Battery aging state detection system, method and device

Abstract

The invention relates to the technical field of power batteries, in particular to a battery aging state detection system, method and device. The method comprises the following steps: obtaining battery pack structure design data of the unmanned aerial vehicle; according to the battery pack structure design data, carrying out single battery health monitoring, and generating a multi-parameter feature vector; establishing a data mapping relation of the battery monomer indexes through the multi-parameter feature vectors, thereby obtaining initial health state data of the battery pack; collecting charging and discharging sample data of the unmanned aerial vehicle under different flight conditions; a dynamic current excitation sequence is constructed based on charging and discharging sample data, and current load changes in different flight stages are simulated, so that a multi-working-condition current disturbance sequence is generated. By simulating the load change of the unmanned aerial vehicle in different flight stages and combining the dynamic response capture capability of the high-frequency current sensor, the electrochemical characteristics of the battery can be extracted in real time, and the aging state of the battery under the complex working condition can be accurately identified.

Description

Technical Field

[0001] The present invention relates to the technical field of power batteries, and in particular to a battery aging status detection system, method and device. Background Art

[0002] Battery Aging State (BAS) refers to the process in which the performance of a battery gradually declines due to chemical, physical and electrical reactions during use. As time goes by, the performance of the battery in terms of charge and discharge efficiency, capacity, cycle life, etc. will be affected to varying degrees. The detection methods for battery aging state mainly include the following: Capacity test method, which is one of the most basic battery detection methods. It tests the actual capacity of the battery under extreme conditions through a complete charge and discharge cycle; monitors the changes in voltage and current, and records the discharge time to calculate the actual available capacity of the battery. Internal resistance measurement method, by applying a known current load and combining the voltage change for calculation, this method can be simply divided into DC internal resistance method and AC internal resistance method. The DC internal resistance method is suitable for short-term measurement, while the AC internal resistance method can obtain more accurate data. Data monitoring method: With the help of an intelligent battery management system, multiple parameters of the battery (such as current, voltage, temperature, etc.) are monitored in real time, and data analysis and judgment are performed in combination with the historical data of the unit. This method is suitable for long-term tracking of the health status of the battery.

[0003] However, battery aging status detection methods often have the following problems: UAV batteries are often in an environment of rapid charging and discharging, high load fluctuations, and high-frequency cycling. Traditional methods are difficult to accurately capture the battery aging characteristics under these complex working conditions, resulting in an incomplete assessment of the battery health status. UAV battery packs are composed of multiple monomers, and the aging speed of each monomer is usually inconsistent, which is prone to imbalance within the battery pack, thereby affecting the overall performance and safety of the battery pack. Traditional methods lack independent monitoring and management methods for single batteries. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a battery aging status detection system, method and device to solve at least one of the above technical problems.

[0005] To achieve the above object, a battery aging status detection method comprises the following steps:

[0006] Step S1: obtaining the battery pack structure design data of the drone; performing battery cell health monitoring according to the battery pack structure design data to generate a multi-parameter feature vector; establishing a data mapping relationship of the battery cell index through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack;

[0007] Step S2: Collecting charging and discharging sample data of different flight conditions of the UAV; constructing a dynamic current excitation sequence based on the charging and discharging sample data to simulate the current load changes in different flight stages, thereby generating a multi-condition current disturbance sequence;

[0008] Step S3: capturing the dynamic response characteristics of the single cell through a high-frequency current sensor according to the multi-operating current disturbance sequence, and performing real-time electrochemical feature extraction to obtain single cell response data; analyzing the key aging indicators of the single cell response data based on internal resistance change, capacity attenuation and temperature anomaly to generate single cell aging risk assessment data;

[0009] Step S4: dynamically adjusting the charge and discharge parameters based on the optimization of charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, thereby obtaining the battery pack balanced charge and discharge strategy data;

[0010] Step S5: Perform real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

[0011] The present invention can understand the design characteristics, load capacity and design parameters of the battery pack from the source by acquiring and analyzing the structural design data of the battery pack, which provides accurate basic data for subsequent monitoring. By generating a multi-parameter feature vector through health monitoring and establishing a data mapping relationship, the health status of the battery cell can be fully reflected and refined monitoring can be achieved. The initial health status data lays a foundation for the dynamic monitoring and evaluation of the entire battery pack, which is helpful for the accurate adjustment of aging characteristics and charge and discharge optimization in the subsequent stage. By collecting the charge and discharge sample data under different flight conditions, the complex current load change characteristics of the UAV in the real flight environment can be fully simulated. The generated multi-condition current disturbance sequence can accurately reflect the actual use of the battery pack of the UAV in various mission scenarios, which helps to simulate and analyze the dynamic response characteristics of the battery, improve the authenticity and coverage of the battery performance evaluation, and thus provide reliable data support for the comprehensive evaluation and prediction of the health status. Based on the real-time capture of the dynamic response characteristics of the single cell based on the high-frequency current sensor, fast and high-precision electrochemical feature extraction can be achieved. By analyzing key aging indicators such as internal resistance change, capacity decay and temperature anomaly, the aging risk and degradation trend of the single cell can be accurately identified. This method can not only accurately quantify the aging of single cells, but also detect battery anomalies in real time, providing a basis for timely optimization of charging and discharging strategies and battery maintenance. Combining the initial health status data and the single cell aging risk assessment data, the charging and discharging parameters can be dynamically adjusted for the imbalance problems that occur during the use of single cells. This balance optimization strategy can slow down the aging rate of single cells, improve the overall performance and service life of the battery pack, and reduce the risk of damage to the overall function of the battery pack due to single cell failure, thereby ensuring the safe operation of the drone. By integrating multi-dimensional aging feature data, a comprehensive health status assessment report can be generated in real time, which can not only fully display the current health status of the battery pack, but also provide life prediction and maintenance recommendations based on the assessment results. This report can provide operators with actionable maintenance guidance, prevent potential failure risks in advance, and optimize the operating efficiency of the battery pack, thereby reducing long-term use costs and improving equipment reliability.

[0012] The present invention also provides a battery aging state detection system, which is used to execute the above-mentioned battery aging state detection method, and the battery aging state detection system comprises:

[0013] The structural feature acquisition module is used to obtain the structural design data of the battery pack of the drone; perform battery cell health monitoring based on the battery pack structural design data to generate a multi-parameter feature vector; establish a data mapping relationship of the battery cell indicators through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack;

[0014] The operating condition feature reconstruction module is used to collect the charging and discharging sample data of different flight conditions of the UAV; based on the charging and discharging sample data, a dynamic current excitation sequence is constructed to simulate the current load changes in different flight stages, thereby generating a multi-operating condition current disturbance sequence;

[0015] The aging feature recognition module is used to capture the dynamic response characteristics of the single cell through a high-frequency current sensor according to the current disturbance sequence of multiple working conditions, and to extract the electrochemical characteristics in real time, so as to obtain the single cell response data; the key aging index analysis based on the internal resistance change, capacity attenuation and temperature anomaly of the single cell response data is performed to generate the single cell aging risk assessment data;

[0016] The performance balance control module is used to dynamically adjust the charge and discharge parameters based on the optimization of the charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, so as to obtain the battery pack balanced charge and discharge strategy data;

[0017] The health assessment and early warning module is used to conduct real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

[0018] The design and implementation of the modules in the present invention greatly improve the health management and performance optimization of the UAV battery pack. The structural feature acquisition module obtains the battery pack structural design data and generates a multi-parameter feature vector in combination with the battery cell health monitoring, thereby accurately mapping the initial health status data of the battery pack, providing a solid foundation for subsequent battery management. The operating condition feature reconstruction module collects the charging and discharging sample data of different flight conditions and constructs a dynamic current excitation sequence, which can simulate the current load changes in different flight stages and generate a multi-condition current disturbance sequence, thereby accurately reflecting the performance of the battery pack in various working environments. The aging feature identification module uses a high-frequency current sensor to capture the dynamic response characteristics of the single cell and extracts the electrochemical features, providing a scientific basis for identifying key aging indicators such as the internal resistance change, capacity attenuation and temperature anomaly of the battery, and generates detailed aging risk assessment data. The performance balance control module dynamically adjusts the charging and discharging parameters according to the initial health status data and aging assessment data of the battery pack, optimizes the balanced charging and discharging strategy of the battery pack, and improves the efficiency and stability of the battery pack. Finally, the health assessment and early warning module generates a comprehensive battery health assessment report by real-time monitoring of the health status of the battery pack, integrating multi-dimensional aging characteristics, and providing accurate data support for battery life prediction and maintenance, ensuring the long-term stable operation of the drone battery pack, extending its service life and reducing maintenance costs. This comprehensive management system effectively improves the safety, performance and reliability of the battery pack, and provides a comprehensive solution for the intelligent management of drone battery packs.

[0019] The present invention also provides a battery aging status detection device, comprising a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the above-mentioned battery aging status detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0021] Figure 1 A schematic diagram of the steps of the battery aging status detection method of the present invention;

[0022] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0023] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0024] 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.

[0025] 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.

[0026] 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.

