used models include linear regression, exponential decay models, and machine learning models (such as support vector machines and random forests). The choice of model should be based on the distribution characteristics and fitting requirements of the data. Use data analysis tools (such as scikit-learn or statsmodels in Python) to fit the model. The specific steps are as follows: Use the driving range data as the dependent variable and relevant influencing factors (such as charging cycles, temperature, etc.) as independent variables for fitting. Train the model and evaluate the fitting effect. Use indicators such as R² value and root mean square error (RMSE) to judge the accuracy of the model. Visualize the fitting result with the original data to generate scatter plots and fitting curves to intuitively display the accuracy of battery life status prediction. Analyze the fitting curve to identify the changing pattern of battery life and determine the evaluation indicators for the comprehensive battery health status, including battery capacity, internal resistance, charge and discharge efficiency, etc. These indicators will help quantify the health status of the battery. Collect historical data related to the battery health status, including the charge and discharge records, internal resistance changes, temperature, and usage environment of the battery. Integrate these data to form a basic dataset for battery health assessment. Use the weighted scoring method or fuzzy logic model to quantitatively evaluate the battery health status. The specific steps include: Set weights for each health status indicator based on expert opinions or data analysis results. Calculate the score for each indicator and calculate the comprehensive health score through weighted averaging. Output the battery health status assessment results, including the scores of each indicator and the comprehensive score, and generate a visualization chart to show the changes in the battery health status to ensure that the results are intuitive and easy to understand for users. Analyze the generated battery health status assessment results to identify the main factors affecting battery health, such as charging habits and environmental conditions, to provide a basis for subsequent management strategies. Organize the battery health status assessment results into a report, including analysis charts, model evaluations, and conclusions, for relevant personnel to refer to and make decisions.

[0036] In this embodiment, refer to Figure 2 , which is a schematic diagram of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle driving range log; Step S12: Filter abnormal parameters from the multi-modal state monitoring parameters of the vehicle battery to obtain filtered and optimized battery state monitoring parameters; Step S13: Analyze the time-series voltage fluctuations of the filtered and optimized battery state monitoring parameters to generate the voltage fluctuation characteristics of the battery during the journey; Step S14: Perform discrete fitting of the battery capacity attenuation for the filtered and optimized battery state monitoring parameters, so as to construct a battery capacity attenuation curve; Step S15: Perform dynamic evolution of the battery state on the battery voltage fluctuation characteristics during the journey state and the battery capacity attenuation curve, in order to obtain the multi-modal state evolution characteristics of the battery.

[0037] In this embodiment, the battery state parameters to be monitored are clarified, including voltage, current, temperature, SOC (State of Charge), SOH (State of Health), etc. In addition, the format and acquisition time interval of the driving range log are determined, and appropriate vehicle battery monitoring devices are equipped, such as intelligent battery management systems (BMS) and data recorders. These devices can record the multi-modal state parameters of the battery in real time. During vehicle driving, the battery state monitoring parameters and driving range log are collected in real time, ensuring the precise setting of the data recording device to guarantee the accuracy and integrity of the data. The collected data is stored in a database for subsequent analysis, ensuring a unified data format and marking time stamps for subsequent time series analysis. According to the normal operating range of the battery, normal thresholds for each monitoring parameter are set. For example, the voltage should be within a certain range, the temperature should be below a certain critical value, and SOC and SOH should also meet the predetermined standards. The collected battery state monitoring parameters are initially cleaned to remove duplicate and missing data records. Statistical methods (such as Z-score or IQR methods) are used to identify abnormal parameters, calculate the mean and standard deviation of each parameter, and determine whether it is an outlier according to the set threshold. The identified abnormal parameters are excluded to form filtered and optimized battery state monitoring parameters, ensuring that the filtered data set contains only valid and reliable monitoring parameters. Appropriate data analysis tools and libraries (such as Pandas, NumPy, and Matplotlib in Python) are selected for time series data analysis. Voltage data is extracted from the filtered and optimized battery state monitoring parameters and sorted by time to form a time series data set. The time series fluctuation characteristics of the voltage are calculated, including statistical indicators such as mean, variance, maximum value, and minimum value. In addition, the change rate and fluctuation amplitude of the voltage are calculated to facilitate the analysis of voltage fluctuation characteristics during the journey. The voltage fluctuation situation is visualized by plotting a time series graph to show the change trend of voltage over time, helping to identify fluctuation characteristics and abnormal situations. Historical data related to battery capacity, especially information on SOC and charge / discharge cycles, is extracted from the monitoring parameters. An appropriate fitting model (such as linear regression, polynomial fitting, or exponential decay model) is selected to describe the decay trend of battery capacity. Statistical methods such as the least squares method are used to fit the collected battery capacity data to obtain a mathematical model of capacity decay, and the fitting parameters are recorded for subsequent analysis. According to the fitting results, a battery capacity decay curve is generated and visualized to intuitively show the change of battery capacity over time or charge / discharge cycles. An appropriate dynamic model (such as a state space model or Kalman filter) is selected to describe the evolution process of the battery state. Combining the voltage fluctuation characteristics and the capacity decay curve, the change of the battery state over time is analyzed, the performance and health of the battery in different states are evaluated, and potential decay trends are identified. The multi-modal state evolution characteristics of the battery are recorded, a report is generated, and the change of the battery state at different time points is shown through charts.Help to understand the health status of the battery and its dynamic evolution characteristics.

