A charging method and device for electric locomotive battery

Abstract

The present application provides a method and device for charging an electric locomotive battery, which relates to the field of battery charging technology, including: data cleaning and feature extraction based on the battery operating parameters after preliminary processing, calculating the battery health state based on the obtained health state, obtaining the health state parameters, calculating the remaining service life based on the obtained life prediction results, obtaining the life prediction parameters, performing abnormality detection and fault identification based on the obtained health state parameters and life prediction parameters, and generating a report based on the obtained health state parameters and life prediction parameters to obtain the first report data. Through the combination of high-precision sensors with filtering and data calibration algorithms, high-frequency noise and system errors in the battery operating parameters are effectively eliminated, thereby significantly improving the calculation accuracy of the health state parameters and life prediction parameters, and ensuring reliability under a variety of complex operating conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of battery charging, and more particularly to a method and device for charging a battery of an electric locomotive. Background Art

[0002] The electric locomotive battery system is a key component of rail transit vehicles and is widely used in shunting locomotives, rail engineering vehicles and new energy rail transit equipment. These battery systems not only require high energy density and long life, but also need to have good environmental adaptability to ensure the safe operation of vehicles under complex working conditions such as extreme cold and extreme heat. However, with the widespread application of battery technology, batteries are prone to capacity attenuation and increased internal resistance during high-frequency charging and discharging, which in turn affects the battery health status and remaining service life. Therefore, it is particularly important to develop efficient charging methods and intelligent management devices for rail power battery systems.

[0003] At present, electric locomotive battery systems mostly use traditional voltage and current monitoring and simple overcharge protection methods, and fail to dynamically adjust the real-time operating status of the battery. For example, the existing charging technology lacks an assessment of the battery health status and cannot accurately predict the remaining battery life. In addition, the existing technology often relies on a single parameter or preset threshold in abnormality detection and fault diagnosis, and it is difficult to comprehensively judge the causes and impacts of multiple faults, resulting in untimely risk management. Especially when facing different battery systems, the existing methods cannot flexibly adjust the charging strategy according to the battery characteristics. Summary of the invention

[0004] The purpose of the present invention is to address the deficiencies in the prior art and provide a charging method and device for an electric locomotive battery. The method aims to obtain battery operating parameters, combine the Kalman filter algorithm and the linear compensation model, perform high-precision filtering and calibration on the data, and generate the optimal charging strategy through real-time evaluation of the battery health status and remaining service life, and accurately identify various fault types through the abnormal detection module. The system supports the charging needs of different battery systems, and can dynamically adjust the maximum charging voltage and minimum discharge voltage according to battery types such as lithium titanate and lithium iron phosphate to ensure the safe operation of the battery. Through intelligent repair charging strategies, the battery life is effectively extended, and the overall reliability of the rail power battery system is improved.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for charging a battery of an electric locomotive, comprising:

[0007] Step S100, obtaining and converting battery operating parameters, performing preliminary processing based on the converted battery operating parameters, monitoring based on the preliminary processed battery operating parameters, and obtaining health status and life prediction results.

[0008] Step S200: perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing, calculate the battery health status based on the obtained health status, obtain health status parameters, calculate the remaining service life based on the obtained life prediction result, obtain life prediction parameters, perform abnormality detection and fault identification based on the obtained health status parameters and life prediction parameters, generate a report based on the obtained health status parameters and life prediction parameters, and obtain first report data.

[0009] Step S300: perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters to obtain lithium dendrite data, electrolyte data and active material data; perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters to obtain repairable data and unrepairable data; define and generate a report of high-risk faults based on the obtained repairable data and unrepairable data to obtain second report data.

[0010] Step S400: generating a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and unrepairable data, and optimizing repair parameters based on the generated repair strategy.

[0011] Step S500: Perform charging control based on the optimized repair strategy, perform abnormality detection on the charging control process, and perform preliminary evaluation on the restored battery.

