Classification and identification method and system of battery gradient and storage medium

Abstract

The invention relates to the technical field of battery gradient classification, in particular to a battery gradient classification identification method and system and a storage medium. The method comprises the following steps: carrying out dynamic workload log collection on a new energy automobile battery pack, and carrying out connection node load power consumption extraction to obtain connection node load power consumption data; theoretical load temperature fluctuation simulation is carried out on the connection node load power consumption data, and environmental influence effect analysis is carried out to obtain load environmental influence effect data; node output power consumption loss quantification is carried out according to load environment influence effect data, then performance output attenuation gradient analysis is carried out on connection node load power consumption data, a battery performance gradient classification and recognition model is constructed, then automatic firmware design is carried out, and battery performance gradient classification and recognition firmware is obtained. According to the method, the battery gradient classification technology is optimized, so that the battery gradient classification technology is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of battery gradient classification, and in particular to a battery gradient classification and identification method, system and storage medium. Background Art

[0002] The gradient classification and identification method of batteries refers to identifying the health status and usage level of different batteries by analyzing multi-dimensional data such as battery performance, status, and usage environment. The implementation of this method can not only improve the utilization efficiency of batteries, but also reduce resource waste and achieve sustainable development. In recent years, with the advancement of sensor technology, data acquisition and processing technology, especially the rise of the Internet of Things (IoT) and big data analysis, new possibilities have been provided for the gradient classification and identification of batteries. The key to battery gradient classification lies in how to select effective features and design a reasonable classification model. Feature selection needs to consider the physical and chemical properties of the battery and its performance under different working conditions. Through a comprehensive analysis of multi-dimensional factors such as the battery life cycle, charge and discharge efficiency, and thermal management, a feature vector with high recognition can be constructed. At the same time, the design of the classification model is also crucial. Commonly used classification algorithms include support vector machines (SVM), decision trees, random forests, etc. These algorithms can establish a mapping relationship between the battery health status and its features by learning historical data. However, a traditional battery gradient classification and identification method has the problem of unclear analysis of the impact of load energy consumption on battery performance under the influence of different environmental factors, thereby affecting the accuracy of classification. Summary of the invention

[0003] Based on this, it is necessary to provide a battery gradient classification and identification method, system and storage medium to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for classifying and identifying battery gradients is provided, the method comprising the following steps:

[0005] Step S1: obtaining the new energy vehicle system monitoring authority; collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain dynamic workload cleaning logs; extracting the connection node load power consumption according to the dynamic workload cleaning logs to obtain the connection node load power consumption data;

[0006] Step S2: performing a theoretical load temperature fluctuation simulation on the load power consumption data of the connection node to obtain theoretical load energy consumption temperature fluctuation data; performing an environmental impact effect analysis on the theoretical load energy consumption temperature fluctuation data to obtain load environmental impact effect data; quantifying the node output power consumption loss according to the load environmental impact effect data to obtain node output power consumption loss data;

[0007] Step S3: performing a performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; constructing a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model;

[0008] Step S4: Automated firmware design is performed on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

[0009] The present invention obtains the monitoring authority of the new energy vehicle system as a prerequisite for ensuring the legality and effectiveness of data collection. With this authority, the system can collect dynamic workload logs for the battery pack and record the working status of the battery under different working conditions. Then, these logs are cleaned and processed to extract the load power consumption data of the connected nodes. These data provide basic information for subsequent performance analysis, which can help engineers understand the energy consumption performance of the battery during actual use, thereby optimizing the battery management system and improving overall efficiency. By simulating the theoretical load temperature fluctuation of the load power consumption data of the connected nodes, the energy consumption performance under different environmental conditions is analyzed. This process can reveal the impact of load changes on battery temperature, and then evaluate the impact of different environmental factors (such as temperature and humidity) on battery performance. Subsequently, the output power consumption loss of the node is quantified through environmental impact effect analysis. These data not only provide an important basis for battery management, but also provide a reference for future optimization strategies to ensure that new energy vehicles can maintain stable performance under various environmental conditions. The node output power consumption loss data is used to analyze the load power consumption data of the connected nodes, thereby obtaining the gradient information of performance output attenuation. This analysis helps engineers identify the performance changes of batteries under different load conditions and provides support for the construction of battery performance gradient classification and recognition models. Through this model, the health status of the battery can be evaluated and classified more accurately, helping to formulate more scientific maintenance and management strategies, thereby extending the service life of the battery and improving the reliability of new energy vehicles. The battery performance gradient classification and identification model is designed with automated firmware, and finally a battery performance gradient classification and identification firmware is formed. After the firmware is embedded in the new energy vehicle system, real-time battery performance monitoring and classification and identification can be achieved. This automated process not only improves the intelligence level of battery management, but also can quickly respond to changes in battery performance, thereby optimizing battery efficiency, reducing energy consumption, and improving the overall performance and user experience of new energy vehicles, laying the foundation for the future development of intelligent driving and electric vehicles. Therefore, the present invention is an improved processing of a traditional battery gradient classification and identification method, which solves the problem that a traditional battery gradient classification and identification method has unclear analysis of the load energy consumption impact of the battery on the battery performance under the influence of different environmental factors, thereby affecting the accuracy of classification, and improves the clarity of the analysis of the load energy consumption impact of the battery on the battery performance under the influence of different environmental factors, and improves the accuracy of the impact classification.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Obtaining new energy vehicle system monitoring authority;

[0012] Step S12: collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain the dynamic workload logs of the battery pack;

[0013] Step S13: performing data cleaning on the dynamic workload log of the battery pack to obtain a dynamic workload cleaning log;

[0014] Step S14: extracting the connection structure of the new energy vehicle battery pack to obtain the battery pack connection structure data;

[0015] Step S15: extract the connection node load power consumption of the battery pack connection structure data according to the dynamic workload cleaning log to obtain the connection node load power consumption data.

[0016] The present invention obtains the monitoring authority of the new energy vehicle system, which is a crucial first step. This step ensures that the system can legally and effectively access the relevant data of the battery pack. By obtaining the authority, engineers can implement comprehensive data monitoring to ensure subsequent data collection. According to the monitoring authority obtained, the dynamic workload log of the new energy vehicle battery pack is collected. This process involves recording the various working states experienced by the battery in actual operation, covering charging, discharging and performance under different load conditions. The dynamic workload log provides key data support for subsequent analysis. Through these logs, engineers can deeply understand the working characteristics and energy consumption of the battery pack, thereby providing a basis for optimizing the battery management strategy. Data cleaning of the dynamic workload log of the battery pack is an important part of ensuring data quality. The cleaning process includes removing redundant information, correcting data errors and filling missing values ​​to ensure the accuracy of subsequent analysis. The effective execution of this process can significantly improve the reliability of the data, so that engineers can obtain real and effective results in subsequent analysis. This not only helps to improve the prediction accuracy of the model, but also helps to make more scientific decisions in practical applications. Extract the connection structure of the new energy vehicle battery pack. By analyzing the connection mode and structural layout inside the battery pack, engineers can obtain the connection structure data of the battery pack. This data is critical to understanding the flow of current and energy within the battery, and helps identify potential performance bottlenecks and failure points. Accurate connection structure data will provide a basis for subsequent energy consumption analysis and optimization, and help improve the overall efficiency and safety of the battery. Using dynamic workload cleaning logs, the connection node load power consumption is extracted from the battery pack connection structure data. This process calculates the energy consumption of each connection node by comprehensively considering the connection structure and workload status of the battery pack. These connection node load power consumption data are the key basis for subsequent performance analysis, and can help engineers accurately understand the energy consumption performance of the battery under different conditions. The acquisition of this data not only helps to optimize the battery management strategy, but also provides an important basis for improving the energy efficiency of new energy vehicles and extending battery life.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: collecting new energy vehicle battery environment data based on the new energy vehicle system monitoring authority to obtain battery environment data;

[0019] Step S22: performing load energy consumption baseline analysis on the load power consumption data of the connection nodes to obtain a node load energy consumption baseline;

[0020] Step S23: performing theoretical load temperature fluctuation simulation on the node load energy consumption baseline to obtain load energy consumption theoretical temperature fluctuation data;

[0021] Step S24: performing environmental impact analysis on the load energy consumption theoretical temperature fluctuation data according to the battery environment data to obtain load environment impact effect data;

[0022] Step S25: quantifying the node output power consumption loss of the connection node load power consumption data according to the load environment impact effect data to obtain the node output power consumption loss data.