[0038] In this embodiment, the specific steps of step S14 are as follows: Collect charge-discharge battery state for the filtered and optimized battery state monitoring parameters to obtain a plurality of charge-discharge battery state parameters; Analyze the change of battery internal resistance for the plurality of charge-discharge battery state parameters to generate battery internal resistance change characteristics; Evaluate the real-time charge transfer performance based on the battery internal resistance change characteristics to obtain a plurality of real-time charge transfer performance evaluation values; Perform polarization voltage evolution on the plurality of charge-discharge battery state parameters to obtain polarization voltage time-series evolution parameters; Calculate the change of battery capacity in a sliding time window for the polarization voltage time-series evolution parameters and the plurality of real-time charge transfer performance evaluation values to generate battery capacity change data for multiple time windows; Analyze the battery capacity attenuation based on the battery capacity change data for multiple time windows to obtain battery capacity attenuation parameters for multiple time windows; Perform discrete time-series fitting on the battery capacity attenuation parameters for multiple time windows to construct a battery capacity attenuation curve.

[0039] In this embodiment, determine the battery state parameters to be monitored, including voltage, current, temperature, internal resistance, etc., ensure that these parameters can comprehensively reflect the charge-discharge state of the battery, select appropriate monitoring devices (such as data collectors, sensors) for real-time monitoring of the battery state, and the devices should have high precision and real-time data transmission capabilities to ensure the accuracy and timeliness of the data. During the battery charge-discharge cycle, regularly collect the above-defined state parameters, use an automated system to record the data, ensure the stability and continuity of the collection process, and after each charge-discharge cycle, organize and save the data for subsequent analysis. Sort and clean the collected data to remove outliers and noise. Through data preprocessing, ensure the data quality for subsequent analysis. Determine the calculation method of battery internal resistance, usually calculated by Ohm's law (R = V / I), where V is voltage and I is current. Combine the voltage and current data during charge and discharge processes to calculate the internal resistance value at each moment. Process the collected charge and discharge state parameters, calculate the internal resistance value at each collection point, organize the calculation results into an internal resistance change data set, and record the internal resistance change at each time point. Use statistical analysis methods (such as mean, standard deviation) and visualization techniques (such as line charts) to analyze the change characteristics of battery internal resistance, identify the trends and patterns of internal resistance changes, and help understand the battery aging process. Determine the evaluation indicators of battery charge transfer performance, such as charge efficiency, charge time, discharge time, etc. These indicators can reflect the performance of the battery under different states. Based on the internal resistance change characteristics and real-time collected current and voltage data, calculate the evaluation value of battery charge transfer performance, and use the formula for calculation: charge efficiency = (discharge capacity / charge capacity) × 100%. Organize the calculated multiple real-time charge transfer performance evaluation values into a data set for visualization. Through chart analysis, evaluate the transfer ability of the battery under different working conditions. The polarization voltage refers to the voltage drop generated inside the battery due to the passage of current during battery charge and discharge. It is one of the important indicators of battery performance. Extract the polarization voltage data from the collected battery state parameters. The calculation method is: during charge or discharge, monitor the voltage drop and record it. Analyze the temporal variation of the polarization voltage to generate the temporal evolution parameters of the polarization voltage. Use a time series diagram to show the change of the polarization voltage of the battery during charge and discharge, and identify the change pattern of battery performance. Set the size of the sliding time window, usually several charge and discharge cycles. Through the sliding time window, observe the change of battery capacity in different time periods. Based on the temporal evolution parameters of the polarization voltage and the real-time charge transfer performance evaluation value, calculate the change of battery capacity within each sliding time window, ΔC = C(t) - C(t - Δt), where C(t) is the current capacity and C(t - Δt) is the capacity of the previous time window. Organize the calculated battery capacity change data of multiple time windows into a data set for subsequent analysis and visualization. Determine the relevant parameters of battery capacity attenuation, usually including attenuation rate, remaining capacity, etc. These parameters can reflect the health status and service life of the battery. By analyzing the battery capacity change data of multiple time windows, calculate the capacity attenuation parameters for each time window, attenuation rate = {C(t) - C(t + Δt)} / C(t) × 100%. Record the capacity attenuation parameters for each time window and conduct statistical analysis. Use charts to show the attenuation trend of battery capacity and help understand the battery aging process. Select a suitable fitting method (such as polynomial fitting, exponential fitting) to perform discrete time series fitting on the capacity attenuation parameters, and determine the order of the fitting model to ensure the fitting effect.Use a fitting algorithm (such as the least squares method) to fit the capacity decay parameters to generate a battery capacity decay curve. Evaluate the fitting effect by calculating the goodness of fit (such as the R² value). Visualize the fitted battery capacity decay curve to facilitate the analysis of the decay characteristics and trends of the battery during use, which will provide an important basis for battery management and optimization.