[0027] To achieve this, please refer to Figures 1 to 3The present invention provides a method for detecting battery aging status, the method comprising the following steps:

[0028] Step S1: obtaining the battery pack structure design data of the drone; performing battery cell health monitoring according to the battery pack structure design data to generate a multi-parameter feature vector; establishing a data mapping relationship of the battery cell index through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack;

[0029] Step S2: Collecting charging and discharging sample data of different flight conditions of the UAV; constructing a dynamic current excitation sequence based on the charging and discharging sample data to simulate the current load changes in different flight stages, thereby generating a multi-condition current disturbance sequence;

[0030] Step S3: capturing the dynamic response characteristics of the single cell through a high-frequency current sensor according to the multi-operating current disturbance sequence, and performing real-time electrochemical feature extraction to obtain single cell response data; analyzing the key aging indicators of the single cell response data based on internal resistance change, capacity attenuation and temperature anomaly to generate single cell aging risk assessment data;

[0031] Step S4: dynamically adjusting the charge and discharge parameters based on the optimization of charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, thereby obtaining the battery pack balanced charge and discharge strategy data;

[0032] Step S5: Perform real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

[0033] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a process flow of a battery aging state detection method according to the present invention. In this example, the battery aging state detection method includes the following steps:

[0034] Step S1: obtaining the battery pack structure design data of the drone; performing battery cell health monitoring according to the battery pack structure design data to generate a multi-parameter feature vector; establishing a data mapping relationship of the battery cell index through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack;

[0035] The embodiment of the present invention obtains the CAD design drawings of the unmanned aerial vehicle battery pack, extracts data such as the number of battery cells, arrangement method, and series-parallel connection structure through the drawings, and generates structural design data of the battery pack; measures the length, width, thickness, and electrode area of ​​the battery cells one by one, and generates a multi-parameter feature vector in combination with the composition information of the battery material and the manufacturing process parameters (such as the electrode coating density and the rolling thickness); trains the feature vector using the support vector machine algorithm, establishes a multivariate data mapping model for the health status of the single cell battery, and calculates the overall performance indicators such as the total internal resistance and the lower limit of the capacity by combining the mapping results with the series and parallel characteristics of the battery pack, and finally obtains the initial health status data of the battery pack.

[0036] Step S2: Collecting charging and discharging sample data of different flight conditions of the UAV; constructing a dynamic current excitation sequence based on the charging and discharging sample data to simulate the current load changes in different flight stages, thereby generating a multi-condition current disturbance sequence;

[0037] The embodiment of the present invention records the current and voltage changes during take-off, cruising, hovering and landing through the UAV flight control system, and uses a high-precision multi-channel sensor to collect corresponding charge and discharge sample data; based on the sample data, a high current density short pulse model in the take-off stage, a stable load model in the cruising stage, a high-frequency alternating load model in the hovering stage and a gradual attenuation model in the landing stage are established to generate corresponding current fluctuation characteristic data; a dynamic current excitation sequence including linear growth, step change and sinusoidal wave disturbance characteristics is designed, and the sequence is discretized into current disturbance sequence data that can be repeatedly loaded.

[0038] Step S3: capturing the dynamic response characteristics of the single cell through a high-frequency current sensor according to the multi-operating current disturbance sequence, and performing real-time electrochemical feature extraction to obtain single cell response data; analyzing the key aging indicators of the single cell response data based on internal resistance change, capacity attenuation and temperature anomaly to generate single cell aging risk assessment data;

[0039] The embodiment of the present invention uses a high-frequency current sensor with a sampling frequency of 10kHz to record the response data of a single cell under the action of a dynamic current disturbance sequence, including transient voltage fluctuations, internal resistance changes, and temperature rise rates; uses a wavelet transform algorithm to remove noise and background interference, and extracts electrochemical characteristics such as the internal resistance change rate, voltage response amplitude, and temperature drift; further analyzes the internal resistance change characteristics based on the AC impedance spectrum, and uses a pulse current response method to quantify the discharge capacity attenuation and evaluate the coulombic efficiency; extracts the abnormal temperature characteristics of the single cell in combination with the thermal runaway monitoring data, integrates them into an aging feature vector, and generates aging risk assessment data.

[0040] Step S4: dynamically adjusting the charge and discharge parameters based on the optimization of charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, thereby obtaining the battery pack balanced charge and discharge strategy data;

[0041] The embodiment of the present invention performs difference analysis on the initial health status data and the single cell aging risk assessment data, calculates the deviations of the internal resistance, capacity and temperature rise between the single cells, and generates the battery pack imbalance data; designs a dynamic charge and discharge optimization strategy based on the imbalance data, such as adjusting the charging current to 5A, extending the constant current charging time by 10 minutes, and applying a pulse balancing current to the aged battery for charging; monitors the internal resistance change and temperature characteristics of the single cell during the optimization process in real time, records the performance changes before and after the optimization, and finally obtains the battery pack balanced charge and discharge strategy data.

[0042] Step S5: Perform real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

[0043] The embodiment of the present invention inputs the balanced charge and discharge strategy data into the battery pack health assessment model, compares the battery pack capacity retention rate, internal resistance change rate and temperature rise curve before and after optimization, integrates multi-dimensional aging characteristics, and generates a comprehensive health status assessment report; the report includes battery pack life prediction data (for example, a predicted remaining life of 200 cycles) and maintenance recommendations (for example, it is recommended to replace battery cells that exceed 80% of the capacity threshold); in actual application scenarios, the assessment report can be pushed in real time through the control terminal of the drone to help the operator adjust the flight mission plan to extend the battery life.

[0044] The present invention can understand the design characteristics, load capacity and design parameters of the battery pack from the source by acquiring and analyzing the structural design data of the battery pack, which provides accurate basic data for subsequent monitoring. By generating a multi-parameter feature vector through health monitoring and establishing a data mapping relationship, the health status of the battery cell can be fully reflected and refined monitoring can be achieved. The initial health status data lays a foundation for the dynamic monitoring and evaluation of the entire battery pack, which is helpful for the accurate adjustment of aging characteristics and charge and discharge optimization in the subsequent stage. By collecting the charge and discharge sample data under different flight conditions, the complex current load change characteristics of the UAV in the real flight environment can be fully simulated. The generated multi-condition current disturbance sequence can accurately reflect the actual use of the battery pack of the UAV in various mission scenarios, which helps to simulate and analyze the dynamic response characteristics of the battery, improve the authenticity and coverage of the battery performance evaluation, and thus provide reliable data support for the comprehensive evaluation and prediction of the health status. Based on the real-time capture of the dynamic response characteristics of the single cell based on the high-frequency current sensor, fast and high-precision electrochemical feature extraction can be achieved. By analyzing key aging indicators such as internal resistance change, capacity decay and temperature anomaly, the aging risk and degradation trend of the single cell can be accurately identified. This method can not only accurately quantify the aging of single cells, but also detect battery anomalies in real time, providing a basis for timely optimization of charging and discharging strategies and battery maintenance. Combining the initial health status data and the single cell aging risk assessment data, the charging and discharging parameters can be dynamically adjusted for the imbalance problems that occur during the use of single cells. This balance optimization strategy can slow down the aging rate of single cells, improve the overall performance and service life of the battery pack, and reduce the risk of damage to the overall function of the battery pack due to single cell failure, thereby ensuring the safe operation of the drone. By integrating multi-dimensional aging feature data, a comprehensive health status assessment report can be generated in real time, which can not only fully display the current health status of the battery pack, but also provide life prediction and maintenance recommendations based on the assessment results. This report can provide operators with actionable maintenance guidance, prevent potential failure risks in advance, and optimize the operating efficiency of the battery pack, thereby reducing long-term use costs and improving equipment reliability.

[0045] Preferably, step S1 comprises the following steps:

[0046] Step S11: Obtain the battery pack CAD design drawing, and extract the number of battery cells, the arrangement in the drone, and the connection type information, so as to obtain the battery pack structure design data;

[0047] Step S12: collecting geometric dimension information of length, width, thickness and electrode area of ​​the single battery in the battery pack according to the battery pack structure design data, so as to obtain battery geometric dimension data;

[0048] Step S13: obtaining material composition information of the single battery and manufacturing process parameters based on electrode coating density and pole piece rolling process parameters, thereby obtaining battery material characteristic data;

[0049] Step S14: obtaining the rated capacity, nominal voltage and internal resistance information of the single battery, thereby obtaining battery calibration data;

[0050] Step S15: combining the battery geometry data, the battery material characteristic data and the battery calibration data into a multi-parameter feature vector;

[0051] Step S16: establishing data mapping of battery cell indicators through multi-parameter feature vectors, thereby obtaining initial health status data of the battery pack.

[0052] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0053] Step S11: Obtain the battery pack CAD design drawing, and extract the number of battery cells, the arrangement in the drone, and the connection type information, so as to obtain the battery pack structure design data;

[0054] The embodiment of the present invention exports the CAD design drawings of the battery pack through the UAV design platform, uses professional CAD parsing software (such as AutoCAD or SolidWorks) to perform structured parsing on the drawings, and extracts the number, arrangement and connection type information of the battery cells. During the parsing process, the design layer where the battery pack is located is first locked, and the electrical connection information marked in the drawing is used to determine the electrical properties of each battery cell (such as the series connection or parallel connection from the positive electrode to the negative electrode). For example, it is parsed from the drawing that a certain UAV battery pack is composed of 12 single cells, of which every 6 cells are first connected in series to form two groups and then connected in parallel. The extracted data is annotated and checked to ensure that there are no omissions or errors, and finally the structural design information is organized into battery pack structural design data and output as an Excel format file for subsequent use.

[0055] Step S12: collecting geometric dimension information of length, width, thickness and electrode area of ​​the single battery in the battery pack according to the battery pack structure design data, so as to obtain battery geometric dimension data;

[0056] The embodiment of the present invention uses a high-precision measuring instrument (such as a laser displacement sensor or a CNC caliper) in a laboratory environment to measure the geometric dimensions of each battery cell, including length, width, thickness, and effective area of ​​the electrode. The specific operation is to place the battery in the measuring fixture, ensure the position consistency by fixing the fixture, and record each parameter in turn. Taking a certain model of battery as an example, its length measurement value is 100.2mm, width is 65.3mm, thickness is 18.1mm, and electrode area is 20.5cm 2 , repeat the measurement three times and take the average value as the final dimension data. All measurement data are recorded in a standardized template and corresponded with the battery pack structure design data to generate battery geometric dimension data.

[0057] Step S13: obtaining material composition information of the single battery and manufacturing process parameters based on electrode coating density and pole piece rolling process parameters, thereby obtaining battery material characteristic data;

[0058] The present invention obtains the material composition information of the single cell by contacting the battery manufacturer or querying its technical specification documents, for example, the positive electrode material is lithium iron phosphate (LiFePO4), the negative electrode material is graphite, and the electrolyte is lithium hexafluorophosphate (LiPF6) solution. Combined with the production process records, key manufacturing parameters are extracted, such as the electrode coating density is 12mg / cm 2 , the rolling thickness of the pole piece is 50μm. To verify the accuracy of the material information, the electrode surface morphology can be checked by scanning electron microscopy (SEM) or the electrolyte composition content can be confirmed by thermogravimetric analyzer (TGA). Ultimately, these material property data are uniformly stored in the battery database and associated with the geometric dimension data to ensure integrity.