[0023] The present invention is very important to collect battery environmental data based on the monitoring authority of the new energy vehicle system. This process involves monitoring the operating status of the battery under different environmental conditions, including external factors such as temperature, humidity, and air pressure. Acquiring these environmental data not only helps to fully understand the working conditions of the battery in actual use, but also provides necessary background information for subsequent analysis and model building. These data will provide an important basis for evaluating battery performance and its performance under different environmental conditions, and help optimize battery management strategies. The load energy consumption baseline is analyzed for the load power consumption data of the connected nodes. This process determines the energy consumption standards of the nodes under different working conditions through statistical analysis of historical data. This baseline not only helps to identify the energy consumption level of the node under normal operating conditions, but also provides a reference for subsequent performance evaluation. By accurately setting the energy consumption baseline, engineers can effectively monitor changes in battery performance, detect abnormal conditions in a timely manner, and take appropriate countermeasures to improve the reliability and efficiency of the battery. The simulation of theoretical load temperature fluctuations for the node load energy consumption baseline is a key link. Through this simulation, engineers can predict the temperature changes faced by the battery under different load conditions. This process can reveal the impact of load fluctuations on battery temperature and provide a theoretical basis for the thermal management strategy of the battery. By understanding these temperature fluctuation data, engineers can better design battery management systems to ensure that the battery operates within a safe temperature range, thereby improving its performance and service life. Environmental impact effect analysis is performed by combining battery environmental data with load energy consumption theoretical temperature fluctuation data. This analysis helps identify the specific impact of external environmental factors on battery performance and reveals the relationship between environmental conditions and energy consumption and temperature changes. Through systematic analysis, engineers can more comprehensively understand the behavioral characteristics of batteries under different environments, thereby optimizing the working status and management strategies of batteries and improving the overall performance of new energy vehicles. This process lays a solid foundation for achieving more efficient battery management. The node output power consumption loss is quantified based on the load environmental impact effect data for the connected node load power consumption data. This process involves calculating the energy consumption loss of each connected node under specific environmental conditions, identifying which nodes perform poorly, and providing a basis for optimizing battery management. This quantitative analysis not only helps to improve the energy efficiency of the battery, but also provides data support for subsequent fault diagnosis and maintenance strategies, ensuring that new energy vehicles maintain stable performance and low energy consumption during long-term use.

[0024] Preferably, step S24 comprises the following steps:

[0025] Step S241: extracting environmental impact factors from battery environmental data to obtain environmental impact factors;

[0026] Step S242: performing fluctuation extreme value scatter point distribution analysis on load energy consumption theoretical temperature fluctuation data to obtain theoretical fluctuation extreme value scatter point distribution data;

[0027] Step S243: performing scatter point nonlinear correlation analysis on the theoretical fluctuation extreme value scatter point distribution data according to the environmental impact factor to obtain scatter point extreme value fluctuation correlation data;

[0028] Step S244: Perform environmental impact analysis on the load environmental impact effect data to obtain the load environmental impact effect data.

[0029] The extraction of environmental impact factors from battery environmental data by the present invention is an important prerequisite for analyzing the impact of environmental factors on battery performance. By identifying and quantifying environmental factors related to battery performance, such as temperature, humidity, vibration, etc., engineers can deeply understand the role of these factors in battery operation. This process provides a clear framework for subsequent analysis, so that all aspects of environmental impact can be taken into account, thereby providing a scientific basis for improving battery management and performance optimization. The load energy consumption theoretical temperature fluctuation data is analyzed for the extreme value scatter distribution of fluctuations. This process helps engineers identify the maximum and minimum values ​​of temperature fluctuations under specific load conditions by statistically analyzing the extreme values ​​of temperature fluctuations under different conditions. Through this scatter distribution analysis, the regularity and abnormality of temperature fluctuations can be intuitively displayed, laying the foundation for subsequent environmental impact analysis. The acquisition of this data enables engineers to formulate more reasonable thermal management strategies, thereby improving the overall performance and safety of the battery. Scatter nonlinear correlation analysis of theoretical fluctuation extreme value scatter distribution data according to environmental impact factors is an important link in further understanding the relationship between environmental factors and temperature fluctuations. This analysis can reveal how different environmental factors affect the extreme value performance of temperature fluctuations in a nonlinear manner. Through this nonlinear correlation analysis, engineers can identify the key environmental factors that affect battery performance and quantify their specific impact on temperature fluctuations. This not only improves the accuracy of the analysis, but also provides strong support for subsequent battery performance optimization. Environmental impact effect analysis of load environmental impact effect data aims to quantify the comprehensive impact of environmental factors on battery performance. This analysis combines the environmental impact factors and temperature fluctuation related data obtained in the previous steps to evaluate the energy consumption performance and working status of the battery under different environmental conditions. Through systematic effect analysis, engineers can identify the most significant influencing factors and provide a scientific basis for optimizing battery management strategies. This process not only helps to improve the operating efficiency of the battery, but also enhances the reliability and safety of new energy vehicles under various environmental conditions.

[0030] Preferably, step S25 comprises the following steps:

[0031] Step S251: evaluating the impact strength between different nodes on the load power consumption data of the connected nodes according to the load environment impact effect data, and obtaining node impact strength evaluation data;

[0032] Step S252: Based on the node impact strength assessment data and the node load power consumption data, node power consumption growth pattern identification is performed to obtain node power consumption growth pattern data;

[0033] Step S253: performing adjacent node energy consumption growth difference identification on the node power consumption growth pattern data to obtain adjacent node energy consumption growth difference data;

[0034] Step S254: identifying abnormal growth trends between different nodes of the load power consumption data of the connected nodes according to the energy consumption growth difference data of the adjacent nodes, and obtaining abnormal growth trend data of the load power consumption;

[0035] Step S255: quantify the node output power consumption loss based on the load power consumption abnormal growth trend data to obtain the node output power consumption loss data.

[0036] The present invention evaluates the impact intensity between different nodes on the load power consumption data of the connected nodes according to the load environment impact effect data, and can identify the degree of mutual influence between the nodes. This evaluation not only reveals the importance of each node in the entire system, but also helps to determine which nodes have a significant impact on power consumption changes under specific environmental conditions. Through this method, engineers can optimize node configuration and management strategies in a targeted manner to reduce energy consumption losses and improve the overall efficiency of the system. By identifying the node power consumption growth pattern based on the node impact intensity evaluation data, it is intended to reveal the power consumption change law of different nodes under load conditions. This process involves analyzing the response characteristics of each node when facing different load environments, so as to help understand which nodes are prone to power consumption growth in actual operation. By identifying these growth patterns, engineers can better predict future energy consumption trends and provide a reliable basis for subsequent energy consumption management and optimization measures. By identifying the difference in energy consumption growth of adjacent nodes based on the node power consumption growth pattern data, the different performances of adjacent nodes in energy consumption growth can be effectively found. This analysis can reveal which nodes have energy efficiency imbalance problems and help identify potential performance bottlenecks. By understanding the energy consumption differences between adjacent nodes, engineers can optimize the coordination between nodes, ensure that the overall energy efficiency of the system is improved, and avoid waste of resources. Abnormal growth trends between different nodes are identified based on the energy consumption growth difference data of adjacent nodes. This process helps engineers find out whether some nodes have abnormal power consumption growth under certain circumstances, so that they can take corresponding measures to adjust. This identification can not only prevent potential equipment failures, but also promote the implementation of more efficient energy use strategies. By monitoring and analyzing these trends, engineers can effectively optimize energy management and improve the overall system operation stability. The quantification of node output power loss based on the abnormal growth trend data of load power consumption aims to provide a quantitative basis for evaluating the degree of node performance degradation. This quantitative analysis can accurately calculate the power consumption loss of each node under abnormal conditions, providing important data support for subsequent maintenance and optimization. In this way, engineers can develop more effective strategies to reduce energy consumption losses, improve the overall efficiency and reliability of the battery management system, and achieve more sustainable energy use.

[0037] Preferably, step S254 includes the following steps:

[0038] Calculate the growth rate ratio of the energy consumption growth difference data of adjacent nodes to obtain the energy consumption growth rate ratio;

[0039] According to the energy consumption growth rate ratio, the energy consumption transmission resistance value is evaluated by comparing the energy consumption growth difference data of adjacent nodes to obtain the node energy consumption transmission resistance value data;

[0040] According to the node energy consumption transmission resistance data and the energy consumption growth rate ratio, the dependency characteristics between adjacent nodes are analyzed by comparing the load power consumption data of the connected nodes to obtain the energy consumption dependency characteristics data of the adjacent nodes;

[0041] The abnormal growth trend between different nodes is identified based on the energy consumption dependency characteristic data of neighboring nodes to obtain the abnormal growth trend data of load power consumption.