[0040] In this embodiment, refer to Figure 3 , which is a schematic diagram of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage log to obtain the maximum endurance mileage at multiple time points; Step S22: Perform successive adjacent endurance change analysis on the maximum endurance mileage at multiple time points to generate the vehicle endurance change characteristics; Step S23: Perform periodic fluctuation evolution on the vehicle endurance change characteristics to generate the vehicle endurance periodic fluctuation characteristics; Step S24: Perform time-series trend evolution on the vehicle endurance periodic fluctuation characteristics to generate the vehicle endurance evolution trend data; Step S25: Perform in-depth feature mining on the vehicle endurance evolution trend data to construct a vehicle endurance evolution trend graph.

[0041] In this embodiment, historical vehicle driving range log data is collected. The data should include the driving range after each charge, charging time, vehicle usage, etc. The data is cleaned to remove duplicates, missing values, or outliers to ensure data accuracy. A method for calculating the maximum driving range at each time point is defined. The maximum driving range within a certain time window (such as daily or weekly) is selected. Using data aggregation methods, the driving range is grouped by time and the maximum value is calculated. A programming tool (such as Python or R) is used to read the cleaned data, and the data is grouped using an aggregation function (such as groupby). The maximum driving range within each time window is calculated, and the result is stored in a new data set, generating a data set of maximum driving ranges at multiple time points, ensuring that the data format is clear for subsequent analysis and visualization. The calculation method for driving range changes is determined. Usually, the driving range at the current time point is subtracted from the driving range at the previous time point to calculate the absolute and percentage changes in driving range. Using the maximum driving range data generated in step S21, ensure that the data is arranged in chronological order for successive adjacent analysis. In a programming environment, loop or vectorized operations are used to calculate the successive adjacent changes in the driving range data, generating new data columns to record the absolute change value and relative change percentage at each time point. The calculated driving range change characteristics are organized into a new data set, including time points, maximum driving ranges, and their change characteristics. Through a data visualization tool (such as Matplotlib or Seaborn), a change trend graph is drawn to visually display the fluctuations in vehicle driving range. Determine the definition of periodic fluctuations, which usually refer to the repeated patterns or periodic changes in the driving range change characteristics. Signal processing methods such as Fourier transform are used to identify periodic characteristics. Using the driving range change characteristic data generated in step S22, ensure that the data format is suitable for periodic analysis. Usually, smoothing processing is required to eliminate noise. Methods such as Fourier transform or wavelet transform are applied to perform frequency domain analysis on the driving range change characteristics to identify the main periodic components, and analyze their corresponding frequencies and amplitudes. The periodic fluctuation characteristics are organized into a data set, including information such as the main period and amplitude, generating a visualization chart to display the periodic fluctuation characteristics of vehicle driving range for subsequent analysis. Determine the definition of the time series trend, which usually refers to the change trend of the driving range fluctuation characteristics over time. Methods such as linear regression and moving average are used to identify the overall trend. Using the periodic fluctuation characteristic data generated in step S23, a suitable time window (such as daily or weekly) is selected for trend analysis. A linear regression model or moving average method is applied to perform time series analysis on the periodic fluctuation characteristics, generating a trend line, and recording parameters such as the slope and intercept of the trend line. Evaluate the rise or fall of the overall trend. The obtained time series trend data is organized into a data set, generating a trend graph to display the trend characteristics of vehicle driving range over time for intuitive understanding of the changes in driving range. Determine the in-depth characteristics to be mined, including the fluctuation amplitude of the driving range, the amplitude of trend changes, the cycle length, etc.These features will help to comprehensively understand the evolution law of vehicle endurance, select suitable data mining methods such as principal component analysis (PCA), clustering analysis or machine learning models (such as decision trees, random forests) for in-depth feature mining. In the programming environment, use the selected method to analyze the time-series trend data, extract potential in-depth features, record the importance of the features and their impact on endurance changes, organize the mined in-depth features into a dataset, generate a feature impact chart to show the contribution of each feature to the evolution of vehicle endurance, and facilitate subsequent decision-making support.

[0042] In this embodiment, refer to Figure 4 , which is a schematic diagram of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: Conduct a non-linear decay trend analysis on the vehicle endurance evolution trend graph to extract the non-linear decay trend of endurance. Step S32: Perform timestamp multi-point marking on the battery multi-modal state evolution characteristics and the non-linear decay trend of endurance to extract multiple timestamp nodes. Step S33: Perform parameter matching based on multiple timestamp nodes to obtain the parameter matching time points. Step S34: Use the parameter matching time points to conduct time-series cross-correlation mining on the battery multi-modal state evolution characteristics and the non-linear decay trend of endurance to obtain the battery state-endurance correlation features. Step S35: Conduct an analysis of the evolution mechanism of battery aging based on the battery state-endurance correlation features to construct a knowledge graph of battery state aging evolution.