[0059] Step S14: obtaining the rated capacity, nominal voltage and internal resistance information of the single battery, thereby obtaining battery calibration data;

[0060] The embodiment of the present invention uses a comprehensive battery tester (such as Neware or Arbin test system) to perform calibration experiments on single cells to obtain parameters such as rated capacity, nominal voltage and internal resistance. During the experiment, the battery is placed in a constant temperature box (the temperature is set to 25°C and the humidity is set to 40%), and the battery is fully charged using a constant current and constant voltage charging mode. Then, a discharge test is performed at a standard 0.5C rate to record the actual discharge capacity. For example, a battery is calibrated to have a rated capacity of 3000mAh, a nominal voltage of 3.7V, and an internal resistance of 20mΩ. After the experiment, the calibration data is compared with the data provided by the manufacturer. After confirming that there is no significant deviation, a calibration data file is generated to support subsequent analysis.

[0061] Step S15: combining the battery geometry data, the battery material characteristic data and the battery calibration data into a multi-parameter feature vector;

[0062] The embodiment of the present invention integrates battery geometry data, battery material characteristic data and calibration data by writing a data processing script (such as Python or MATLAB) to generate a standardized multi-parameter feature vector. For example, for a single battery, the feature vector may be represented as: {length: 100.2mm, width: 65.3mm, thickness: 18.1mm, positive electrode material: LiFePO4, coating density: 12mg / cm 2 , rated capacity: 3000mAh, internal resistance: 20mΩ}. To ensure the compatibility of data of different dimensions, a normalization method (such as Min-Max normalization) is used to scale all parameter values ​​to the range of 0-1 and store them in JSON format files for subsequent model training and analysis.

[0063] Step S16: establishing data mapping of battery cell indicators through multi-parameter feature vectors, thereby obtaining initial health status data of the battery pack.

[0064] The embodiment of the present invention is based on a machine learning regression model (such as random forest regression or support vector regression), inputs the multi-parameter feature vector data generated in step S15, and constructs a data mapping relationship of battery cell indicators through training and verification. For example, the input features included in the training set include length, width, coating density, internal resistance, etc., and the target output is a health status indicator (such as capacity decay rate or internal resistance increase rate). After the model training is completed, the feature vector is input into the model to obtain the health status score of each single battery. Combined with the series-parallel structure of the battery pack, the initial health status data of the entire battery pack is comprehensively calculated. For example, the total capacity of a battery pack is 34Ah, and the equivalent internal resistance is 15mΩ. The data is presented in a visual form (such as a line chart or a heat map), and the output is a PDF format health status report for evaluating the performance of the battery pack and subsequent maintenance.

[0065] The present invention can understand the design characteristics, load capacity and design parameters of the battery pack from the source by acquiring and analyzing the structural design data of the battery pack, which provides accurate basic data for subsequent monitoring. By generating a multi-parameter feature vector through health monitoring and establishing a data mapping relationship, the health status of the battery cell can be fully reflected and refined monitoring can be achieved. The initial health status data lays a foundation for the dynamic monitoring and evaluation of the entire battery pack, which is helpful for the accurate adjustment of aging characteristics and charge and discharge optimization in the subsequent stage. By collecting the charge and discharge sample data under different flight conditions, the complex current load change characteristics of the UAV in the real flight environment can be fully simulated. The generated multi-condition current disturbance sequence can accurately reflect the actual use of the battery pack of the UAV in various mission scenarios, which helps to simulate and analyze the dynamic response characteristics of the battery, improve the authenticity and coverage of the battery performance evaluation, and thus provide reliable data support for the comprehensive evaluation and prediction of the health status. Based on the real-time capture of the dynamic response characteristics of the single cell based on the high-frequency current sensor, fast and high-precision electrochemical feature extraction can be achieved. By analyzing key aging indicators such as internal resistance change, capacity decay and temperature anomaly, the aging risk and degradation trend of the single cell can be accurately identified. This method can not only accurately quantify the aging of single cells, but also detect battery anomalies in real time, providing a basis for timely optimization of charging and discharging strategies and battery maintenance. Combining the initial health status data and the single cell aging risk assessment data, the charging and discharging parameters can be dynamically adjusted for the imbalance problems that occur during the use of single cells. This balance optimization strategy can slow down the aging rate of single cells, improve the overall performance and service life of the battery pack, and reduce the risk of damage to the overall function of the battery pack due to single cell failure, thereby ensuring the safe operation of the drone. By integrating multi-dimensional aging feature data, a comprehensive health status assessment report can be generated in real time, which can not only fully display the current health status of the battery pack, but also provide life prediction and maintenance recommendations based on the assessment results. This report can provide operators with actionable maintenance guidance, prevent potential failure risks in advance, and optimize the operating efficiency of the battery pack, thereby reducing long-term use costs and improving equipment reliability.

[0066] Preferably, step S16 comprises the following steps:

[0067] A multivariate data mapping model of the health status of a single cell based on a support vector machine is constructed according to the multi-parameter feature vector, thereby obtaining the health status data of the single cell; the health status data of the single cell is mapped to the overall performance index of the battery pack according to the connection structure of the battery pack in the battery pack structure design data, thereby obtaining the initial health status data of the battery pack, wherein the connection structure of the battery pack includes series connection and parallel connection, and for the series structure, the total internal resistance and the lower limit of the capacity are calculated, and for the parallel structure, the current distribution and balance analysis are performed.

[0068] The embodiment of the present invention utilizes multi-parameter feature vector data (such as length, width, material properties, internal resistance, capacity, etc.) to construct a multivariate data mapping model for the health status of a single cell. Specifically, a support vector machine (SVM) algorithm is used for model training, with the actual health status data of the single cell as a label, and a multi-parameter feature vector is input as a feature. The kernel function type (such as RBF kernel) and hyperparameters (such as C value and gamma value) of the SVM model are optimized through GridSearchCV to ensure the classification accuracy and generalization ability of the model. For example, the characteristics of a single cell in the training set are {length: 100.2mm, coating density: 12mg / cm 2 , internal resistance: 20mΩ, rated capacity: 3000mAh}, the health status label is "good" or "abnormal". After the model training is completed, the test data is input for prediction to obtain the health status data of the single cell. For example, the health status of a single cell is predicted to be "capacity decay rate less than 5%". After integrating all the single cell health status data, store them in CSV format for subsequent analysis. According to the connection structure information recorded in the battery pack structure design data, the single cell health status data is mapped to the overall performance indicators of the battery pack. For the series structure, first extract the internal resistance data of the single cells connected in series and calculate the total internal resistance. For example, a battery pack containing 5 series cells has a single internal resistance of 15mΩ, 16mΩ, 15mΩ, 14mΩ and 17mΩ, and the total internal resistance is 15+16+15+14+17=77mΩ. Then calculate the capacity lower limit and select the minimum capacity value in the series battery as the overall capacity lower limit. For example, the capacities of these 5 batteries are 3000mAh, 2950mAh, 3000mAh, 2900mAh and 3000mAh, respectively, and the capacity lower limit is 2900mAh. For the parallel structure, perform current distribution analysis based on the internal resistance and health status of each battery. For example, the internal resistances of 3 parallel batteries are 20mΩ, 18mΩ and 22mΩ, respectively. Calculate the corresponding current distribution ratio through Ohm's law, and adjust and analyze the unbalanced part to confirm whether the battery with poor health status may affect the overall performance. Finally, the series-parallel results are combined to obtain the initial health status data of the entire battery pack. For example, the total capacity of the battery pack is calculated to be 45Ah and the total internal resistance is 10mΩ, and a health status assessment report is generated for system analysis.

[0069] The SVM of the present invention is good at processing multi-dimensional nonlinear data, and can efficiently extract the complex mapping relationship between multi-parameter feature vectors (such as geometric dimensions, material properties and calibration data) and health status (such as internal resistance change, capacity attenuation, temperature anomaly), thereby improving the accuracy of single cell health status assessment. By introducing kernel functions, SVM can effectively deal with the nonlinear distribution and noise interference of multi-parameter feature data, and improve the stability and versatility of the model. Through the optimal selection of support vectors, SVM can reduce the computational complexity and adapt to the application scenarios of limited samples, thereby providing a lightweight solution for the dynamic monitoring of drone batteries. The single cell health status data generated by the model covers internal resistance, capacity and other aging indicators, providing a unified and standardized data input for the subsequent calculation of the overall performance indicators of the battery pack. In the series structure, the sum of internal resistance directly affects the energy transmission efficiency and heating performance of the battery pack. By calculating the sum of internal resistance, the voltage drop and heating risk of the series battery pack can be accurately evaluated, thereby providing a key basis for thermal management design and power optimization. Accurate calculation of the sum of internal resistance helps to discover abnormal single cells (such as excessive internal resistance) and deal with them in a timely manner to prevent them from having an adverse effect on the performance of the entire battery pack. In a series structure, the overall capacity is determined by the minimum capacity of the single cell. By calculating the lower limit of capacity, the bottleneck cells that restrict the overall performance of the battery pack can be identified, thereby optimizing the configuration of the single cells and improving the balance and reliability of the battery pack. In a parallel structure, the current distribution of each single cell is directly related to the load balance and life of the battery pack. By calculating the current distribution based on the health status data, the rationality of the current distribution can be ensured to avoid overloading or over-discharging of some single cells. Accurate current distribution calculation can also optimize the charging and discharging efficiency of parallel battery packs and improve the endurance of drones. In a parallel structure, single cells with large differences in health status will cause uneven voltage, which in turn affects the stability of the charging and discharging process. Through balance analysis, inconsistent single cells can be found and targeted charge and discharge balance strategies can be formulated to improve the overall performance of the battery pack. Balance analysis can also reduce the inconsistency of single cell aging in the battery pack and extend the overall service life of the battery pack.