[0042] In the process of calculating the growth rate ratio of the energy consumption growth difference data of adjacent nodes, the present invention enables engineers to intuitively compare the energy consumption growth rates between different nodes. This comparison not only reveals the relative intensity of the node energy consumption growth, but also provides basic data for the subsequent energy consumption transmission evaluation. By quantifying the energy consumption changes of each node, managers can more effectively identify nodes with excellent and poor performance, provide strong support for the formulation of subsequent optimization measures, and promote the improvement of the overall system energy efficiency. According to the calculated energy consumption growth rate ratio, the energy consumption transmission resistance value of the energy consumption growth difference data of adjacent nodes is evaluated, and the energy consumption transmission efficiency between each node can be quantified. This evaluation helps identify which nodes have bottlenecks in energy transmission, and then reveals the obstacles encountered when energy consumption is transmitted between nodes. By understanding these transmission resistance values, engineers can improve system design or adjust load distribution in a targeted manner to achieve more efficient energy use and lower energy consumption losses. The dependence characteristics analysis between adjacent nodes based on the node energy consumption transmission resistance value data and the energy consumption growth rate ratio can reveal the mutual influence and dependence between nodes. This analysis allows managers to understand which nodes' energy consumption changes will have a significant impact on other nodes under specific load conditions, thereby helping to carry out effective energy management and scheduling. The identification of this dependency characteristic not only improves the understanding of the energy consumption relationship between nodes, but also provides a basis for achieving overall optimization and coordinated operation of the system. When identifying abnormal growth trends between different nodes in the energy consumption dependency characteristic data of adjacent nodes, potential energy consumption anomalies can be discovered early. This process helps engineers quickly locate nodes that show abnormal growth under specific conditions, so that timely measures can be taken to adjust or optimize them. By monitoring these abnormal growth trends, not only can equipment failures be prevented, but the overall energy consumption management strategy can also be optimized, the safety and stability of the system can be improved, and a good foundation can be laid for sustainable development.

[0043] Preferably, step S3 comprises the following steps:

[0044] Step S31: matching the battery node output with the connected node load power consumption data according to the node output power loss data to obtain node output matching data;

[0045] Step S32: Perform performance output attenuation gradient analysis on the node output matching data to obtain performance output attenuation gradient data;

[0046] Step S33: identifying the critical state of battery performance attenuation according to the performance output attenuation gradient data to obtain the critical state data of battery performance attenuation;

[0047] Step S34: constructing a battery performance gradient classification and identification model based on the battery performance attenuation critical state data to obtain a battery performance gradient classification and identification model.

[0048] The present invention can effectively optimize the energy distribution of the battery under different load conditions by matching the battery node output with the connected node load power consumption data according to the node output power loss data. This process ensures that the battery can adjust its output according to actual needs, thereby avoiding energy waste caused by power consumption mismatch. Through precise matching, the system can extend the service life of the battery while maintaining the normal operation of the equipment, improve the overall energy efficiency, and lay a solid foundation for subsequent energy management. By performing performance output attenuation gradient analysis on the node output matching data, it is intended to identify and quantify the performance changes of the battery under different working conditions. This analysis not only reveals the changing trend of the battery output performance, but also helps engineers understand the impact of different load conditions on battery performance. Through this analysis, managers can better predict the performance attenuation of the battery, so as to take effective maintenance measures to ensure that the battery is always in the best working state and improve the reliability and efficiency of the system. According to the performance output attenuation gradient data, the critical state of battery performance attenuation is identified, and the key critical point of battery performance degradation can be accurately located. This identification process is an important part of evaluating the health status of the battery, helping managers understand when the battery needs to be replaced or maintained. By clarifying these critical states, engineers can intervene in time to avoid equipment failures caused by insufficient battery performance, thereby improving the overall safety and stability of the system. A battery performance gradient classification and identification model is constructed based on the critical state data of battery performance attenuation. The establishment of this model can not only provide a clear classification of battery performance status, but also provide strong support for future performance prediction and management decisions. By classifying and identifying battery performance, managers can more effectively formulate targeted maintenance strategies and optimization plans, thereby achieving more efficient energy management, extending battery life, reducing operating costs, and contributing to the sustainable development of the system.

[0049] Preferably, step S33 includes the following steps:

[0050] Step S331: performing time-dependent performance attenuation evaluation on the performance output attenuation gradient data to obtain time-dependent performance attenuation evaluation data;

[0051] Step S332: constructing a battery performance index according to the time-dependent performance attenuation evaluation data to obtain battery performance index data;

[0052] Step S333: performing a battery aging pre-response factor analysis based on the time-dependent performance attenuation evaluation data and the battery performance index data to obtain a battery aging pre-response factor;

[0053] Step S334: performing mechanistic induction on the battery aging pre-response factor to obtain aging pre-response mechanism induction data;

[0054] Step S335: Summarize the data based on the aging pre-response mechanism to identify the critical state of battery performance degradation, and obtain the critical state data of battery performance degradation.

[0055] The present invention performs a time-dependent performance attenuation evaluation on the performance output attenuation gradient data, which can effectively quantify the performance changes of the battery during long-term use. This evaluation not only helps to understand the attenuation trend of battery performance over time, but also can identify the performance influencing factors under specific use conditions. By timely obtaining the time-dependent performance attenuation evaluation data, managers can formulate corresponding maintenance plans, optimize the use and management strategies of the battery, extend the actual service life of the battery, and ensure the stable operation of the system. The battery performance index is constructed based on the time-dependent performance attenuation evaluation data, aiming to quantify the health status of the battery into an indicator that is easy to understand and analyze. This performance index provides a comprehensive evaluation method to help users quickly understand the current performance level of the battery and its potential attenuation risk. By establishing a battery performance index, managers can monitor the battery status more intuitively and make timely adjustments when necessary, thereby achieving more efficient energy management and resource allocation. By performing a battery aging pre-response factor analysis based on the time-dependent performance attenuation evaluation data and the battery performance index data, it is intended to identify the key influencing factors in the battery aging process. This analysis can help managers understand which factors most accelerate battery aging under different environments and usage conditions. By identifying these pre-response factors, engineers can develop more targeted preventive measures and optimize battery usage strategies to slow down the aging process and improve the overall performance and reliability of the battery. Mechanistic induction of battery aging pre-response factors helps to reveal the inherent mechanisms and laws of battery aging. This induction process not only strengthens the scientific understanding of aging phenomena, but also provides theoretical support for battery management. By deeply analyzing how each factor interacts and affects the performance of the battery, managers can optimize battery design and usage conditions in a targeted manner, further improve the stability and performance of the battery, and make the battery performance more predictable and reliable during use. According to the aging pre-response mechanism induction data, the critical state of battery performance degradation can be identified, which can effectively determine the critical point of battery performance degradation. This identification process provides an important basis for battery status monitoring, enabling managers to take timely intervention measures when the battery performance is close to the critical state to prevent the performance loss of the equipment due to battery failure. By clarifying these critical states, the intelligence level of battery management can be improved, thereby optimizing the overall energy efficiency and system stability.

[0056] Preferably, the present invention further provides a battery gradient classification and identification system, which is used to execute the battery gradient classification and identification method as described above, and the battery gradient classification and identification system comprises:

[0057] The power consumption extraction module is used to obtain the new energy vehicle system monitoring authority; according to the new energy vehicle system monitoring authority, the dynamic workload log of the new energy vehicle battery pack is collected to obtain the dynamic workload cleaning log; according to the dynamic workload cleaning log, the connection node load power consumption is extracted to obtain the connection node load power consumption data;

[0058] The output power loss quantification module is used to simulate the theoretical load temperature fluctuation of the load power consumption data of the connection node to obtain the theoretical temperature fluctuation data of the load energy consumption; analyze the environmental impact effect of the theoretical temperature fluctuation data of the load energy consumption to obtain the load environmental impact effect data; quantify the node output power loss according to the load environmental impact effect data to obtain the node output power loss data;

[0059] The performance gradient classification and identification model building module is used to perform performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; and to construct a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model;

[0060] The automated firmware design module is used to perform automated firmware design on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

[0061] Preferably, a storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the battery gradient classification and identification method described in any one of items 1 to 8 above.