[0012] Step S600: perform an extended evaluation on the battery recovery effect, generate a repair effect report, and output the integrated report.

[0013] As a preferred solution of the present invention, step S100 is specifically:

[0014] Step S100.1, obtaining battery operating parameters and converting them.

[0015] Battery operating parameters include: voltage, current, temperature and internal resistance.

[0016] The voltage sensor is used to set the error range to ≤±0.01V to obtain voltage parameters, the current sensor is used to set the error range to ≤±0.1A to obtain current, the temperature sensor is used to set the error range to ≤±0.5°C to obtain temperature, and the internal resistance sensor is used to set the frequency response range to 1Hz-10kHz to obtain internal resistance.

[0017] The obtained battery operating parameters are converted into digital signals using an analog-to-digital converter.

[0018] Step S100.2: Perform preliminary processing based on the converted battery operating parameters.

[0019] Preliminary processing includes: filtering and data calibration.

[0020] The high-frequency noise in the battery operating parameters is removed by Kalman filtering, and the battery operating parameters after the high-frequency noise is removed are corrected by a linear compensation model. After correction based on the linear compensation model, a timestamp and a unique identifier are added to the battery operating parameters.

[0021] It should be noted that, since the preliminary processing is a technology well known to those skilled in the art, it will not be described in detail here.

[0022] Step S100.3: Monitor the battery operating parameters based on the preliminary processed battery operating parameters.

[0023] The BMS is used to detect battery types including lithium titanate, lithium iron phosphate and ternary lithium batteries, and the maximum and minimum allowable voltages of lithium titanate, lithium iron phosphate and ternary lithium batteries are extracted from the database. If the current voltage V 当前 Meet V 当前 >V 允许最大 or V 当前＜ V 允许最小 When the alarm is triggered,

[0024] Trigger warning includes: voltage close to V 允许最大 or V 允许最小 When the voltage exceeds V 允许最大 or V 允许最小 When the alarm is triggered, the sound and light alarm is triggered, and the abnormal information is sent to the mobile device. If the alarm persists, the battery connection is disconnected;

[0025] Based on the battery operating parameters after preliminary processing, the integral method is used to calculate the current battery SOC value, obtain the SOC parameter data, and define the SOC safety range for the battery type: 5%≤SOC≤95% for lithium batteries, 10%≤SOC≤90% for lithium iron phosphate batteries, and 20%≤SOC≤80% for lithium titanate batteries;

[0026] When the SOC exceeds the safety range, an early warning is triggered, including: issuing a reminder and triggering an alarm and limiting charging and discharging operations;

[0027] The instrument panel is used to monitor the battery operating parameters and SOC parameter data, and provide health status and life prediction results. As a preferred solution of the present invention, the step S200 is specifically as follows:

[0028] Step S200.1: Perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing.

[0029] Data cleaning uses statistical methods to remove outliers and interpolates the battery operating parameters after preliminary processing to fill in missing values.

[0030] It should be noted here that since data cleaning is a technology well known to people in this field, it will not be elaborated here.

[0031] Feature extraction includes: voltage change rate, temperature gradient and internal resistance change rate, and voltage change parameters, temperature gradient parameters and internal resistance change parameters are obtained.

[0032] Step S200.2: Calculate the battery health state based on the obtained health state.

[0033] Through the health status definition, the health status parameters are obtained.

[0034] Step S200.3: Calculate the remaining service life based on the obtained life prediction result.

[0035] The life prediction parameters are obtained through the definition of life prediction results.

[0036] Step S200.4: perform abnormality detection and fault identification based on the obtained health status parameters and life prediction parameters.

[0037] Statistical methods were used to detect abnormalities in health status parameters and life prediction parameters.

[0038] The fault type is identified through characteristic pattern matching. When the temperature rises abnormally, there is a risk of thermal runaway. When the internal resistance increases abnormally, it indicates active material depletion.

[0039] Step S200.5: Generate a report based on the obtained health status parameters and life prediction parameters to obtain report data.