[0070] Preferably, step S2 comprises the following steps:

[0071] Step S21: using a high-precision multi-channel current sensor to collect charge and discharge sample data of the drone during takeoff, cruising, hovering and landing conditions during actual flight;

[0072] Step S22: Based on the charge and discharge sample data, a mathematical model reflecting the load characteristics of the battery cell under different working conditions is established, and the corresponding current fluctuation characteristics are defined respectively, so as to obtain the current fluctuation characteristic data of multiple working conditions, wherein the take-off condition corresponds to the discharge characteristics of high current density and short pulse, the cruise condition corresponds to the continuous discharge characteristics of stable medium load, the hovering condition corresponds to the high frequency alternating load and low current fluctuation characteristics, and the landing condition corresponds to the current attenuation characteristics of gradually reducing the load;

[0073] Step S23: designing a dynamic current excitation sequence covering different load amplitudes and change rates according to the current fluctuation characteristic data of multiple working conditions, thereby obtaining current excitation sequence data, wherein the dynamic current excitation sequence includes linear growth, step change and sinusoidal wave disturbance characteristics;

[0074] Step S24: Load the current excitation sequence data into the test system in segments, and use the channel high-precision current and voltage sensors to synchronously record the response signals of the battery pack, so as to generate a multi-operating condition current disturbance model;

[0075] Step S25: performing parameter discretization processing on the multi-operating-condition current disturbance model, discretizing the continuous current change into a current excitation sequence that can be repeatedly sampled, thereby obtaining a multi-operating-condition current disturbance sequence.

[0076] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0077] Step S21: using a high-precision multi-channel current sensor to collect charge and discharge sample data of the drone during takeoff, cruising, hovering and landing conditions during actual flight;

[0078] Embodiments of the present invention

[0079] Step S22: Based on the charge and discharge sample data, a mathematical model reflecting the load characteristics of the battery cell under different working conditions is established, and the corresponding current fluctuation characteristics are defined respectively, so as to obtain the current fluctuation characteristic data of multiple working conditions, wherein the take-off condition corresponds to the discharge characteristics of high current density and short pulse, the cruise condition corresponds to the continuous discharge characteristics of stable medium load, the hovering condition corresponds to the high frequency alternating load and low current fluctuation characteristics, and the landing condition corresponds to the current attenuation characteristics of gradually reducing the load;

[0080] The embodiment of the present invention uses a high-precision multi-channel current sensor (such as a Hall effect sensor with a sampling frequency of 1kHz and a measurement accuracy of ±0.1%) installed on the battery output port of the drone to record the current curve and discharge characteristics of the take-off, cruising, hovering and landing conditions during the actual flight process. In the actual flight test, the peak current of the drone during take-off reached 70A and lasted for about 2 seconds; the current was maintained between 20A±1A during the cruising stage and lasted for 10 minutes; high-frequency current fluctuations with a frequency of 50Hz and an amplitude of 15A±3A were recorded during the hovering stage; the current gradually decreased from 15A to 5A during the landing stage, and lasted for about 1 minute. The data was exported and stored in CSV format for subsequent analysis.

[0081] Step S23: designing a dynamic current excitation sequence covering different load amplitudes and change rates according to the current fluctuation characteristic data of multiple working conditions, thereby obtaining current excitation sequence data, wherein the dynamic current excitation sequence includes linear growth, step change and sinusoidal wave disturbance characteristics;

[0082] According to the collected charge and discharge sample data, the embodiment of the present invention uses Matlab to establish a mathematical model of the load characteristics of a single battery. For example, in the take-off stage, a high current density (70A) and a short pulse (2 seconds) are used as model inputs to define the high current load characteristics of the take-off condition; in the cruise stage, a medium load current (20A±1A) and a stable characteristic of a duration of 10 minutes are used; in the hovering stage, a high-frequency alternating load characteristic with a frequency of 50Hz and an amplitude of 15A±3A is used; in the landing stage, a linear model with gradual load attenuation is used to represent the current attenuation characteristics. The model parameters are fitted by the least squares method to obtain the current fluctuation characteristic data of multiple working conditions for the excitation design.

[0083] Step S24: Load the current excitation sequence data into the test system in segments, and use the channel high-precision current and voltage sensors to synchronously record the response signals of the battery pack, so as to generate a multi-operating condition current disturbance model;

[0084] The embodiment of the present invention loads the designed dynamic current excitation sequence data into the test system in segments, and synchronously records the dynamic response signal of the battery pack on the test bench through channel high-precision current and voltage sensors (such as the accuracy of the current sensor is 0.1%, and the accuracy of the voltage sensor is 0.05%). For example, when loading the takeoff condition excitation signal, it is recorded that the voltage of the single cell drops from 3.7V to 3.6V, and the response time is 1.8 seconds; when loading the hovering condition signal, it is recorded that the voltage changes at a high frequency within the range of 3.65V±0.02V. The recorded multi-condition response signals generate a current disturbance model and are saved as a database format file.

[0085] Step S25: performing parameter discretization processing on the multi-operating-condition current disturbance model, discretizing the continuous current change into a current excitation sequence that can be repeatedly sampled, thereby obtaining a multi-operating-condition current disturbance sequence.

[0086] The embodiment of the present invention discretizes the parameters of the multi-operating condition current disturbance model, and uses Python script to segment the continuous current curve into discrete repeatable sampling sequences, for example, decomposing the linear growth signal of the take-off condition into 5 incremental points (0A, 14A, 28A, 42A, 56A, 70A); discretizing the sinusoidal wave signal of the hovering condition into 10 characteristic points (such as 0A, +15A, 0A, -15A, etc.). The discretized data is saved in JSON format and imported into the simulation platform for verification, generating the final multi-operating condition current disturbance sequence for evaluating battery performance.

[0087] The load characteristics of the drone of the present invention vary in different flight stages. This step realizes high-resolution acquisition of multi-condition current data through high-precision sensors, providing real and reliable basic data for subsequent model construction. Directly collecting data during the actual flight process ensures the authenticity and representativeness of the sample, and avoids the deviation between the test data under laboratory conditions and the actual usage. It can record the charge and discharge characteristics of multiple battery cells at the same time, which is convenient for analyzing the performance differences between the cells and improving the comprehensiveness and accuracy of modeling. The current characteristics (such as high current density, pulse characteristics, stable load, etc.) are defined for take-off, cruising, hovering and landing conditions respectively, which can fully describe the current change law of each stage and provide a clear reference for the design of dynamic current excitation sequence. Through the precise fitting of the mathematical model, the nonlinear load characteristics of the battery under complex working conditions can be simulated, and the accuracy of subsequent testing and analysis can be improved. The current fluctuation characteristic data of multiple working conditions are obtained, covering indicators such as load intensity, change rate and periodicity, providing multi-dimensional support for further dynamic analysis and optimization. The designed dynamic current excitation sequence covers characteristics such as linear growth, step change and sinusoidal wave disturbance, which can simulate the actual current fluctuation of the drone under different working conditions. Through the combination of different load amplitudes and change rates, a comprehensive evaluation of the performance of the battery pack under extreme and common working conditions is achieved, and the adaptability verification capability of the battery pack is improved. The design standardization and systematization of the dynamic current excitation sequence improves the consistency and efficiency of the test and supports repeatable experimental operations. By recording the real-time response of current and voltage through high-precision sensors, the performance changes and adaptability of the battery under different loads can be accurately analyzed. The segmented loading current excitation sequence can be tested in detail for each specific load change, which helps to discover potential performance problems under specific working conditions. The synchronously recorded current and voltage response signals provide high-quality experimental data for model verification and parameter optimization. After the continuous parameters are discretized, the generated current perturbation sequence can be used as a standard test sequence to ensure the repeatability of the experiment and the comparability of the results. The discretization process reduces the complexity of the model calculation and facilitates the application of the current perturbation sequence to real-time monitoring and control systems. The discretized current perturbation sequence realizes the standardized characterization of load characteristics under different working conditions, providing a unified reference for the horizontal performance comparison of different battery packs.

[0088] Preferably, step S23 includes the following steps:

[0089] Step S231: designing a high current density, short pulse linear growth current excitation sequence according to the current fluctuation characteristic data of the take-off condition, simulating the current change characteristics of the UAV during the rapid acceleration phase, and obtaining a take-off condition current excitation subsequence;

[0090] The embodiment of the present invention designs a linear growth current excitation sequence based on the current fluctuation characteristic data of the take-off condition. Write a script in the Matlab environment, set the starting current to 0A, the final current to 70A, the time range to 2 seconds, and generate a linear growth current signal with a time resolution of 0.01 seconds. Use it as a take-off condition current excitation subsequence to simulate the current characteristic changes in the rapid acceleration stage of the drone. Load this signal in the bench test, record the voltage and current changes of the battery output, and verify its consistency with the actual flight data.

[0091] Step S232: constructing a stable medium load step change current excitation sequence according to the current fluctuation characteristic data of the cruise condition, simulating the continuous discharge characteristics of the UAV during uniform flight, and thus obtaining a cruise condition current excitation subsequence;

[0092] The embodiment of the present invention constructs a step change current excitation sequence for a stable medium load based on the current fluctuation characteristic data of the cruise condition. The step current signal is generated using the NumPy library in Python, the basic current value is set to 20A, and the step change is ±1A every 5 minutes to generate a sequence containing 6 step points. This sequence is used to simulate the continuous discharge characteristics of the drone during the uniform flight stage. In the simulation environment, the subsequence is loaded into the battery test platform, and the working parameters of the battery pack under this load are collected to ensure signal stability and load response authenticity.