[0043] In this embodiment, obtaining the data for the vehicle's endurance evolution trend graph, including the variation of the endurance mileage over time, ensuring the integrity of the data, removing outliers and missing data to improve the accuracy of the analysis, selecting a suitable non-linear regression model (such as polynomial regression, exponential regression) to analyze the attenuation trend of the endurance. The non-linear model can better capture the complexity of the endurance mileage changing over time, especially in the case of battery aging. Using data analysis tools (such as scikit-learn in Python or the nls function in R) to perform non-linear regression analysis, taking the endurance mileage as the dependent variable and time as the independent variable, fitting the non-linear model, evaluating the goodness of fit of the model (for example, the R² value) to ensure the effectiveness of the model, extracting the non-linear attenuation trend from the fitting results, and generating the corresponding visualization chart to intuitively display the non-linear attenuation characteristics of the endurance mileage. Collecting the data of the multi-modal state evolution characteristics of the battery (such as voltage, current, temperature, etc.) and the data of the non-linear attenuation trend of the endurance, ensuring that the timestamp formats of the two sets of data are consistent for subsequent analysis. Determining the timestamp nodes that need to be marked, usually selecting time points with important significance, such as charging cycles, mileage change points, or important events (such as accidents, repairs, etc.). Using programming tools (such as Python) to traverse the data, marking each timestamp node, and recording the corresponding values of the battery state and the endurance trend at each node to form a comprehensive data set. Outputting the data set containing multiple timestamp nodes, ensuring the accuracy and integrity of the data for subsequent analysis. Clearly defining the parameters that need to be matched, including the multi-modal state characteristics of the battery (such as internal resistance, temperature) and the data related to the endurance mileage, ensuring that these parameters can reflect the relationship between battery performance and endurance. Selecting a suitable parameter matching algorithm, usually using time series matching methods (such as DTW, Dynamic Time Warping) to compare the parameters at different timestamp nodes. Using programming tools to implement the parameter matching algorithm, traversing all timestamp nodes, performing parameter matching, and recording the parameter matching results at each timestamp node to form a new data set. Outputting the data set of the parameter matching time points, ensuring the accuracy of the data, and conducting preliminary analysis to verify the rationality of the matching results. Clearly defining the content of the cross-correlation analysis to be carried out, that is, exploring the relationship between the battery state and the endurance mileage, identifying potential influencing factors and patterns. Selecting a suitable time series analysis method, such as Cross-Correlation or Granger Causality analysis, to evaluate the impact of the battery state on the endurance mileage. Using programming tools to perform cross-correlation mining, inputting the data of the parameter matching time points, performing cross-correlation analysis, identifying and evaluating the relationship between the battery state and the endurance, recording and visualizing the analysis results, and generating a battery state-endurance correlation feature graph to intuitively display the relationship between the battery state change and the endurance mileage. Determining the mechanism of battery aging, including chemical reactions, physical changes, etc., and clarifying how these mechanisms affect the battery performance and endurance mileage. Selecting a suitable method for constructing a knowledge graph,Use a graph database (such as Neo4j) or a knowledge graph construction tool (such as GraphDB) to visualize the association between battery status and endurance. Utilize the collected battery status-endurance association features to construct a knowledge graph of the evolution mechanism of battery aging. The graph should display the relationships among battery status, endurance changes, and aging mechanisms. Output the constructed knowledge graph, ensuring its clear structure for facilitating the understanding and analysis of the battery aging process and its impact on endurance. This knowledge graph will provide an important basis for subsequent battery management and optimization.

[0044] In this embodiment, step S4 includes the following steps: Step S41: Extract historical vehicle driving environment parameters according to the historical vehicle endurance mileage log; Step S42: Calculate the environmental temperature change for the historical vehicle driving environment parameters to obtain an environmental temperature change curve; Step S43: Identify different driving scene features for the historical vehicle driving environment parameters; Step S44: Conduct an analysis of the impact of environmental changes on the vehicle endurance evolution trend graph based on the environmental temperature change curve and different driving scene features, thereby obtaining the endurance trend-environmental change impact law; Step S45: Based on the endurance trend-environmental change impact law and the knowledge graph of battery state aging evolution, perform dynamic vehicle state modeling on the multi-modal state monitoring parameters of the vehicle battery, and construct a dynamic vehicle twin model.