[0062] The beneficial effect of the present invention is that obtaining the monitoring authority of the new energy vehicle system is a prerequisite for ensuring the legality and effectiveness of data collection. With this authority, the system can collect dynamic workload logs for the battery pack and record the working status of the battery under different working conditions. Then, these logs are cleaned and processed to extract the load power consumption data of the connected nodes. These data provide basic information for subsequent performance analysis, which can help engineers understand the energy consumption performance of the battery during actual use, thereby optimizing the battery management system and improving overall efficiency. By simulating the theoretical load temperature fluctuation of the load power consumption data of the connected nodes, the energy consumption performance under different environmental conditions is analyzed. This process can reveal the impact of load changes on battery temperature, and then evaluate the impact of different environmental factors (such as temperature and humidity) on battery performance. Subsequently, the output power consumption loss of the node is quantified through environmental impact effect analysis. These data not only provide an important basis for battery management, but also provide a reference for future optimization strategies to ensure that new energy vehicles can maintain stable performance under various environmental conditions. The node output power consumption loss data is used to analyze the load power consumption data of the connected nodes to obtain the gradient information of performance output attenuation. This analysis helps engineers identify the performance changes of batteries under different load conditions and provides support for the construction of battery performance gradient classification and recognition models. Through this model, the health status of the battery can be evaluated and classified more accurately, helping to formulate more scientific maintenance and management strategies, thereby extending the service life of the battery and improving the reliability of new energy vehicles. The battery performance gradient classification and identification model is designed with automated firmware, and finally a battery performance gradient classification and identification firmware is formed. After the firmware is embedded in the new energy vehicle system, real-time battery performance monitoring and classification and identification can be achieved. This automated process not only improves the intelligence level of battery management, but also can quickly respond to changes in battery performance, thereby optimizing battery efficiency, reducing energy consumption, and improving the overall performance and user experience of new energy vehicles, laying the foundation for the future development of intelligent driving and electric vehicles. Therefore, the present invention is an improved processing of a traditional battery gradient classification and identification method, which solves the problem that a traditional battery gradient classification and identification method has unclear analysis of the load energy consumption impact of the battery on the battery performance under the influence of different environmental factors, thereby affecting the accuracy of classification, and improves the clarity of the analysis of the load energy consumption impact of the battery on the battery performance under the influence of different environmental factors, and improves the accuracy of the impact classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic flow chart of the steps of a battery gradient classification and identification method;

[0064] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0068] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0069] To achieve this, please refer to Figure 1 to Figure 2 , a classification and identification method for battery gradients, the method comprising the following steps:

[0070] Step S1: obtaining the new energy vehicle system monitoring authority; collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain dynamic workload cleaning logs; extracting the connection node load power consumption according to the dynamic workload cleaning logs to obtain the connection node load power consumption data;

[0071] Step S2: performing a theoretical load temperature fluctuation simulation on the load power consumption data of the connection node to obtain theoretical load energy consumption temperature fluctuation data; performing an environmental impact effect analysis on the theoretical load energy consumption temperature fluctuation data to obtain load environmental impact effect data; quantifying the node output power consumption loss according to the load environmental impact effect data to obtain node output power consumption loss data;

[0072] Step S3: performing a performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; constructing a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model;

[0073] Step S4: Automated firmware design is performed on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

[0074] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a battery gradient classification and identification method of the present invention. In this example, the battery gradient classification and identification method includes the following steps:

[0075] Step S1: obtaining the new energy vehicle system monitoring authority; collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain dynamic workload cleaning logs; extracting the connection node load power consumption according to the dynamic workload cleaning logs to obtain the connection node load power consumption data;

[0076] In an embodiment of the present invention, during the monitoring process of a new energy vehicle, it is first necessary to obtain the monitoring authority of the system. This process communicates with the internal management system of the new energy vehicle and uses a dedicated identity authentication protocol to ensure that only authorized users can access relevant data. After completing the authority verification, a dedicated interface is used to collect the dynamic workload log of the battery pack. In specific operations, the monitoring frequency is set to once per minute to record the real-time load of the battery pack under different working conditions, including current, voltage and temperature data during charging and discharging. These data are organized into a dynamic workload cleaning log, which records the status information of all abnormal data and normal data within the cycle. Subsequently, based on the valid data in the cleaning log, the load power consumption of the connection node is extracted to obtain the load power consumption data of the connection node. This data records the power consumption performance of the connection node under different load conditions, providing a basis for subsequent analysis.

[0077] Step S2: performing a theoretical load temperature fluctuation simulation on the load power consumption data of the connection node to obtain theoretical load energy consumption temperature fluctuation data; performing an environmental impact effect analysis on the theoretical load energy consumption temperature fluctuation data to obtain load environmental impact effect data; quantifying the node output power consumption loss according to the load environmental impact effect data to obtain node output power consumption loss data;

[0078] In an embodiment of the present invention, the load power consumption data of the connection node is used to carry out a theoretical load temperature fluctuation simulation. This simulation adopts the heat conduction theory, and by establishing a heat transfer model of the node, the temperature change under different power consumption conditions is calculated. Using Fourier's law, combined with the thermal conductivity of the node material and the ambient temperature, the temperature fluctuation of the battery under different workloads is simulated, and the load energy consumption theoretical temperature fluctuation data is generated. After obtaining the temperature fluctuation data, an environmental impact effect analysis is performed, considering the impact of environmental factors such as external temperature, humidity and wind speed on the load temperature, and using the linear regression analysis method, the influencing factors are extracted and their actual impact on the load is quantified, thereby obtaining the load environmental impact effect data. Finally, the node output power consumption loss is quantified based on these data, and the energy balance method is used to calculate the energy loss caused by environmental factors to obtain the node output power consumption loss data.

[0079] Step S3: performing a performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; constructing a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model;

[0080] In an embodiment of the present invention, a performance output attenuation gradient analysis is performed based on the node output power consumption loss data. First, the output power consumption loss data is gradually derived using the differential method, and the changing trend of the power consumption loss in the time series is analyzed to obtain the performance output attenuation gradient data. This data describes how the performance output of the node decays over time under a specific load. Next, a battery performance gradient classification and recognition model is constructed based on the performance output attenuation gradient data. By adopting a multi-layer neural network algorithm, the input layer is set to the power consumption loss data, the hidden layer is adjusted according to the feature dimension to identify batteries of different performance levels, and the output layer is the classification result. After model training, a battery performance gradient classification and recognition model is obtained, which is used for subsequent rapid identification and classification of battery performance.

[0081] Step S4: Automated firmware design is performed on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

[0082] In an embodiment of the present invention, an automated firmware design is performed for a battery performance gradient classification and recognition model. During the design process, the model is first converted into a firmware format suitable for embedded system operation, and the program is written in C language to ensure high efficiency and real-time performance. The firmware includes a data input interface, a model inference engine, and an output result interface. The data input interface is responsible for obtaining battery performance data from the new energy vehicle system in real time. The inference engine processes the input data according to a preset recognition algorithm, and finally returns the recognition result to the system through the output result interface. The design of the firmware follows the principles of real-time and low power consumption to ensure that it does not affect the overall performance during the operation of the new energy vehicle. After the firmware design is completed, it is embedded in the new energy vehicle system so that it can perform classification and recognition of battery gradients in actual operation, thereby improving the system's response speed and accuracy to battery performance monitoring.

[0083] Preferably, step S1 comprises the following steps:

[0084] Step S11: Obtaining new energy vehicle system monitoring authority;

[0085] Step S12: collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain the dynamic workload logs of the battery pack;

[0086] Step S13: performing data cleaning on the dynamic workload log of the battery pack to obtain a dynamic workload cleaning log;

[0087] Step S14: extracting the connection structure of the new energy vehicle battery pack to obtain the battery pack connection structure data;

[0088] Step S15: extract the connection node load power consumption of the battery pack connection structure data according to the dynamic workload cleaning log to obtain the connection node load power consumption data.