[0093] Step S233: generating a sinusoidal wave disturbance current excitation sequence of a high-frequency alternating load according to the current fluctuation characteristic data of the hovering condition, simulating the small current fluctuation characteristics during the hovering of the UAV, and thus obtaining a hovering condition current excitation subsequence;

[0094] The embodiment of the present invention generates a sinusoidal wave perturbation current excitation sequence of a high-frequency alternating load based on the current fluctuation characteristic data of the hovering condition. A sinusoidal wave signal with a frequency of 50Hz and an amplitude of 15A is generated by using the sin function in Matlab, and it is superimposed on the DC baseline current of 3A to obtain a hovering condition current excitation subsequence. This subsequence is used to simulate the high-frequency small current fluctuation characteristics of the drone during hovering. In the experiment, the hovering condition signal is loaded and tested, and the transient response and thermal changes of the battery are recorded to ensure that the data is true and valid.

[0095] Step S234: designing a gradually decaying composite current excitation sequence according to the current fluctuation characteristic data of the landing condition, simulating the process of the current gradually decreasing when the UAV decelerates and descends, thereby obtaining a landing condition current excitation subsequence;

[0096] The embodiment of the present invention designs a gradually decaying composite current excitation sequence based on the current fluctuation characteristic data of the landing condition. Matlab is used to generate an exponential decay signal with an initial current of 15A and a final current of 5A, and a short pulse current (amplitude 3A, duration 0.5 seconds) is superimposed every 10 seconds. It is defined as a landing condition current excitation subsequence to simulate the gradual current reduction characteristics when the drone decelerates and descends. The sequence is loaded on the test platform, the discharge efficiency and residual capacity changes of the battery are collected, and the changes in battery characteristics during the decay process are analyzed.

[0097] Step S235: combining and standardizing the take-off condition current excitation subsequence, the cruise condition current excitation subsequence, the hover condition current excitation subsequence, and the landing condition current excitation subsequence to generate a comprehensive dynamic current excitation sequence covering different load amplitudes and change rates, thereby obtaining current excitation sequence data.

[0098] The embodiment of the present invention combines and standardizes the current excitation subsequences of the four working conditions of take-off, cruising, hovering and landing to generate a comprehensive dynamic current excitation sequence. In Python, the pandas library is used to integrate each subsequence into a set of time series data, standardize the current amplitude and time range, limit the amplitude to the range of 0-1, and unify the time step to 0.01 seconds. This comprehensive sequence is used to cover different load amplitudes and change rates to form the final current excitation sequence data, which is stored in JSON format and imported into the drone battery test environment to verify its adaptability to multi-condition load changes.

[0099] The linear growth current excitation sequence of the present invention accurately reproduces the rapidly changing current demand of the drone during take-off, and provides reliable data for analyzing the instantaneous response capability and power output performance of the battery. The sequence can test the instantaneous discharge capability and stability of the battery at high current density, and identify the performance bottleneck under extreme working conditions. It provides a targeted experimental basis for improving battery materials and improving conductivity efficiency, and optimizes the design of drone batteries to meet the needs of the take-off stage. The step-change current excitation sequence accurately simulates the current demand of the drone during the cruising stage, and can effectively test the capacity retention ability of the battery under long-term stable load. Evaluating the energy conversion efficiency and thermal management performance of the battery pack under stable power output helps to improve the endurance. Analyzing the balance of single cells under medium load, reducing the endurance problems caused by inconsistent performance, and improving the overall system reliability. The sinusoidal wave perturbation excitation sequence can test the battery's response capability to high-frequency load changes and identify the dynamic performance of the battery under low current fluctuations. By simulating high-frequency alternating loads, the electrochemical stability and charge and discharge efficiency of the battery pack under rapidly changing conditions are tested, providing a basis for improving frequency response characteristics. It provides support for precise control of the drone during the hovering stage, ensuring that the power system runs smoothly under high-frequency fluctuating loads. The composite excitation sequence reproduces the entire process of the current gradually decaying from medium load to low load, and tests the transition performance of the battery under dynamic load reduction conditions. Evaluate the output stability and remaining capacity estimation ability of the battery under gradually reducing load conditions, and reduce the uncertainty in the landing phase. Provide key data for analyzing the remaining battery life and its self-discharge characteristics under light load conditions, so as to facilitate the formulation of a reasonable battery replacement strategy. The comprehensive dynamic current excitation sequence covers the typical operating conditions of the entire UAV flight process, and can comprehensively evaluate the dynamic performance and comprehensive adaptability of the battery. The standardized sequence has a unified amplitude and change rate reference, which is convenient for horizontal comparison of the performance of different battery packs and improves the credibility of the experimental results. Provide a flexible and repeatable test framework to facilitate the adjustment of the excitation sequence according to specific UAV mission requirements and optimize the battery pack design and management strategy.

[0100] Preferably, step S3 comprises the following steps:

[0101] Step S31: Based on the multi-operating current disturbance, the dynamic response signal of the single cell is collected in real time by using a high-frequency current sensor to obtain the original response data of the single cell, wherein the sampling frequency of the high-frequency current sensor is not less than 10kHz, and the dynamic response signal includes transient voltage fluctuation, internal resistance change and temperature rise rate.

[0102] The embodiment of the present invention is based on multi-operating current disturbances, and uses a high-frequency current sensor to collect the dynamic response signal of the single cell in real time. A high-precision current sensor is used, and its sampling frequency is 20kHz to ensure that high-frequency transient changes can be captured. The sensor is connected to the positive and negative terminals of the battery cell, and the collected dynamic response signals include transient voltage fluctuations, internal resistance changes, and temperature rise rates. To ensure the accuracy of the data, a constant current source is used during the test to apply current loads for different operating conditions such as takeoff, cruising, hovering, and landing, and the voltage and temperature changes of the battery are recorded in real time. At the same time, the internal resistance change is calculated by the instantaneous changes in the battery terminal voltage and current, and the data is stored in a high-precision data recorder for further processing.

[0103] Step S32: using a wavelet transform method to remove high-frequency noise and background interference from the original response data, thereby obtaining pre-processed response data;

[0104] The embodiment of the present invention utilizes the wavelet transform method to remove high-frequency noise and background interference from the original response data. First, the collected battery voltage and current data are decomposed using discrete wavelet transform (DWT), a suitable wavelet basis (such as Haar wavelet or Daubechies wavelet) is selected, and the number of decomposition layers is set to 4. Threshold processing is performed on the detail coefficients after wavelet transform to remove high-frequency noise and background interference and retain the low-frequency part of the signal. The processed data is then reconstructed through inverse wavelet transform (IDWT) to obtain preprocessed response data. This method effectively reduces the high-frequency noise in the signal and provides a clear signal basis for subsequent feature extraction.

[0105] Step S33: extracting electrochemical characteristics of the pre-processed response data based on the internal resistance change rate, voltage response amplitude and temperature drift, thereby obtaining single cell response data;

[0106] The embodiment of the present invention extracts electrochemical features based on the internal resistance change rate, voltage response amplitude and temperature drift. First, the battery internal resistance change rate in the preprocessed response data is calculated, and the rate is represented by the ratio of the voltage response to the current change rate. Then, the voltage response amplitude is extracted, that is, the voltage fluctuation amplitude of the battery during the charge and discharge process. Finally, by analyzing the battery temperature response curve, the temperature drift is extracted. The temperature drift is defined as the difference between the maximum value and the initial value of the battery temperature during the test. These features are quantified into a set of electrochemical characteristic data to characterize the electrochemical characteristics of the single cell under different load conditions, providing data support for subsequent aging evaluation.

[0107] Step S34: quantifying the law of the change of the internal resistance of the single cell with the working condition through AC impedance spectrum analysis and pulse current response method according to the single cell response data, so as to obtain the internal resistance change characteristic data;

[0108] The embodiment of the present invention quantifies the law of changes in the internal resistance of a single cell with operating conditions through AC impedance spectrum analysis and pulse current response methods based on the response data of the single cell. Use a frequency response analyzer to measure the AC impedance spectrum of the battery, apply a small amplitude AC signal at different frequencies (frequency range is 1Hz to 100kHz), record the voltage response of the battery, and calculate the AC impedance of the battery. Further combined with the pulse current response method, by applying a short-time pulse current (such as 5A per second) and recording the battery voltage change, the internal resistance response under the pulse current is obtained. By comparing the internal resistance change characteristics under different working conditions, the internal resistance change characteristic data is plotted to reflect the aging trend of the battery under different flight conditions.

[0109] Step S35: analyzing the capacity decay characteristics of the single cell according to the single cell response data, and comparing the discharge capacity and coulomb efficiency under different working conditions, thereby establishing a quantitative evaluation model for capacity degradation and obtaining capacity decay index data;

[0110] The embodiment of the present invention performs capacity decay characteristic analysis based on the single cell response data. Using a battery pack test system, multiple charge and discharge cycles are performed under different operating conditions (for example, 10 charge and discharge cycles under takeoff conditions), and the battery discharge capacity and coulomb efficiency of each charge and discharge are recorded. By comparing the discharge capacity and coulomb efficiency under different operating conditions, the capacity decay characteristics of the battery are evaluated. Based on these data, a regression analysis method (such as linear regression or polynomial regression) is used to establish a quantitative evaluation model for capacity decay, and capacity decay index data is obtained. This model can effectively predict the life decay of the battery under different usage conditions.

[0111] Step S36: monitoring the abnormal temperature characteristics of the single cell according to the single cell response data, and analyzing the temperature distribution, heating rate, and thermal runaway risk, thereby obtaining temperature abnormality assessment data;

[0112] The embodiment of the present invention monitors temperature anomaly characteristics based on the response data of the single cell. By recording the temperature change rate of the battery in real time, the temperature distribution of the battery under different loads is analyzed. An infrared thermal imager or a high-precision temperature sensor is used to monitor the surface temperature of the battery to obtain the heating rate of the battery under various working conditions. By comparing with the standard temperature rise curve, temperature anomalies are identified and their impact on battery performance is analyzed. Combined with the overall heat dissipation of the battery pack, temperature anomaly assessment data is generated, including the temperature rise curve and the prediction of thermal runaway risk, to provide a basis for subsequent battery life prediction and risk assessment.