[0045] In this embodiment, historical vehicle driving range logs and related environmental parameter data are collected. These parameters include environmental temperature, humidity, air pressure, wind speed, etc. These factors can affect the vehicle's driving range. Data related to the driving environment is extracted from the driving range logs. Through data fusion technology, data from different sources (such as meteorological data, sensor data, etc.) are integrated into a unified dataset. The extracted environmental parameters are cleaned to remove outliers and missing data to ensure the accuracy and consistency of the data. The cleaned data is organized for subsequent analysis, generating a dataset containing historical vehicle driving environment parameters, ensuring that the data format is clear for subsequent processing and analysis. Determine the method for calculating the change in environmental temperature. Use a simple difference calculation method to analyze the temperature changes at different time points. Use programming tools (such as Python) to read the historical environmental parameter data and calculate the temperature change within each time period. Calculate the daily average, weekly average, or average of other time periods of the temperature and record the changes. Visualize the calculated temperature change data to generate an environmental temperature change curve. Use a line chart to show the trend of temperature change over time to help understand the impact of the environment on the vehicle's driving range. Output the generated environmental temperature change curve and conduct a preliminary analysis to identify the potential impact of temperature changes on the vehicle's driving range. Determine the types of driving scenarios, such as urban driving, highway driving, mountain driving, etc. Clearly define the characteristics of each scenario and its impact on the driving range. Extract features related to the driving scenario from the historical vehicle driving environment parameters, such as driving speed, acceleration, braking frequency, etc. These features will be used to identify different driving scenarios. Select appropriate machine learning algorithms (such as K-means clustering, support vector machines, etc.) for driving scenario identification. Train the model to identify different driving scenarios. Apply the trained model to classify the historical environmental parameter data and mark the characteristics of different driving scenarios. Record the relevant parameters of each driving scenario for subsequent analysis. Output the driving scenario identification results and verify the accuracy of the identification to ensure the rationality of the results. Determine the analysis framework, including the impact of environmental temperature changes and driving scenarios on the evolution of the driving range. Clearly define the analysis objective, that is, to identify which environmental factors have the greatest impact on the driving range. Select a suitable statistical analysis model (such as linear regression, analysis of variance, etc.) for environmental change impact analysis. Set the independent variables (environmental temperature, driving scenario) and the dependent variable (driving range). Use programming tools (such as the statsmodels library in Python) for data analysis. Combine the environmental temperature change curve and driving scenario characteristics with the vehicle driving range evolution trend data for regression analysis. Record and visualize the analysis results to generate a chart of the driving range trend - environmental change impact law, intuitively showing the impact of environmental changes on the driving range. Determine the framework for dynamic vehicle state modeling, including input parameters (driving range trend, environmental changes, battery state) and output results (vehicle state assessment).Combine the battery state aging evolution knowledge graph with the law of the impact of endurance trend - environmental change to form a complete power system model, help understand the relationship between battery state and endurance, select appropriate modeling methods (such as state space model, dynamic system modeling, etc.), perform dynamic modeling based on the collected data, implement the dynamic model using programming tools, update the vehicle state in real time according to the input parameters, the model should have self - adaptive ability and be able to adjust according to new data, output the constructed dynamic vehicle twin model, and verify the model to ensure its accuracy and effectiveness, generate a visual display of the model to help understand the changes and influencing factors of the vehicle state.

[0046] In this embodiment, step S5 includes the following steps: Step S51: Calculate the driving scenario frequency of the historical vehicle endurance mileage log to obtain different driving scenario frequency parameters; Step S52: Calculate the driving scenario frequency peak based on different driving scenario frequency parameters and extract the normalized driving scenario; Step S53: Perform multi - period endurance evolution simulation on the dynamic vehicle twin model based on the normalized driving scenario to extract multi - period endurance evolution simulation data; Step S54: Analyze the current scenario endurance situation of the multi - period endurance evolution simulation data to obtain the current scenario endurance situation characteristics; Step S55: Perform dynamic endurance rolling prediction based on the current scenario endurance situation characteristics to obtain long - term endurance trend prediction data.

[0047] In this embodiment, historical vehicle endurance mileage log data is collected to ensure that it includes information such as the time of each trip, driving scenario, and driving mileage. These data will be used to analyze the frequency of different driving scenarios. The driving scenarios are classified into different types (such as urban driving, highway driving, mountain driving, etc.), and classification criteria are set for each scenario. Programming tools (such as Python) are used to read the data and count the occurrence frequency of each scenario within a specific time period. Using data analysis tools, calculate the number of occurrences of each driving scenario in the entire dataset to generate frequency parameters. The value_counts() method of the pandas library is used to achieve fast statistics. Output a dataset containing the frequency parameters of different driving scenarios, ensuring that the data is clear and easy to read, and conduct a preliminary analysis to identify which driving scenarios are the most common. Determine the calculation method for the driving scenario frequency peak. Usually, select the several scenarios with the highest frequency values as the normalized driving scenarios. Use programming tools to sort the scenario frequency data generated in the previous step, extract the frequency peaks, set a threshold value (such as the top 20% of the scenarios) to ensure the representativeness of the selected normalized scenarios. Record the normalized driving scenarios and generate a visualization chart to show the frequency distribution and its peaks of each scenario, facilitating the understanding of normalized driving behaviors. Output a list of normalized driving scenarios for subsequent analysis and simulation. Determine the simulation time period and parameters, including the driving scenario, environmental conditions, and battery status for each time period. Using the dynamic vehicle twin model, conduct endurance evolution simulations for multiple time periods based on the normalized driving scenarios. Input the parameters of each scenario, run the model, and record the simulation results. Collect the endurance evolution simulation data from different time periods and organize it into a unified format for subsequent analysis. Output the endurance evolution simulation data for multiple time periods, ensuring that the data is clear and facilitating visualization and subsequent analysis. Determine the objectives of the current scenario endurance situation analysis, including identifying endurance trends, potential risks, and influencing factors. Select suitable analysis methods, such as time series analysis, regression analysis, etc. Analyze the relationship between the endurance evolution simulation data and the current environmental factors. Use data analysis tools to analyze the simulation data and evaluate the endurance situation characteristics of different time periods. Use a visualization tool (such as matplotlib) to generate a trend chart to intuitively show the endurance changes. According to the current scenario endurance situation characteristics, select a suitable prediction model, such as a time series prediction model (ARIMA, LSTM, etc.), and define the input parameters. Divide the historical data into a training set and a test set. Use the training set to train the prediction model and verify the accuracy of the model through the test set. Use the trained model to conduct dynamic endurance rolling predictions. Input the latest endurance situation characteristics to generate long-term endurance trend prediction data. Output the long-term endurance trend prediction data and generate a visualization chart to help understand the future endurance change trend.