[0089] In the embodiment of the present invention, the process of obtaining the monitoring authority of the new energy vehicle system first involves effective communication with the internal management system of the new energy vehicle. In the specific implementation, an encryption protocol based on identity authentication is adopted to ensure that only devices or users with corresponding permissions can request access to the monitoring data by allocating a unique access token. This process requires the establishment of a diagnostic interface connected to the new energy vehicle, the implementation of security authentication measures, and the recording of all access logs for subsequent review. The core of this step is to verify and confirm the user's identity, ensure the security of the system and the accuracy of the data, and avoid the potential risks of unauthorized access to the system. After obtaining the monitoring authority, the dynamic workload log of the battery pack is collected. This process monitors the operating status of the battery pack in real time through embedded sensors, including multiple parameters such as current, voltage, and temperature. The data collection frequency is set to once per second to ensure that the changes in the battery under different working conditions can be captured. The collected data is recorded in time series to form a dynamic workload log of the battery pack, which includes power consumption data and timestamps during charging and discharging. The data format must comply with a unified standard for subsequent processing and analysis. After obtaining the dynamic workload log of the battery pack, data cleaning is performed to eliminate invalid and abnormal data. First, the data is preliminarily screened using the set threshold to exclude values ​​that are beyond the normal operating range, such as excessive temperature or abnormal voltage. Next, the interpolation method is applied to fill in the missing data points to ensure the integrity and continuity of the data. Finally, the data distribution is further checked by the standard deviation analysis method, and outliers exceeding three standard deviations are eliminated, and finally a dynamic workload cleaning log is generated. This cleaning process ensures that subsequent analysis is based on high-quality data and improves the reliability of the analysis results. In the process of extracting the connection structure of the battery pack, the physical structure diagram and connection diagram of the battery pack must first be obtained to facilitate the analysis of its internal connection nodes. The relative position and electrical characteristics of each connection point, including the resistance and inductance between the connection points, are recorded using physical measurement tools. Next, the node analysis method in graph theory is used to construct a connection network model of the battery pack to identify each connection node and their mutual relationship. This process ensures that each connection of the battery pack is marked in detail to form the battery pack connection structure data for subsequent power extraction and performance analysis. According to the dynamic workload cleaning log, the load power consumption of the connection node is extracted. This step needs to be combined with the battery pack connection structure data, and the power consumption of each connected node in different time periods is calculated one by one according to the timestamp and power consumption information in the data. In specific implementation, the weighted average method is applied to calculate the power consumption value of each node according to the actual load of each node during the operation of the battery pack. By allocating the overall power consumption of the battery pack according to the connected nodes, the load power consumption data of the connected nodes is obtained. This data provides a basis for subsequent analysis and helps to identify the energy consumption characteristics of different nodes during operation.

[0090] Preferably, step S2 comprises the following steps:

[0091] Step S21: collecting new energy vehicle battery environment data based on the new energy vehicle system monitoring authority to obtain battery environment data;

[0092] Step S22: performing load energy consumption baseline analysis on the load power consumption data of the connection nodes to obtain a node load energy consumption baseline;

[0093] Step S23: performing theoretical load temperature fluctuation simulation on the node load energy consumption baseline to obtain load energy consumption theoretical temperature fluctuation data;

[0094] Step S24: performing environmental impact analysis on the load energy consumption theoretical temperature fluctuation data according to the battery environment data to obtain load environment impact effect data;

[0095] Step S25: quantifying the node output power consumption loss of the connection node load power consumption data according to the load environment impact effect data to obtain the node output power consumption loss data.

[0096] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0097] Step S21: collecting new energy vehicle battery environment data based on the new energy vehicle system monitoring authority to obtain battery environment data;

[0098] In an embodiment of the present invention, after obtaining the monitoring authority of the new energy vehicle system, the battery environment data is collected. The process first implements sensor data collection through an interface connection with the battery management system to record environmental parameters, including temperature, humidity, air pressure, and air flow rate around the battery pack. The collection frequency of each parameter is set to once per minute to capture the environmental changes of the battery under different working conditions. The data is transmitted in real time through the physical interface and stored in a time series format to form a battery environment data set. This data set is used for subsequent analysis to ensure a comprehensive understanding of the battery environment status and provide a basis for evaluating the relationship between battery performance and the environment.

[0099] Step S22: performing load energy consumption baseline analysis on the load power consumption data of the connection nodes to obtain a node load energy consumption baseline;

[0100] In an embodiment of the present invention, a load energy consumption baseline analysis is performed on the acquired load power consumption data of the connected nodes. First, the power consumption data over a period of time is collected, and a reference time window is set, such as a week of data, to establish the normal working range of the node. A statistical analysis method is used to calculate the mean and standard deviation of the power consumption data within the time window, and determine the power consumption baseline value under normal working conditions. By comparing the actual collected power consumption data with the baseline value, power consumption fluctuations that exceed the normal range are identified. This analysis not only provides a reference value for subsequent load temperature fluctuation simulations, but also lays the foundation for identifying potential performance issues.

[0101] Step S23: performing theoretical load temperature fluctuation simulation on the node load energy consumption baseline to obtain load energy consumption theoretical temperature fluctuation data;

[0102] In an embodiment of the present invention, a theoretical load temperature fluctuation simulation is performed based on the node load energy consumption baseline. The simulation first establishes a heat conduction model, taking into account the thermal characteristics of the node and the influencing factors of the external environment. The heat balance equation is applied, and the corresponding temperature changes are calculated in combination with the power consumption of the node under different load conditions. Through the numerical simulation method, the load conditions under different working conditions are set, and the temperature fluctuations of the nodes under these conditions are simulated. The simulation results form the load energy consumption theoretical temperature fluctuation data, record the theoretical temperature changes of the nodes at different time points, and provide data support for subsequent environmental impact analysis.

[0103] Step S24: performing environmental impact analysis on the load energy consumption theoretical temperature fluctuation data according to the battery environment data to obtain load environment impact effect data;

[0104] In an embodiment of the present invention, after obtaining the theoretical temperature fluctuation data of the load energy consumption, an environmental impact effect analysis is performed. This step evaluates the degree of influence of environmental factors on the load temperature fluctuation based on the collected battery environment data. First, multiple environmental factors such as temperature, humidity and air flow velocity are set as input variables, and a multivariate linear regression model is used to correlate the environmental data with the temperature fluctuation data to quantify the impact of each environmental factor on the temperature fluctuation. By analyzing the regression results, load environmental impact effect data is generated to describe the changing characteristics of the load temperature fluctuation under specific environmental conditions, providing the necessary data for the subsequent quantification of node power consumption losses.

[0105] Step S25: quantifying the node output power consumption loss of the connection node load power consumption data according to the load environment impact effect data to obtain the node output power consumption loss data.

[0106] In an embodiment of the present invention, load environment impact effect data is introduced to perform node power consumption analysis. After determining the function and load characteristics of each connected node, the power consumption data of each node is collected and combined with the load environment impact effect data. Next, the weighted average method is used to quantify the node power consumption loss. This method takes into account the difference between the actual power consumption and the expected power consumption of the node under different load conditions. By setting the weight coefficient, it is ensured that the power consumption loss under high load and low load conditions is accurately reflected. During the implementation process, the power consumption loss of each node is statistically analyzed and visualized using histograms and box plots to identify potential high-loss nodes. Finally, the node output power consumption loss data is generated to provide a basis for subsequent energy efficiency improvement measures. This process ensures that each step of data conversion and processing is based on rigorous mathematical theory and practical operations, avoiding ambiguity and uncertainty.

[0107] Preferably, step S24 comprises the following steps:

[0108] Step S241: extracting environmental impact factors from battery environmental data to obtain environmental impact factors;

[0109] Step S242: performing fluctuation extreme value scatter point distribution analysis on load energy consumption theoretical temperature fluctuation data to obtain theoretical fluctuation extreme value scatter point distribution data;

[0110] Step S243: performing scatter point nonlinear correlation analysis on the theoretical fluctuation extreme value scatter point distribution data according to the environmental impact factor to obtain scatter point extreme value fluctuation correlation data;

[0111] Step S244: Perform environmental impact analysis on the load environmental impact effect data to obtain the load environmental impact effect data.

[0112] In the embodiment of the present invention, in this step, environmental factors related to battery performance are first extracted from the collected battery environmental data. These factors include temperature, humidity, air pressure, and the strength of the surrounding electromagnetic field. This process is carried out through feature selection technology, and the environmental data is reduced in dimension using principal component analysis (PCA) to identify the variables that have the greatest impact on the load energy consumption fluctuation. During implementation, the variability of each environmental factor needs to be calculated, and the factor with a cumulative variance contribution rate of 85% is selected as the environmental impact factor. Subsequently, the extracted factors are standardized to ensure that they are compared under the same dimension. Finally, the obtained environmental impact factors provide basic data for subsequent analysis, so that the load energy consumption fluctuation under different environmental conditions can be accurately quantified. The load energy consumption theoretical temperature fluctuation data is subjected to extreme value scatter distribution analysis to identify extreme points in the data. First, the basic statistical characteristics of the fluctuation data are determined by calculating the mean and standard deviation of the fluctuation data. Then, the Z-score method is used to calculate the deviation of each fluctuation point relative to the mean, and the extreme points exceeding the set threshold are identified. In this step, a 90% confidence interval is used as a threshold to ensure that the selected extreme value points are statistically significant. Subsequently, these extreme value points are plotted into a scatter plot to display their distribution in a visual way and analyze their relationship with the time series. Finally, the generated theoretical fluctuation extreme value scatter distribution data provides a basis for subsequent nonlinear correlation analysis, ensuring the reliability and accuracy of the analysis results. Scatter nonlinear correlation analysis is performed on the environmental impact factors extracted in step S241 and the theoretical fluctuation extreme value scatter distribution data obtained in step S242. First, the Spearman rank correlation coefficient is used to evaluate the correlation between environmental impact factors and fluctuation extreme values. This method can effectively capture nonlinear relationships. During implementation, the environmental impact factors and fluctuation extreme value data are sorted, the rank difference of each pair of variables is calculated, and then substituted into the Spearman correlation formula to obtain the correlation coefficient. Then, through visualization means, such as scatter plots and curve fitting, the scatter distribution of the data and its nonlinear relationship are deeply observed. After the analysis is completed, the scatter extreme value fluctuation correlation data is obtained, which provides sufficient support for further environmental impact effect analysis. The scatter point extreme value fluctuation related data obtained in step S243 is used to comprehensively analyze the load environmental impact effect. First, the scatter point extreme value fluctuation related data is combined with the extracted environmental impact factors to construct a multivariate regression model, aiming to reveal the specific impact of environmental factors on load energy consumption. In this process, a multicollinearity test is performed to ensure that there is no significant correlation between the respective variables. Next, the regression analysis method is used to estimate the degree of influence of each environmental factor on load energy consumption and obtain the environmental impact effect analysis results. Finally, the analysis results are normalized for comparison with other data, and finally the load environmental impact effect data is generated.This data provides an important reference for the classification and identification method of battery gradients, ensuring the scientificity and accuracy of subsequent operations.