[0113] Step S37: extracting aging characteristics from the internal resistance change characteristic data, the capacity attenuation index data, and the temperature anomaly assessment data, thereby obtaining an aging characteristic vector;

[0114] The embodiment of the present invention combines the internal resistance change characteristic data, capacity decay index data and temperature anomaly assessment data to extract aging characteristics. The internal resistance change, capacity decay and temperature anomaly are used as input features, and the principal component analysis (PCA) or multidimensional normalization method is used to reduce the dimension of these characteristic data to extract the key aging feature vector. Through this method, the aging behavior of the battery under different working conditions can be effectively integrated to obtain representative aging characteristic data. This aging feature vector is used for subsequent risk assessment and life prediction.

[0115] Step S38: Draw the dynamic response curve of the single cell according to the aging characteristic vector, and perform health status risk grading to obtain single cell aging risk assessment data, wherein the dynamic response curve includes the change of voltage response amplitude over time, the trend of internal resistance change rate, and the temperature rise curve.

[0116] The embodiment of the present invention draws the dynamic response curve of the single cell battery according to the aging characteristic vector, and performs health status risk classification. First, the dynamic response curve of the battery is drawn by comparing the voltage response amplitude, internal resistance change rate and temperature rise curve of the battery under different working conditions. Then, the health status risk classification is performed based on these curves, and the health status of the battery is divided into different risk levels (such as low risk, medium risk and high risk) by using machine learning methods such as fuzzy logic algorithm or support vector machine (SVM). According to the risk classification, the aging risk assessment data of the single cell battery is generated, and decision support is provided for battery maintenance and replacement.

[0117] The present invention realizes a comprehensive evaluation of battery performance and accurate monitoring of health status by constructing a complete single cell dynamic response and aging feature analysis process. The high-frequency current sensor collects dynamic response signals and can capture key information such as transient voltage fluctuations, internal resistance changes and temperature rise rate, laying the foundation for subsequent analysis. The high-frequency noise and background interference are removed by wavelet transform, which improves the purity of the data and the accuracy of the analysis. Electrochemical feature extraction can accurately characterize the response characteristics of the battery under different working conditions and provide support for further quantitative analysis. The AC impedance spectrum analysis and pulse current response method reveal the law of internal resistance changes with working conditions, which is helpful to evaluate the conductivity and electrochemical stability of the battery. The capacity decay feature analysis quantifies the capacity degradation trend of the battery under different working conditions, constructs a quantitative evaluation model for the comparison of coulomb efficiency and discharge capacity, and provides a basis for life prediction. Temperature anomaly monitoring enhances the ability to identify battery thermal management and safety hazards by analyzing temperature distribution and thermal runaway risks. The extraction of aging feature vectors integrates the characteristics of internal resistance change, capacity decay and temperature anomaly to form a comprehensive aging feature characterization system. Based on this, dynamic response curves are drawn and health status risk classification is performed, which intuitively shows the aging trend and health risk of batteries under different working conditions, providing important support for long-term reliability analysis and optimized design of batteries. This method not only comprehensively covers the electrochemical, thermal and dynamic performance evaluation of single cells, but also realizes quantitative risk assessment of health status through systematic modeling, significantly improving the intelligent level of battery management and safety assurance capabilities.

[0118] Preferably, step S33 includes the following steps:

[0119] Step S331: Calculate the instantaneous internal resistance based on the dynamic current and voltage response of the preprocessed response data, and analyze its change trend under different working conditions to obtain the internal resistance change rate data;

[0120] The embodiment of the present invention calculates the instantaneous internal resistance based on the dynamic current and voltage response. The current and voltage data of the battery are collected in real time, and the internal resistance value at each moment is calculated using Ohm's law. In the specific operation, the pre-processed response data is first segmented, and a suitable time window is selected (for example, a time period every 500ms), and the current value and the corresponding voltage response are recorded in each time period. Then, the internal resistance is calculated using the instantaneous values ​​of voltage and current: R(t) = V(t) / I(t), where V(t) is the voltage value and I(t) is the current value. Through continuous internal resistance calculation, the trend of internal resistance change over time can be obtained. Further, the internal resistance changes under different working conditions are analyzed, the internal resistance fluctuations under high current load and the stability changes under low current conditions are identified, and finally the internal resistance change rate data is obtained as an important evaluation indicator of the battery health status.

[0121] Step S332: recording the maximum voltage offset of the preprocessed response data under different current loading conditions, and extracting the voltage response amplitude characteristics, so as to obtain voltage response amplitude data reflecting the charge and discharge capacity of the battery;

[0122] The embodiment of the present invention records the maximum voltage offset of the preprocessed response data under different current loading conditions, and extracts the voltage response amplitude characteristics. First, according to different charging and discharging conditions (for example, take-off conditions, cruising conditions, etc.), different current load amplitudes (for example, 0.5A, 2A, 5A, etc.) are set, and current loads of different amplitudes are applied under each condition. Then, the voltage response of the battery is recorded, and the maximum offset of the battery voltage after each current loading is calculated to obtain the amplitude of the voltage change. By comparing the voltage response amplitudes under different conditions, the charge and discharge capacity characteristics of the battery can be extracted, that is, the voltage fluctuation range that the battery can withstand under different load conditions. This data provides a basis for battery performance evaluation and can reflect the charge and discharge capabilities of the battery under different working conditions.

[0123] Step S333: Calculating the temperature rise and cooling rate under different working conditions according to the preprocessed response data, thereby obtaining temperature drift data;

[0124] The embodiment of the present invention calculates the temperature rise and cooling rate under different working conditions based on the preprocessed response data. First, a high-precision temperature sensor is used to monitor the temperature change of the battery in real time, and the temperature data is recorded when different current loads are applied (for example, the time interval for monitoring the temperature is 1 second). By analyzing the temperature rise and cooling process of the battery during the charging and discharging process, the temperature rise rate (such as how many degrees the temperature rises per minute) and the cooling rate (the rate at which the temperature drops) are calculated. For example, under takeoff conditions, due to the large load, the battery temperature may rise rapidly, while under cruising conditions, the temperature rise is relatively stable. Under landing conditions, the cooling process of the battery is more significant. By calculating the temperature rise and cooling rates under different working conditions, the temperature drift data is obtained, which reflects the thermal management performance and thermal stability of the battery.

[0125] Step S334: establishing a characteristic correlation mapping according to the internal resistance change rate data, the voltage response amplitude data and the temperature drift data, thereby obtaining comprehensive electrochemical characteristic data;

[0126] The embodiment of the present invention establishes a feature association mapping based on the internal resistance change rate data, the voltage response amplitude data and the temperature drift data. Using the multivariate linear regression or support vector machine (SVM) method, the internal resistance change rate, the voltage response amplitude and the temperature drift are used as input features to establish a mapping model. Through model training, the correlation between these three electrochemical characteristics is obtained. For example, the following model can be established by the linear regression method: internal resistance change rate = α1 × voltage response amplitude + α2 × temperature drift + β, where α1 and α2 are the weight coefficients of the model, and β is the deviation term. The training data is used to fit the model to obtain comprehensive electrochemical characteristic data. These feature association maps can help to gain a deep understanding of the electrochemical performance of the battery and provide a comprehensive basis for battery health assessment.

[0127] Step S335: Standardize and dimensionlessly transform the comprehensive electrochemical characteristic data to generate comparable electrochemical characteristic vectors, thereby obtaining single cell response data.

[0128] The embodiment of the present invention performs standardization processing and dimensionless conversion on the comprehensive electrochemical characteristic data. First, the Z-score standardization method is used to standardize characteristic data such as the internal resistance change rate, voltage response amplitude, and temperature drift, and obtain standardized data with a mean of 0 and a variance of 1. Then, in order to eliminate the influence of the differences in different characteristic units, the standardized data is dimensionally converted, for example, the Min-Max normalization method is used to convert the characteristic data into a numerical value in the range of [0,1]. This operation ensures the consistency of different characteristic data in terms of dimensions, so that each characteristic is comparable in subsequent analysis. Through this processing, a comparable electrochemical characteristic vector is generated, which serves as an important basis for the performance evaluation of single cells and provides accurate data support for subsequent health assessments and aging predictions.

[0129] The present invention realizes the accurate extraction and comprehensive quantification of the electrochemical characteristics of the single cell by hierarchical analysis and processing of pre-processed response data, providing a comprehensive and efficient technical means for battery performance evaluation. The instantaneous internal resistance calculation and change trend analysis can dynamically capture the characteristics of the internal resistance changing with the working conditions, and provide key data for evaluating the battery's conductivity and decay characteristics. The maximum voltage offset recording and voltage response amplitude extraction effectively characterize the battery's charging and discharging capabilities under different current loading conditions, reflecting the battery's power output performance and capacity utilization efficiency. The temperature rise and cooling rate calculations reveal the thermal behavior of the battery under dynamic conditions, providing basic support for evaluating thermal management efficiency and thermal runaway risks. Through the correlation mapping between the internal resistance change rate, the voltage response amplitude and the temperature drift, comprehensive electrochemical characteristic data is formed, which can reveal the intrinsic relationship between the multi-parameter characteristics of the battery and improve the systematicness and integrity of feature extraction. Finally, through the standardization and dimensionless processing of the comprehensive electrochemical characteristic data, a consistent and comparable electrochemical characteristic vector is generated, which lays a unified quantitative foundation for the performance comparison and health status evaluation of different batteries under multiple working conditions. This process not only achieves high-precision extraction and modeling of battery electrochemical characteristics, but also improves the reliability and applicability of the evaluation results through multi-dimensional feature fusion, providing strong technical support for the intelligent analysis of battery management systems.