[0048] In this embodiment, step S6 includes the following steps: Step S61: Calculate the battery life attenuation rate for the long-term battery life trend prediction data to obtain the battery life attenuation rate; Step S62: Perform a prediction fitting of the battery life status based on the battery life attenuation rate to obtain the battery life status prediction curve; Step S63: Quantitatively evaluate the overall battery health status for the battery life status prediction data to generate the battery health status evaluation result.

[0049] In this embodiment, collect the long-term battery life trend prediction data to ensure the integrity and accuracy of the data. Such data should include the battery life at each time point and the corresponding timestamp. Determine the calculation formula for the battery life attenuation rate. A common method is to calculate the change rate of the battery life within each time period. The attenuation rate = {C(t0) - C(t1)} / (t1 - t0), where C(t0) and C(t1) are the battery life at the initial and end time points respectively. Use programming tools (such as Python or R) to read the long-term battery life trend prediction data, traverse the data set, calculate the attenuation rate for each time period, store the results in a new data set, output the data set containing the battery life attenuation rate, and generate a visualization chart to show the change trend of the battery life attenuation rate over time. This helps to identify the patterns and influencing factors of the battery life attenuation. Determine the prediction targets for the battery life status, including the remaining battery life, attenuation time, etc. The battery life status prediction aims to evaluate the actual remaining usage time and performance of the battery. Select a suitable fitting model (such as the exponential decay model, linear regression model, etc.) to describe the relationship between the battery life attenuation rate and the battery life status. Use the historical data and the calculated battery life attenuation rate, and apply the selected fitting model for curve fitting. Use data analysis tools (such as scikit-learn or statsmodels) for model training and validation, output the battery life status prediction curve, and generate a visualization chart to show the relationship between the battery life and the battery life attenuation rate. This provides an intuitive reference for subsequent battery management. Determine the evaluation indicators for the overall battery health status, including the battery capacity, internal resistance, charge and discharge efficiency, etc. Each indicator will affect the overall health status of the battery. Select a suitable quantitative evaluation model (such as the weighted scoring method, fuzzy logic system, etc.) to comprehensively evaluate the battery health status. The model should be able to convert multiple evaluation indicators into a comprehensive health score. According to the battery life status prediction data and the health status evaluation indicators, calculate the score for each indicator and sum them up to obtain the comprehensive health status score. Use programming tools to implement this calculation process, output the battery health status evaluation result, including the scores of each indicator and the comprehensive score, and generate a visualization chart to intuitively show the battery health status. These results will provide an important basis for battery maintenance and replacement.

[0050] In this embodiment, a new energy vehicle battery health state evaluation system is provided, which is used to execute the new energy vehicle battery health state evaluation method as described above, and includes: A state evolution module, configured to obtain multi-modal state monitoring parameters of the vehicle battery and historical vehicle cruising range logs; perform dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain multi-modal state evolution characteristics of the battery; A cruising range evolution trend module, configured to calculate the maximum cruising range at multiple time points for the historical vehicle cruising range logs, and perform time series trend evolution to construct a vehicle cruising range evolution trend graph; A non-linear attenuation trend module, configured to perform non-linear attenuation trend analysis on the vehicle cruising range evolution trend graph, and perform battery aging mechanism evolution analysis, so as to construct a battery state aging evolution knowledge graph; A twin model module, configured to perform environmental change impact analysis on the vehicle cruising range evolution trend graph according to the battery state aging evolution knowledge graph, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; A cruising range trend prediction module, configured to perform cruising range evolution simulation for multiple time periods on the dynamic vehicle twin model, and perform dynamic cruising range rolling prediction, so as to obtain long-term cruising range trend prediction data; A health state evaluation module, configured to perform battery life state prediction fitting on the long-term cruising range trend prediction data, and perform quantitative evaluation of the comprehensive health state of the battery to generate a battery health state evaluation result.