[0113] Preferably, step S25 comprises the following steps:

[0114] Step S251: evaluating the impact strength between different nodes on the load power consumption data of the connected nodes according to the load environment impact effect data, and obtaining node impact strength evaluation data;

[0115] Step S252: Based on the node impact strength assessment data and the node load power consumption data, node power consumption growth pattern identification is performed to obtain node power consumption growth pattern data;

[0116] Step S253: performing adjacent node energy consumption growth difference identification on the node power consumption growth pattern data to obtain adjacent node energy consumption growth difference data;

[0117] Step S254: identifying abnormal growth trends between different nodes of the load power consumption data of the connected nodes according to the energy consumption growth difference data of the adjacent nodes, and obtaining abnormal growth trend data of the load power consumption;

[0118] Step S255: quantify the node output power consumption loss based on the load power consumption abnormal growth trend data to obtain the node output power consumption loss data.

[0119] In an embodiment of the present invention, firstly, the load environment impact effect data and the load power consumption data of the connection nodes are collected. According to the load environment impact effect data, the environmental factors affecting the power consumption of each node are identified. Then, the weight normalization method is used to convert the environmental impact effect data into a standardized form to ensure that the environmental impact factors between different nodes can be compared. Subsequently, the power consumption data of each node is evaluated by the weighted average method to calculate the influence intensity of each node on other nodes. Specifically, by setting an influence matrix, each element in the matrix represents the influence degree of one node on another node. During the calculation process, it is necessary to ensure that the connection relationship between the nodes is clear, and the connected graph algorithm is used to ensure that all nodes can be effectively connected. Finally, based on the calculation results, the node impact intensity evaluation data is generated for subsequent analysis. Using the node impact intensity evaluation data obtained in step S251, the load power consumption data of each connection node is analyzed to identify the growth mode of the node power consumption. First, the cluster analysis technology is applied to divide the power consumption data into several categories according to the growth trend. Using the K-means algorithm, the nodes are classified into different power consumption growth modes by calculating the change rate of the historical data of each node power consumption. In this process, it is necessary to set a reasonable number of clusters and verify them according to indicators such as the silhouette coefficient to ensure the validity of the clustering results. Subsequently, the characteristics of the node power consumption in each category are analyzed, and the node power consumption growth pattern data is generated to describe the power consumption change characteristics of different nodes in a specific time period. These data provide a basis for further identifying energy consumption differences and abnormal trends. Based on the node power consumption growth pattern data obtained in step S252, the energy consumption growth differences between adjacent nodes are analyzed. The power consumption growth rates of adjacent nodes are compared using a differentiation analysis method. Specifically, by calculating the difference in the power consumption growth of adjacent nodes, the growth difference of each pair of adjacent nodes is obtained. In the implementation process, a statistical significance test (such as a t-test) is used to verify whether the energy consumption growth difference between adjacent nodes is statistically significant. At the same time, in order to show the difference more intuitively, a box plot or a bar chart is used for visualization so as to quickly identify node pairs with significant growth differences. Finally, adjacent node energy consumption growth difference data is generated to provide data support for the next step of abnormal growth trend identification. According to the adjacent node energy consumption growth difference data obtained in step S253, abnormal growth trends between different nodes are identified. First, a growth threshold is set to determine what degree of growth is considered abnormal. Using the control chart method, an energy consumption growth control chart is constructed, and abnormal points are identified by statistically analyzing the upper and lower control limits of the control chart. During the analysis process, the energy consumption growth rate of each node is calculated and compared with the set control limits to identify nodes that exceed the control limits. This step also needs to be combined with time series analysis to ensure that the identified abnormal growth trend has temporal coherence and avoid short-term fluctuation interference. Finally, the load power consumption abnormal growth trend data is generated to provide necessary information for the subsequent quantification of power consumption losses.According to the abnormal growth trend data of load power consumption obtained in step S254, the output power consumption loss of the node is quantified. First, the power consumption data of each node in the abnormal state and the power consumption baseline data in the normal state are collected. Then, the power consumption loss of each node in the abnormal state is calculated using the formula: power consumption loss = abnormal power consumption - normal power consumption. For each node, ensure that the calculation of power consumption loss takes into account the average value of multiple measurements to improve the stability and reliability of the results. During the implementation process, data cleaning is also required to remove outliers and noise to ensure the accuracy of the calculation results. Finally, the node output power consumption loss data is generated to provide a detailed basis for subsequent energy efficiency management and improvement measures.

[0120] Preferably, step S254 includes the following steps:

[0121] Calculate the growth rate ratio of the energy consumption growth difference data of adjacent nodes to obtain the energy consumption growth rate ratio;

[0122] According to the energy consumption growth rate ratio, the energy consumption transmission resistance value is evaluated by comparing the energy consumption growth difference data of adjacent nodes to obtain the node energy consumption transmission resistance value data;

[0123] According to the node energy consumption transmission resistance data and the energy consumption growth rate ratio, the dependency characteristics between adjacent nodes are analyzed by comparing the load power consumption data of the connected nodes to obtain the energy consumption dependency characteristics data of the adjacent nodes;

[0124] The abnormal growth trend between different nodes is identified based on the energy consumption dependency characteristic data of neighboring nodes to obtain the abnormal growth trend data of load power consumption.

[0125] In an embodiment of the present invention, the energy consumption growth difference data of adjacent nodes is first collected to ensure that the data includes the energy consumption changes of each node at different time points. By calculating the energy consumption growth rate of each node, the growth rate is compared with the adjacent nodes to identify the relative change of energy consumption. In specific operation, the formula is used: growth rate ratio = (node ​​A growth rate) / (node ​​B growth rate). In the calculation process, it is necessary to ensure that the time series data between nodes are aligned to avoid calculation errors caused by time dislocation. Finally, through the calculation of this ratio, it is identified which nodes show a strong relative influence in energy consumption growth, and energy consumption growth rate ratio data is formed for subsequent analysis. The next step is to evaluate the energy consumption transmission resistance. According to the energy consumption growth rate ratio obtained in the step, the energy consumption transmission resistance of adjacent nodes is evaluated by the proportional method. In specific operation, the energy consumption transmission resistance of a reference node is set to 1, and the resistance of other nodes is adjusted according to their growth rate ratio. For example, for a certain node, its transmission resistance can be calculated by the formula: transmission resistance = 1 / (growth rate ratio + ε) (ε is a small constant to avoid division by zero). This evaluation ensures that the relative intensity of the energy consumption impact between nodes can be quantified, and finally forms the node energy consumption transmission resistance data, which provides a basis for further analysis. Based on the obtained node energy consumption transmission resistance data, the load power consumption data of the connected nodes is analyzed for dependency characteristics. First, a dependency matrix is ​​established, and each element in the matrix represents the degree of dependence of one node on another node. The degree of dependence is calculated by the following formula: Dependence = energy consumption transmission resistance × load power consumption of the corresponding node. This process needs to ensure the timeliness of all data and consider multiple dependencies between nodes. By analyzing the dependency matrix, it is possible to identify which nodes show obvious dependency characteristics in power consumption changes. Finally, the energy consumption dependency characteristic data of neighboring nodes is formed, which provides a basis for further abnormal trend identification. Using the obtained energy consumption dependency characteristic data of neighboring nodes, the abnormal growth trend of load power consumption is identified. First, the energy consumption dependency characteristics of neighboring nodes are comprehensively analyzed, and a time series model is established to analyze the energy consumption change trend between nodes within a specific time range. Use time series analysis methods, such as the autoregressive moving average (ARMA) model, to evaluate the normal fluctuation range of node power consumption. On this basis, an abnormal threshold is set to determine what degree of deviation is considered abnormal. By comparing with the threshold, the growth trend beyond the normal range is identified. Finally, the abnormal growth trend data of load power consumption is generated, providing important information for subsequent power loss quantification and energy efficiency management. This series of operations ensures that the identification of abnormal growth is highly accurate and reliable.