[0130] Preferably, step S4 comprises the following steps:

[0131] Step S41: Analyze the performance difference and charge-discharge inconsistency between single cells on the initial health status data and single cell aging risk assessment data, so as to obtain battery pack imbalance data;

[0132] The embodiment of the present invention analyzes the performance differences and charge and discharge inconsistency between single cells based on the initial health status data and the single cell aging risk assessment data. First, by collecting the initial health status data (such as internal resistance, voltage response amplitude, temperature drift, etc.) and aging risk assessment data of each single cell in the battery pack, the performance differences of each single cell under different charge and discharge conditions are analyzed. In specific operations, the single cells in the battery pack can be classified by cluster analysis or K-means algorithm to identify the performance differences between the batteries. Secondly, the charge and discharge data are used to analyze the charge and discharge inconsistency of the battery pack to quantify the differences in charging current, discharging current and voltage response between the batteries. Through these data, the imbalance data of the battery pack is finally obtained, which provides a basis for subsequent optimization of charging and discharging parameters and adjustment of battery pack balance.

[0133] Step S42: designing a dynamic charging and discharging parameter adjustment strategy according to the battery pack imbalance data, thereby obtaining a charging and discharging parameter optimization scheme, wherein the charging and discharging parameters include a balanced charging current, a constant current charging time, and a balanced charging pulse parameter;

[0134] The embodiment of the present invention designs a dynamic charging and discharging parameter adjustment strategy based on the imbalance data of the battery pack. First, based on the imbalance data obtained in step S41, identify which single cells have large health differences or charging and discharging inconsistencies, and classify these batteries as priority optimization objects. Then, according to the performance characteristics of different single cells such as voltage, internal resistance, temperature rise, etc., design an adjustment scheme for charging current, constant current charging time and balanced charging pulse. Specifically, for single cells with poor health, their charging current can be reduced and the constant current charging time can be extended to reduce their charging pressure; and for single cells with better performance, the charging current can be appropriately increased to improve the charging efficiency. In addition, by setting the balanced charging pulse parameters, such as pulse frequency and pulse width, it is ensured that each single cell in the battery pack can reach a balanced state during the charging process. The final charging and discharging parameter optimization scheme can be dynamically adjusted in the battery pack management system.

[0135] Step S43: performing variable pulse balanced charging according to the charging and discharging parameter optimization scheme, and performing differentiated charging and discharging processing on single cells with different aging degrees, so as to obtain charging and discharging balance optimization data;

[0136] The embodiment of the present invention performs variable pulse balanced charging according to the charging and discharging parameter optimization scheme, and performs differentiated charging and discharging treatments on single cells with different degrees of aging. First, the charging and discharging management system is used to charge the battery pack according to the charging and discharging optimization scheme designed in step S42. During the charging process, different charging strategies are adopted for single cells with different degrees of aging. For single cells with more serious aging, a lower current is used for charging, and the constant current charging time is extended; while for batteries with less aging, a higher current is used for charging, and the constant current charging time is shortened. At the same time, the battery is balanced by a balanced charging pulse technology (such as a pulse charging method). Through this series of differentiated charging and discharging treatments, it is ensured that each single cell in the battery pack can be in the best working state, thereby achieving the overall performance optimization of the battery pack. The charge and discharge balance optimization data finally obtained will be reflected in the performance parameters such as the voltage and internal resistance of each single cell.

[0137] Step S44: monitoring the real-time response characteristics of the single cells during the optimization process according to the charge-discharge balance optimization data, and recording the performance changes before and after the charge-discharge parameters are adjusted, thereby obtaining balance adjustment effect data;

[0138] The embodiment of the present invention monitors the real-time response characteristics of the single cell during the optimization process based on the charge and discharge balance optimization data, and records the performance changes before and after the charge and discharge parameters are adjusted. In specific operations, high-precision sensors (such as voltage, current and temperature sensors) are used to monitor the voltage, internal resistance, temperature change and other response characteristics of the single cell in real time. During the charge and discharge process, these data are collected regularly and compared with the data before the charge and discharge parameters are adjusted. By comparing the internal resistance, voltage response amplitude, temperature rise and other characteristics before and after the adjustment, the effect of charge and discharge balance optimization can be intuitively evaluated. If the battery exhibits a lower internal resistance and a more stable voltage response during the charging process, it means that the charge balance effect is better; if the temperature rise rate is low, it means that the thermal management of the battery has been optimized. The final balance adjustment effect data provides a basis for the next step of generating a charge and discharge strategy.

[0139] Step S45: generating battery pack balanced charge and discharge strategy data based on the balance adjustment effect data.

[0140] The embodiment of the present invention generates balanced charge and discharge strategy data for the battery pack based on the balance adjustment effect data. According to the balance adjustment effect data obtained in step S44, a charge and discharge strategy for the battery pack is generated by a data analysis method (such as regression analysis or a neural network algorithm). These strategies include charging current, constant current charging time, balanced charging pulse parameters, etc. According to the health status, internal resistance, temperature rise and other data of different single cells in the battery pack, the system can dynamically adjust the charge and discharge strategy so that each battery can obtain a charging solution suitable for its health status. Ultimately, through the optimization of the charge and discharge strategy, the overall balance of the battery pack is improved, and the service life of the battery pack is extended.

[0141] The present invention further optimizes the charging and discharging process of the battery pack by accurately analyzing the performance difference and charging and discharging inconsistency of the battery pack, and improves the balance and overall performance of the battery pack. By analyzing the initial health status data and the single cell aging risk assessment data, the performance difference and charging and discharging imbalance problem in the battery pack are identified, which provides an important basis for subsequent balanced adjustment. According to the battery pack imbalance data, a dynamic charging and discharging parameter adjustment strategy is designed, which can effectively adjust the charging current, constant current charging time and balanced charging pulse parameters to ensure that the charging and discharging process of each single cell under different working conditions tends to be consistent. Through variable pulse balanced charging, differentiated processing is achieved between single cells with different aging degrees, thereby optimizing the balance of charging and discharging, and maximizing the service life and performance of the battery. Real-time monitoring of the battery response characteristics during the equalization process provides an intuitive basis for the performance changes before and after the adjustment of the charging and discharging parameters, so that the charging and discharging strategy can be accurately adjusted and the optimization effect is ensured. Finally, based on the balance adjustment effect data, the balanced charging and discharging strategy data for the battery pack is generated, which further improves the overall balance and system efficiency of the battery pack. This process not only optimizes the charge and discharge management of single cells, but also improves the stability and life of the battery pack under complex working conditions, providing a more refined and intelligent adjustment strategy for the battery management system.

[0142] The present invention also provides a battery aging state detection system, which is used to execute the above-mentioned battery aging state detection method, and the battery aging state detection system comprises:

[0143] The structural feature acquisition module is used to obtain the structural design data of the battery pack of the drone; perform battery cell health monitoring based on the battery pack structural design data to generate a multi-parameter feature vector; establish a data mapping relationship of the battery cell indicators through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack;

[0144] The operating condition feature reconstruction module is used to collect the charging and discharging sample data of different flight conditions of the UAV; based on the charging and discharging sample data, a dynamic current excitation sequence is constructed to simulate the current load changes in different flight stages, thereby generating a multi-operating condition current disturbance sequence;

[0145] The aging feature recognition module is used to capture the dynamic response characteristics of the single cell through a high-frequency current sensor according to the current disturbance sequence of multiple working conditions, and to extract the electrochemical characteristics in real time, so as to obtain the single cell response data; the key aging index analysis based on the internal resistance change, capacity attenuation and temperature anomaly of the single cell response data is performed to generate the single cell aging risk assessment data;

[0146] The performance balance control module is used to dynamically adjust the charge and discharge parameters based on the optimization of the charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, so as to obtain the battery pack balanced charge and discharge strategy data;

[0147] The health assessment and early warning module is used to conduct real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

[0148] The design and implementation of the modules in the present invention greatly improve the health management and performance optimization of the UAV battery pack. The structural feature acquisition module obtains the battery pack structural design data and generates a multi-parameter feature vector in combination with the battery cell health monitoring, thereby accurately mapping the initial health status data of the battery pack, providing a solid foundation for subsequent battery management. The operating condition feature reconstruction module collects the charging and discharging sample data of different flight conditions and constructs a dynamic current excitation sequence, which can simulate the current load changes in different flight stages and generate a multi-condition current disturbance sequence, thereby accurately reflecting the performance of the battery pack in various working environments. The aging feature identification module uses a high-frequency current sensor to capture the dynamic response characteristics of the single cell and extracts the electrochemical features, providing a scientific basis for identifying key aging indicators such as the internal resistance change, capacity attenuation and temperature anomaly of the battery, and generates detailed aging risk assessment data. The performance balance control module dynamically adjusts the charging and discharging parameters according to the initial health status data and aging assessment data of the battery pack, optimizes the balanced charging and discharging strategy of the battery pack, and improves the efficiency and stability of the battery pack. Finally, the health assessment and early warning module generates a comprehensive battery health assessment report by real-time monitoring of the health status of the battery pack, integrating multi-dimensional aging characteristics, and providing accurate data support for battery life prediction and maintenance, ensuring the long-term stable operation of the drone battery pack, extending its service life and reducing maintenance costs. This comprehensive management system effectively improves the safety, performance and reliability of the battery pack, and provides a comprehensive solution for the intelligent management of drone battery packs.

[0149] The present invention also provides a battery aging status detection device, comprising a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the above-mentioned battery aging status detection method.