[0051] Through the real-time collection of multi-modal state parameters, the present invention can comprehensively monitor the performance of the battery under different working conditions, providing data support for subsequent health assessment, analyzing the dynamic evolution of the battery state, accurately identifying the change trend of battery health, and providing an important basis for subsequent battery maintenance and management. Through the time series analysis of historical vehicle cruising ranges, comprehensively understand the performance of the battery over a period of time in the past, help identify which factors have a greater impact on cruising range, and the cruising range evolution trend chart can provide a scientific historical basis for subsequent cruising range prediction, helping vehicle owners understand the cruising range change trend of the battery in advance. The non-linear decay trend analysis can help identify the complex patterns of battery degradation, thus providing more accurate degradation prediction, which is crucial for accurately evaluating the battery health status. By constructing a knowledge graph of battery state aging evolution, the internal mechanism of battery aging can be deeply understood, providing a theoretical basis for battery maintenance and optimization. By analyzing the impact of environmental changes (such as temperature, humidity, etc.) on cruising range, the performance of the battery under different driving conditions can be predicted more accurately, which helps to maximize battery performance. By using a twin model to simulate the dynamic performance of the battery under various environmental changes, real-time and accurate prediction of the battery health status can be provided for the battery management system, helping users better cope with the impact of complex environments on battery performance. Through the simulation of the cruising range evolution in multiple periods, long-term cruising range prediction results can be obtained, helping vehicle owners make travel plans in advance. The long-term cruising range prediction data provides an important decision-making basis for the battery management system, and the system optimizes the charging, discharging, and battery temperature control strategies according to the prediction results to improve battery performance and extend service life. The health status assessment module can comprehensively and accurately evaluate the current health status of the battery by combining multiple factors such as cruising range trend prediction and battery decay rate. The health status assessment results provide a scientific basis for the battery management system, and the system makes adjustments based on these assessment results, such as optimizing the charge and discharge cycles and adjusting the battery maintenance frequency, thereby effectively extending the battery service life and improving the overall performance.

[0052] Therefore, in any aspect, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed by the present invention.

[0053] As described above are only the specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the state of health of a new energy vehicle battery, characterized in that, Including the following steps: Step S1: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle endurance mileage log; perform dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain the multi-modal state evolution characteristics of the battery; Step S2: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage log, and perform time-series trend evolution to construct a vehicle endurance evolution trend graph; Step S3: Perform non-linear attenuation trend analysis on the vehicle endurance evolution trend graph, and perform battery aging mechanism evolution analysis, so as to construct a battery state aging evolution knowledge graph; Step S4: According to the battery state aging evolution knowledge graph, perform environmental change impact analysis on the vehicle endurance evolution trend graph, conduct dynamic vehicle state modeling, and construct a dynamic vehicle twin model; Step S5: Perform endurance evolution simulation for multiple time periods on the dynamic vehicle twin model, and perform dynamic endurance rolling prediction to obtain long-term endurance trend prediction data; Step S6: Perform battery life state prediction fitting on the long-term endurance trend prediction data, and perform quantitative evaluation of the comprehensive battery health state to generate a battery health state evaluation result.

2. The method for evaluating the state of health of a new energy vehicle battery according to claim 1, wherein The specific steps of Step S1 are: Step S11: Obtain the multi-modal state monitoring parameters of the vehicle battery and the historical vehicle endurance mileage log; Step S12: Filter abnormal parameters from the multi-modal state monitoring parameters of the vehicle battery to obtain filtered and optimized battery state monitoring parameters; Step S13: Perform driving voltage time-series fluctuation analysis on the filtered and optimized battery state monitoring parameters to generate the battery voltage fluctuation characteristics of the driving state; Step S14: Perform discrete fitting of battery capacity attenuation on the filtered and optimized battery state monitoring parameters to construct a battery capacity attenuation curve; Step S15: Perform dynamic evolution of the battery state on the battery voltage fluctuation characteristics of the driving state and the battery capacity attenuation curve to obtain the multi-modal state evolution characteristics of the battery.

3. The method for evaluating the health state of a new energy vehicle battery according to claim 2, wherein The specific steps of Step S14 are: Collect the charge and discharge battery state from the filtered and optimized battery state monitoring parameters to obtain multiple charge and discharge battery state parameters; Perform battery internal resistance change analysis on multiple charge and discharge battery state parameters to generate battery internal resistance change characteristics; Based on the battery internal resistance change characteristics, perform real-time charge transfer performance evaluation to obtain multiple real-time charge transfer performance evaluation values; Perform polarization voltage evolution on multiple charge and discharge battery state parameters to obtain polarization voltage time-series evolution parameters; Perform sliding time window battery capacity change calculation on the polarization voltage time-series evolution parameters and multiple real-time charge transfer performance evaluation values to generate battery capacity change data for multiple time windows; Based on the battery capacity change data for multiple time windows, perform battery capacity attenuation analysis to obtain battery capacity attenuation parameters for multiple time windows; Perform discrete time-series fitting on the battery capacity attenuation parameters for multiple time windows to construct a battery capacity attenuation curve.