[0126] Preferably, step S3 comprises the following steps:

[0127] Step S31: matching the battery node output with the connected node load power consumption data according to the node output power loss data to obtain node output matching data;

[0128] Step S32: Perform performance output attenuation gradient analysis on the node output matching data to obtain performance output attenuation gradient data;

[0129] Step S33: identifying the critical state of battery performance attenuation according to the performance output attenuation gradient data to obtain the critical state data of battery performance attenuation;

[0130] Step S34: constructing a battery performance gradient classification and identification model based on the battery performance attenuation critical state data to obtain a battery performance gradient classification and identification model.

[0131] In an embodiment of the present invention, node output power loss data is used, which includes the power loss of each node in a specific time period. Then, the data is used to match and analyze the load power consumption data of the connected node. In specific operations, it is necessary to ensure that the power consumption data and loss data of the node are aligned in time series to avoid mismatching caused by time dislocation. The output power loss data and the load power consumption data are fitted using a linear regression method to determine the relationship between the two. During the fitting process, the correlation coefficient needs to be calculated to ensure the validity of the fitting. By establishing a regression equation, the node output power loss data is quantitatively matched with the load power consumption data of the connected node, thereby generating node output matching data. This data provides a basis for subsequent performance analysis, ensuring that the calculation of subsequent steps can be carried out with the support of valid data. The obtained node output matching data is analyzed, and the specific steps include using a multivariate linear regression analysis method to determine the degree of influence of different factors on battery performance. Select appropriate performance indicators, such as battery charge and discharge efficiency, cycle life, etc., build a regression model, and calculate the attenuation trend of each performance indicator. On this basis, the gradient information of the data is extracted through statistical analysis software to generate performance output attenuation gradient data. The gradient data reflects the rate at which the performance output changes over time or usage conditions. By drawing a performance decay curve, its changing rules are further analyzed to provide basic data for subsequent critical state identification. According to the performance output decay gradient data obtained in step S32, the critical state of battery performance decay is identified. First, a threshold range is set to determine which nodes' decay gradients are considered critical. The mean and standard deviation of the decay gradient data are calculated using statistical methods, and the upper and lower limits of the critical state are set to mean ±2 times the standard deviation. Then, the decay gradient of each node is analyzed and compared with the set threshold to identify which nodes' performance has entered the critical state of decay. This process needs to ensure the integrity and timeliness of the data to avoid misjudgment caused by incomplete historical data. Finally, the critical state data of battery performance decay is generated to provide a clear identification for subsequent performance classification identification and ensure the accuracy of subsequent analysis. According to the critical state data of battery performance decay generated in step S33, the battery performance gradient classification and identification model is constructed. First, a suitable classification algorithm, such as a decision tree algorithm, is selected to construct a classification tree model by analyzing the critical state data. The nodes of the model represent different performance gradient features, and each branch represents a decision rule. The model is trained using the training set data to ensure that the model can effectively identify the battery performance in different gradient states. The accuracy of the model is evaluated through cross-validation to ensure the stability of the classification effect. After the model is built, a battery performance gradient classification and recognition model is generated. This model can identify the battery performance gradient in real time based on the input battery status data, providing support for subsequent battery management and maintenance.This process needs to ensure the applicability and effectiveness of the model to ensure that it can quickly respond to changes in battery status in practical applications.

[0132] Preferably, step S33 includes the following steps:

[0133] Step S331: performing time-dependent performance attenuation evaluation on the performance output attenuation gradient data to obtain time-dependent performance attenuation evaluation data;

[0134] Step S332: constructing a battery performance index according to the time-dependent performance attenuation evaluation data to obtain battery performance index data;

[0135] Step S333: performing a battery aging pre-response factor analysis based on the time-dependent performance attenuation evaluation data and the battery performance index data to obtain a battery aging pre-response factor;

[0136] Step S334: performing mechanistic induction on the battery aging pre-response factor to obtain aging pre-response mechanism induction data;

[0137] Step S335: Summarize the data based on the aging pre-response mechanism to identify the critical state of battery performance degradation, and obtain the critical state data of battery performance degradation.

[0138] In an embodiment of the present invention, first, the time-related performance information is extracted from the performance output attenuation gradient data obtained in step S32. In order to evaluate the performance changes of the battery in different time periods, a set of time series analysis methods needs to be established. The performance output attenuation gradient data is grouped by time, and time scales such as days, weeks, and months can usually be selected for analysis. The data in each time period is statistically calculated to find the average attenuation rate and standard deviation to quantify the characteristics of battery performance changes over time. Next, the regression analysis method is used to explore the relationship between time and performance attenuation. Select a suitable regression model, such as linear regression or polynomial regression, to fit the time and performance attenuation gradient data to obtain a regression curve reflecting the performance attenuation trend. By fitting the curve, the attenuation speed of different time periods can be intuitively displayed, and the future performance attenuation can be predicted. Finally, combined with these analysis results, the time-related performance attenuation evaluation data is formed, specifically including the average attenuation rate, fluctuation range, and regression analysis results in different time periods, which provide the necessary quantitative basis for the subsequent construction of the battery performance index. The battery performance index is constructed according to the time-related performance attenuation evaluation data generated in step S331. First, determine the components of the battery performance index, which include charge and discharge efficiency, remaining capacity, number of cycles, internal resistance, etc. Each factor needs to be standardized to eliminate the influence between different dimensions. Next, the aging performance attenuation evaluation data is associated with the determined performance indicators, and a weighted summation method is used to construct a comprehensive battery performance index. Specifically, the corresponding weights are assigned according to the importance of each battery performance indicator, and these weights can be determined through historical data analysis and expert experience. Finally, the battery performance index data is generated to reflect the overall performance of the battery under specific conditions. This index will provide important basic data for the subsequent battery aging pre-response factor analysis to ensure that the subsequent analysis can be based on accurate performance evaluation results. Through the above process, the performance evaluation is clearly and effectively combined with the battery status to form a systematic battery performance monitoring solution. First, the aging performance attenuation evaluation data obtained in step S331 and the battery performance index data generated in step S332 need to be collected. The combination of these two data sets will be used to identify the pre-response factors related to the battery aging process. The aging performance attenuation data and the battery performance index data are correlated to determine which factors have a significant impact on the battery performance attenuation. Statistical methods such as Pearson correlation coefficient or Spearman rank correlation coefficient can be used to calculate the correlation between each performance indicator and the attenuation data, and screen out factors with high correlation. Perform regression analysis on the screened factors, establish a linear or nonlinear regression model, and clarify the degree of influence of each factor on battery performance. Through the coefficients of the regression model, the contribution of each factor to the attenuation of battery performance can be quantified. Next, sort the aging pre-response factors, determine the most important influencing factors, and establish a response factor matrix to link each factor with the corresponding performance changes.Finally, the analysis results of the battery aging pre-response factor are formed to provide necessary data support for the subsequent mechanism induction. In this step, a systematic mechanism induction is performed based on the battery aging pre-response factor data generated in step S333. First, the relationship between each pre-response factor is analyzed to identify potential causal relationships. The structural equation model (SEM) can be used for analysis to understand the mutual influence between different factors and their comprehensive effect on battery performance degradation. In the process of mechanism induction, the physical and chemical reaction mechanisms in the aging process are further confirmed in combination with literature in related fields. Through literature comparison, the rationality of the aging pre-response factor is verified, and the actual analysis results are combined with theoretical research to summarize the mechanism that affects battery performance degradation. At the same time, the relationship between the factors and their overall impact on battery performance are presented in a graphical manner to form a comprehensive aging pre-response mechanism induction data. These data not only reveal the inherent mechanism of battery aging, but also provide theoretical support for the subsequent identification of critical states of performance degradation. In this step, the critical state of battery performance degradation is identified based on the aging pre-response mechanism induction data generated in step S334. First, set the normal range and critical state thresholds for battery performance. These thresholds can be defined based on historical performance data and industry standards. Use the analysis results of the aging pre-response factor and the battery performance index to determine the key parameters that affect battery performance. Next, use the cutoff value analysis method to identify samples that show obvious attenuation characteristics based on the collected battery performance data. By monitoring the performance changes of the battery under different working conditions, identify when the performance drops to a critical state, and use statistical analysis methods such as control charts or root cause analysis to conduct in-depth research on the causes of performance degradation. Finally, generate battery performance degradation critical state data, specifically describing the manifestation of the critical state and its related factors. This data provides an important basis for the formulation of battery management and maintenance strategies, so that potential performance degradation risks can be effectively predicted and intervened during battery operation. Ensure the safety and reliability of the battery during use through clear thresholds and identification methods.