[0150] 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 intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0151] 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 aging status detection method, characterized in that: The following steps are involved: Step S1: obtaining the battery pack structure design data of the drone; performing battery cell health monitoring according to the battery pack structure design data to generate a multi-parameter feature vector; establishing a data mapping relationship of the battery cell index through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack; Step S2: Collecting charging and discharging sample data of different flight conditions of the UAV; constructing a dynamic current excitation sequence based on the charging and discharging sample data to simulate the current load changes in different flight stages, thereby generating a multi-condition current disturbance sequence; Step S3: capturing the dynamic response characteristics of the single cell through a high-frequency current sensor according to the multi-operating current disturbance sequence, and performing real-time electrochemical feature extraction to obtain single cell response data; analyzing the key aging indicators of the single cell response data based on internal resistance change, capacity attenuation and temperature anomaly to generate single cell aging risk assessment data; Step S4: dynamically adjusting the charge and discharge parameters based on the optimization of charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, thereby obtaining the battery pack balanced charge and discharge strategy data; Step S5: Perform real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

2. The battery aging status detection method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtain the battery pack CAD design drawing, and extract the number of battery cells, the arrangement in the drone, and the connection type information, so as to obtain the battery pack structure design data; Step S12: collecting geometric dimension information of length, width, thickness and electrode area of ​​the single battery in the battery pack according to the battery pack structure design data, so as to obtain battery geometric dimension data; Step S13: obtaining material composition information of the single battery and manufacturing process parameters based on electrode coating density and pole piece rolling process parameters, thereby obtaining battery material characteristic data; Step S14: obtaining the rated capacity, nominal voltage and internal resistance information of the single battery, thereby obtaining battery calibration data; Step S15: combining the battery geometry data, the battery material characteristic data and the battery calibration data into a multi-parameter feature vector; Step S16: establishing data mapping of battery cell indicators through multi-parameter feature vectors, thereby obtaining initial health status data of the battery pack.

3. The battery aging status detection method according to claim 2, characterized in that: Step S16 includes the following steps: A multivariate data mapping model of the health status of a single cell based on a support vector machine is constructed according to the multi-parameter feature vector, thereby obtaining the health status data of the single cell; the health status data of the single cell is mapped to the overall performance index of the battery pack according to the connection structure of the battery pack in the battery pack structure design data, thereby obtaining the initial health status data of the battery pack, wherein the connection structure of the battery pack includes series connection and parallel connection, and for the series structure, the total internal resistance and the lower limit of the capacity are calculated, and for the parallel structure, the current distribution and balance analysis are performed.

4. The battery aging status detection method according to claim 3, characterized in that: Step S2 includes the following steps: Step S21: using a high-precision multi-channel current sensor to collect charge and discharge sample data of the drone during takeoff, cruising, hovering and landing conditions during actual flight; Step S22: Based on the charge and discharge sample data, a mathematical model reflecting the load characteristics of the battery cell under different working conditions is established, and the corresponding current fluctuation characteristics are defined respectively, so as to obtain the current fluctuation characteristic data of multiple working conditions, wherein the take-off condition corresponds to the discharge characteristics of high current density and short pulse, the cruise condition corresponds to the continuous discharge characteristics of stable medium load, the hovering condition corresponds to the high frequency alternating load and low current fluctuation characteristics, and the landing condition corresponds to the current attenuation characteristics of gradually reducing the load; Step S23: designing a dynamic current excitation sequence covering different load amplitudes and change rates according to the current fluctuation characteristic data of multiple working conditions, thereby obtaining current excitation sequence data, wherein the dynamic current excitation sequence includes linear growth, step change and sinusoidal wave disturbance characteristics; Step S24: Load the current excitation sequence data into the test system in segments, and use the channel high-precision current and voltage sensors to synchronously record the response signals of the battery pack, so as to generate a multi-operating condition current disturbance model; Step S25: performing parameter discretization processing on the multi-operating-condition current disturbance model, discretizing the continuous current change into a current excitation sequence that can be repeatedly sampled, thereby obtaining a multi-operating-condition current disturbance sequence.

5. The battery aging status detection method according to claim 4, characterized in that: Step S23 includes the following steps: Step S231: designing a high current density, short pulse linear growth current excitation sequence according to the current fluctuation characteristic data of the take-off condition, simulating the current change characteristics of the UAV during the rapid acceleration phase, and obtaining a take-off condition current excitation subsequence; Step S232: constructing a stable medium load step change current excitation sequence according to the current fluctuation characteristic data of the cruise condition, simulating the continuous discharge characteristics of the UAV during uniform flight, and thus obtaining a cruise condition current excitation subsequence; Step S233: generating a sinusoidal wave disturbance current excitation sequence of a high-frequency alternating load according to the current fluctuation characteristic data of the hovering condition, simulating the small current fluctuation characteristics during the hovering of the UAV, and thus obtaining a hovering condition current excitation subsequence; Step S234: designing a gradually decaying composite current excitation sequence according to the current fluctuation characteristic data of the landing condition, simulating the process of the current gradually decreasing when the UAV decelerates and descends, thereby obtaining a landing condition current excitation subsequence; Step S235: combining and standardizing the take-off condition current excitation subsequence, the cruise condition current excitation subsequence, the hover condition current excitation subsequence, and the landing condition current excitation subsequence to generate a comprehensive dynamic current excitation sequence covering different load amplitudes and change rates, thereby obtaining current excitation sequence data.

6. The battery aging status detection method according to claim 5, characterized in that: Step S3 includes the following steps: Step S31: Based on the multi-operating current disturbance, the dynamic response signal of the single cell is collected in real time by using a high-frequency current sensor to obtain the original response data of the single cell, wherein the sampling frequency of the high-frequency current sensor is not less than 10kHz, and the dynamic response signal includes transient voltage fluctuation, internal resistance change and temperature rise rate. Step S32: using a wavelet transform method to remove high-frequency noise and background interference from the original response data, thereby obtaining pre-processed response data; Step S33: extracting electrochemical characteristics of the pre-processed response data based on the internal resistance change rate, voltage response amplitude and temperature drift, thereby obtaining single cell response data; Step S34: quantifying the law of the change of the internal resistance of the single cell with the working condition through AC impedance spectrum analysis and pulse current response method according to the single cell response data, so as to obtain the internal resistance change characteristic data; Step S35: analyzing the capacity decay characteristics of the single cell according to the single cell response data, and comparing the discharge capacity and coulomb efficiency under different working conditions, thereby establishing a quantitative evaluation model for capacity degradation and obtaining capacity decay index data; Step S36: monitoring the abnormal temperature characteristics of the single cell according to the single cell response data, and analyzing the temperature distribution, heating rate, and thermal runaway risk, thereby obtaining temperature abnormality assessment data; Step S37: extracting aging characteristics from the internal resistance change characteristic data, the capacity attenuation index data, and the temperature anomaly assessment data, thereby obtaining an aging characteristic vector; Step S38: Draw the dynamic response curve of the single cell according to the aging characteristic vector, and perform health status risk grading to obtain single cell aging risk assessment data, wherein the dynamic response curve includes the change of voltage response amplitude over time, the trend of internal resistance change rate, and the temperature rise curve.

7. The battery aging status detection method according to claim 6, characterized in that: Step S33 includes the following steps: Step S331: Calculate the instantaneous internal resistance based on the dynamic current and voltage response of the preprocessed response data, and analyze its change trend under different working conditions to obtain the internal resistance change rate data; Step S332: recording the maximum voltage offset of the preprocessed response data under different current loading conditions, and extracting the voltage response amplitude characteristics, so as to obtain voltage response amplitude data reflecting the charge and discharge capacity of the battery; Step S333: Calculating the temperature rise and cooling rate under different working conditions according to the preprocessed response data, thereby obtaining temperature drift data; Step S334: establishing a characteristic correlation mapping according to the internal resistance change rate data, the voltage response amplitude data and the temperature drift data, thereby obtaining comprehensive electrochemical characteristic data; Step S335: Standardize and dimensionlessly transform the comprehensive electrochemical characteristic data to generate comparable electrochemical characteristic vectors, thereby obtaining single cell response data.

8. The battery aging status detection method according to claim 7, characterized in that: Step S4 includes the following steps: Step S41: Analyze the performance difference and charge-discharge inconsistency between single cells on the initial health status data and single cell aging risk assessment data, so as to obtain battery pack imbalance data; Step S42: designing a dynamic charging and discharging parameter adjustment strategy according to the battery pack imbalance data, thereby obtaining a charging and discharging parameter optimization scheme, wherein the charging and discharging parameters include a balanced charging current, a constant current charging time, and a balanced charging pulse parameter; Step S43: performing variable pulse balanced charging according to the charging and discharging parameter optimization scheme, and performing differentiated charging and discharging processing on single cells with different aging degrees, so as to obtain charging and discharging balance optimization data; Step S44: monitoring the real-time response characteristics of the single cells during the optimization process according to the charge-discharge balance optimization data, and recording the performance changes before and after the charge-discharge parameters are adjusted, thereby obtaining balance adjustment effect data; Step S45: generating battery pack balanced charge and discharge strategy data based on the balance adjustment effect data.

9. A battery aging status detection system, characterized in that: Used to execute the battery aging status detection method according to claim 1, the battery aging status detection system comprises: The structural feature acquisition module is used to obtain the structural design data of the battery pack of the drone; perform battery cell health monitoring based on the battery pack structural design data to generate a multi-parameter feature vector; establish a data mapping relationship of the battery cell indicators through the multi-parameter feature vector, thereby obtaining the initial health status data of the battery pack; The operating condition feature reconstruction module is used to collect the charging and discharging sample data of different flight conditions of the UAV; based on the charging and discharging sample data, a dynamic current excitation sequence is constructed to simulate the current load changes in different flight stages, thereby generating a multi-operating condition current disturbance sequence; The aging feature recognition module is used to capture the dynamic response characteristics of the single cell through a high-frequency current sensor according to the current disturbance sequence of multiple working conditions, and to extract the electrochemical characteristics in real time, so as to obtain the single cell response data; the key aging index analysis based on the internal resistance change, capacity attenuation and temperature anomaly of the single cell response data is performed to generate the single cell aging risk assessment data; The performance balance control module is used to dynamically adjust the charge and discharge parameters based on the optimization of the charge and discharge balance of the single cell according to the initial health status data and the single cell aging risk assessment data, so as to obtain the battery pack balanced charge and discharge strategy data; The health assessment and early warning module is used to conduct real-time health assessment of the battery pack based on the battery pack balanced charge and discharge strategy data, and integrate multi-dimensional aging characteristics to generate a comprehensive battery health status assessment report to provide battery pack life prediction and maintenance recommendations.

10. A battery aging status detection device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the battery aging status detection method as claimed in any one of claims 1 to 8.

Images