4. The method for evaluating the health state of a new energy vehicle battery according to claim 1, wherein, The specific steps of Step S2 are: Step S21: Perform multi-point maximum endurance calculation on the historical vehicle endurance mileage log to obtain the maximum endurance mileage at multiple time points; Step S22: Conduct successive adjacent endurance changes analysis on the maximum endurance mileage at multiple time points to generate vehicle endurance change characteristics; Step S23: Conduct periodic fluctuation evolution on the vehicle endurance change characteristics to generate vehicle endurance periodic fluctuation characteristics; Step S24: Conduct time series trend evolution on the vehicle endurance periodic fluctuation characteristics to generate vehicle endurance evolution trend data; Step S25: Conduct in-depth feature mining on the vehicle endurance evolution trend data to construct a vehicle endurance evolution trend graph.

5. The method for evaluating the health state of a new energy vehicle battery according to claim 1, wherein, The specific steps of Step S3 are as follows: Step S31: Conduct non-linear attenuation trend analysis on the vehicle endurance evolution trend graph to extract the endurance non-linear attenuation trend; Step S32: Conduct timestamp multi-point marking on the battery multi-modal state evolution characteristics and the endurance non-linear attenuation trend to extract multiple timestamp nodes; Step S33: Conduct parameter matching based on multiple timestamp nodes to obtain parameter matching time points; Step S34: Use the parameter matching time points to conduct time series cross-correlation mining on the battery multi-modal state evolution characteristics and the endurance non-linear attenuation trend to obtain battery state-endurance correlation characteristics; Step S35: Conduct battery aging mechanism evolution analysis based on the battery state-endurance correlation characteristics to construct a battery state aging evolution knowledge graph.

6. The method for evaluating the health state of a new energy vehicle battery according to claim 1, wherein, The specific steps of Step S4 are as follows: Step S41: Extract historical vehicle driving environment parameters according to the historical vehicle endurance mileage log; Step S42: Conduct environmental temperature change calculation on the historical vehicle driving environment parameters to obtain an environmental temperature change curve; Step S43: Conduct driving scenario recognition on the historical vehicle driving environment parameters to obtain different driving scenario characteristics; Step S44: Conduct environmental change impact analysis on the vehicle endurance evolution trend graph based on the environmental temperature change curve and different driving scenario characteristics to obtain endurance trend-environmental change impact rules; Step S45: Based on the endurance trend-environmental change impact rules and the battery state aging evolution knowledge graph, conduct dynamic vehicle state modeling on the vehicle battery multi-modal state monitoring parameters to construct a dynamic vehicle twin model.

7. The method for evaluating the state of health of a new energy vehicle battery according to claim 1, characterized in that, The specific steps of Step S5 are as follows: Step S51: Conduct driving scenario frequency calculation on the historical vehicle endurance mileage log to obtain different driving scenario frequency parameters; Step S52: Conduct driving scenario frequency peak calculation based on different driving scenario frequency parameters to extract the normalized driving scenario; Step S53: Conduct endurance evolution simulation for multiple time periods on the dynamic vehicle twin model based on the normalized driving scenario to extract endurance evolution simulation data for multiple time periods; Step S54: Conduct current scenario endurance situation analysis on the endurance evolution simulation data for multiple time periods to obtain current scenario endurance situation characteristics; Step S55: Conduct dynamic endurance rolling prediction based on the current scenario endurance situation characteristics to obtain long-term endurance trend prediction data.

8. The method for evaluating the health state of a new energy vehicle battery according to claim 1, wherein The specific steps of Step S6 are as follows: Step S61: Conduct endurance decay rate calculation on the long-term endurance trend prediction data to obtain the endurance decay rate; Step S62: Conduct battery life state prediction fitting based on the endurance decay rate to obtain a battery life state prediction curve; Step S63: Quantitatively evaluate the comprehensive battery health state of the battery life state prediction data to generate a battery health state evaluation result.

9. A new energy vehicle battery health state evaluation system, characterized in that, An apparatus for performing the new energy vehicle battery health state evaluation method as claimed in claim 1, comprising: A state evolution module, configured to obtain multi-modal state monitoring parameters of a vehicle battery and historical vehicle cruising range logs; perform dynamic evolution of the battery state on the multi-modal state monitoring parameters of the vehicle battery to obtain multi-modal battery state evolution features; A cruising range evolution trend module, configured to perform multi-point maximum cruising range calculation on the historical vehicle cruising range logs, and perform time series trend evolution to construct a vehicle cruising range evolution trend graph; A non-linear decay trend module, configured to perform non-linear decay trend analysis on the vehicle cruising range evolution trend graph, and perform battery aging mechanism evolution analysis, thereby constructing a battery state aging evolution knowledge graph; A twin model module, configured to perform environmental change impact analysis on the vehicle cruising range evolution trend graph according to the battery state aging evolution knowledge graph, perform dynamic vehicle state modeling, and construct a dynamic vehicle twin model; A cruising range trend prediction module, configured to perform cruising range evolution simulation for multiple periods on the dynamic vehicle twin model, and perform dynamic cruising range rolling prediction, thereby obtaining long-term cruising range trend prediction data; A health state evaluation module, configured to perform battery life state prediction fitting on the long-term cruising range trend prediction data, and perform quantitative evaluation of the comprehensive battery health state to generate a battery health state evaluation result.

Images