[0139] Preferably, the present invention further provides a battery gradient classification and identification system, which is used to execute the battery gradient classification and identification method as described above, and the battery gradient classification and identification system comprises:

[0140] The power consumption extraction module is used to obtain the new energy vehicle system monitoring authority; according to the new energy vehicle system monitoring authority, the dynamic workload log of the new energy vehicle battery pack is collected to obtain the dynamic workload cleaning log; according to the dynamic workload cleaning log, the connection node load power consumption is extracted to obtain the connection node load power consumption data;

[0141] The output power loss quantification module is used to simulate the theoretical load temperature fluctuation of the load power consumption data of the connection node to obtain the theoretical temperature fluctuation data of the load energy consumption; analyze the environmental impact effect of the theoretical temperature fluctuation data of the load energy consumption to obtain the load environmental impact effect data; quantify the node output power loss according to the load environmental impact effect data to obtain the node output power loss data;

[0142] The performance gradient classification and identification model building module is used to perform performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; and to construct a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model;

[0143] The automated firmware design module is used to perform automated firmware design on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

[0144] Preferably, a storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the battery gradient classification and identification method described in any one of items 1 to 8 above.

[0145] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0146] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A battery gradient classification and identification method, characterized in that: The following steps are involved: Step S1: obtaining the new energy vehicle system monitoring authority; collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain dynamic workload cleaning logs; extracting the connection node load power consumption according to the dynamic workload cleaning logs to obtain the connection node load power consumption data; Step S2: performing a theoretical load temperature fluctuation simulation on the load power consumption data of the connection node to obtain theoretical load energy consumption temperature fluctuation data; performing an environmental impact effect analysis on the theoretical load energy consumption temperature fluctuation data to obtain load environmental impact effect data; Quantify the node output power loss according to the load environment impact effect data to obtain the node output power loss data; Step S3: performing performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; A battery performance gradient classification and identification model is constructed according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model; Step S4: Automated firmware design is performed on the battery performance gradient classification and identification model to obtain battery performance gradient classification and identification firmware; The battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

2. The battery gradient classification and identification method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtaining new energy vehicle system monitoring authority; Step S12: collecting dynamic workload logs of the new energy vehicle battery pack according to the new energy vehicle system monitoring authority to obtain the dynamic workload logs of the battery pack; Step S13: performing data cleaning on the dynamic workload log of the battery pack to obtain a dynamic workload cleaning log; Step S14: extracting the connection structure of the new energy vehicle battery pack to obtain the battery pack connection structure data; Step S15: extract the connection node load power consumption of the battery pack connection structure data according to the dynamic workload cleaning log to obtain the connection node load power consumption data.

3. The battery gradient classification and identification method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: collecting new energy vehicle battery environment data based on the new energy vehicle system monitoring authority to obtain battery environment data; Step S22: performing load energy consumption baseline analysis on the load power consumption data of the connection nodes to obtain a node load energy consumption baseline; Step S23: performing theoretical load temperature fluctuation simulation on the node load energy consumption baseline to obtain load energy consumption theoretical temperature fluctuation data; Step S24: performing environmental impact analysis on the load energy consumption theoretical temperature fluctuation data according to the battery environment data to obtain load environment impact effect data; Step S25: quantifying the node output power consumption loss of the connection node load power consumption data according to the load environment impact effect data to obtain the node output power consumption loss data.

4. The battery gradient classification and identification method according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: extracting environmental impact factors from battery environmental data to obtain environmental impact factors; Step S242: performing fluctuation extreme value scatter point distribution analysis on load energy consumption theoretical temperature fluctuation data to obtain theoretical fluctuation extreme value scatter point distribution data; Step S243: performing scatter point nonlinear correlation analysis on the theoretical fluctuation extreme value scatter point distribution data according to the environmental impact factor to obtain scatter point extreme value fluctuation correlation data; Step S244: Perform environmental impact analysis on the load environmental impact effect data to obtain the load environmental impact effect data.

5. The battery gradient classification and identification method according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: evaluating the impact strength between different nodes on the load power consumption data of the connected nodes according to the load environment impact effect data, and obtaining node impact strength evaluation data; Step S252: Based on the node impact strength assessment data and the node load power consumption data, node power consumption growth pattern identification is performed to obtain node power consumption growth pattern data; Step S253: performing adjacent node energy consumption growth difference identification on the node power consumption growth pattern data to obtain adjacent node energy consumption growth difference data; Step S254: identifying abnormal growth trends between different nodes of the load power consumption data of the connected nodes according to the energy consumption growth difference data of the adjacent nodes, and obtaining abnormal growth trend data of the load power consumption; Step S255: quantify the node output power consumption loss based on the load power consumption abnormal growth trend data to obtain the node output power consumption loss data.

6. The battery gradient classification and identification method according to claim 5, characterized in that: Step S254 includes the following steps: Calculate the growth rate ratio of the energy consumption growth difference data of adjacent nodes to obtain the energy consumption growth rate ratio; According to the energy consumption growth rate ratio, the energy consumption transmission resistance value is evaluated by comparing the energy consumption growth difference data of adjacent nodes to obtain the node energy consumption transmission resistance value data; According to the node energy consumption transmission resistance data and the energy consumption growth rate ratio, the dependency characteristics between adjacent nodes are analyzed by comparing the load power consumption data of the connected nodes to obtain the energy consumption dependency characteristics data of the adjacent nodes; The abnormal growth trend between different nodes is identified based on the energy consumption dependency characteristic data of neighboring nodes to obtain the abnormal growth trend data of load power consumption.

7. The battery gradient classification and identification method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: matching the battery node output with the connected node load power consumption data according to the node output power loss data to obtain node output matching data; Step S32: Perform performance output attenuation gradient analysis on the node output matching data to obtain performance output attenuation gradient data; Step S33: identifying the critical state of battery performance attenuation according to the performance output attenuation gradient data to obtain the critical state data of battery performance attenuation; Step S34: constructing a battery performance gradient classification and identification model based on the battery performance attenuation critical state data to obtain a battery performance gradient classification and identification model.

8. The battery gradient classification and identification method according to claim 7, characterized in that: Step S33 includes the following steps: Step S331: performing time-dependent performance attenuation evaluation on the performance output attenuation gradient data to obtain time-dependent performance attenuation evaluation data; Step S332: constructing a battery performance index according to the time-dependent performance attenuation evaluation data to obtain battery performance index data; Step S333: performing a battery aging pre-response factor analysis based on the time-dependent performance attenuation evaluation data and the battery performance index data to obtain a battery aging pre-response factor; Step S334: performing mechanistic induction on the battery aging pre-response factor to obtain aging pre-response mechanism induction data; Step S335: Summarize the data based on the aging pre-response mechanism to identify the critical state of battery performance degradation, and obtain the critical state data of battery performance degradation.

9. A battery gradient classification and identification system, characterized in that: For executing the battery gradient classification and identification method as claimed in claim 1, the battery gradient classification and identification system comprises: The power consumption extraction module is used to obtain the new energy vehicle system monitoring authority; according to the new energy vehicle system monitoring authority, the dynamic workload log of the new energy vehicle battery pack is collected to obtain the dynamic workload cleaning log; according to the dynamic workload cleaning log, the connection node load power consumption is extracted to obtain the connection node load power consumption data; The output power loss quantification module is used to simulate the theoretical load temperature fluctuation of the load power consumption data of the connection node to obtain the theoretical temperature fluctuation data of the load energy consumption; analyze the environmental impact effect of the theoretical temperature fluctuation data of the load energy consumption to obtain the load environmental impact effect data; quantify the node output power loss according to the load environmental impact effect data to obtain the node output power loss data; The performance gradient classification and identification model building module is used to perform performance output attenuation gradient analysis on the connected node load power consumption data according to the node output power consumption loss data to obtain performance output attenuation gradient data; and to construct a battery performance gradient classification and identification model according to the performance output attenuation gradient data to obtain a battery performance gradient classification and identification model; The automated firmware design module is used to perform automated firmware design on the battery performance gradient classification and identification model to obtain the battery performance gradient classification and identification firmware; the battery performance gradient classification and identification firmware is embedded into the new energy vehicle system to perform battery gradient classification and identification.

10. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the battery gradient classification and identification method according to any one of claims 1 to 8.

Images