[0040] The report content includes: current values ​​and predicted trends of health status parameters and life prediction parameters, SOC parameter data and detected anomalies and fault types.

[0041] The time series curves of health status parameters and life prediction parameters and abnormal event markers are displayed in graphs.

[0042] As a preferred solution of the present invention, step S300 is specifically:

[0043] Step S300.1, perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters.

[0044] Fault diagnosis includes: detection of lithium dendrite growth, detection of electrolyte aging and detection of active material consumption, and obtaining lithium dendrite data, electrolyte data and active material data.

[0045] Step S300.2: Perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters.

[0046] Repairable means that according to the lithium dendrite data, it is repaired through high-frequency pulse current, and according to the electrolyte data, the electrolyte is regenerated through long-term charging with low current to obtain the repairable data.

[0047] Irreparable means that large-scale loss of active materials and internal short circuits lead to thermal runaway, resulting in irreparable data.

[0048] Step S300.3: Define high-risk faults and generate reports based on the obtained repairable data and unrepairable data.

[0049] When the temperature rises abnormally, the internal resistance increases sharply, or the remaining life approaches zero, it is defined as a high-risk fault and is handled by stopping the charging or discharging process or triggering an emergency alarm.

[0050] The report generated includes: fault type, related parameters and unrepairable fault report.

[0051] Fault types include thermal runaway and internal short circuit. Related parameters include health status parameters, life prediction parameters, temperature, internal resistance and diagnostic timestamp. Unrepairable fault reports include current status parameters and recommended solutions.

[0052] As a preferred solution of the present invention, the step S400 is specifically:

[0053] Step S400.1, generating a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and unrepairable data.

[0054] Repair strategies include: lithium dendrite repair, electrolyte aging repair and material wear repair.

[0055] Lithium dendrite repair, dissolving early lithium dendrites through high-frequency pulsed current while limiting heat generation.

[0056] To repair electrolyte aging, low current and long-term constant voltage charging is used to restore the chemical activity of the electrolyte.

[0057] Active material data, the active material in the edge region is activated by step current charging.

[0058] During the repair charging process, the SOC parameter data is monitored through the instrument panel. If the SOC is close to the upper or lower limit, the charging strategy is dynamically adjusted, including: when the SOC is close to the upper limit, the charging current is reduced, and when the SOC is close to the lower limit, the discharge is limited.

[0059] Step S400.2: Optimize repair parameters based on the generated repair strategy.

[0060] Genetic algorithm is used to optimize the repair parameters, specifically:

[0061] ;

[0062] Where: F 优化 is the optimization objective function, which means maximizing the improvement of health status in unit time while minimizing power loss. ΔSOH is the improvement of health status, which means the improvement of health status after repair. t 修复 is the repair time, which indicates the length of time it takes to complete the charging repair. 损耗 is the power loss, which indicates the power consumed during the repair process, is the power loss weight factor used to balance the relationship between health status improvement and power consumption.

[0063] As a preferred solution of the present invention, step S500 is specifically:

[0064] Step S500.1: Perform charging control based on the optimized repair strategy.

[0065] A DC / DC converter is used to control current and voltage. The DC / DC converter control includes pulse mode, constant voltage low current mode and step current mode.

[0066] The pulse mode is used to repair lithium dendrites, the constant voltage and low current mode is used to repair electrolyte aging, and the step current mode is used to repair material loss.

[0067] When in the charging state, the changes in battery performance are monitored, the battery temperature is detected, and the reasonable range of internal resistance is verified.

[0068] The current is adjusted dynamically according to the detected battery temperature, specifically:

[0069] ;

[0070] Where: I 调整 is the output current after temperature adjustment, indicating the actual output current value after temperature correction, I 输出 is the original output current, indicating the output current value set by the repair charging strategy. is the temperature correction coefficient, which indicates the sensitivity of temperature to current adjustment. is the temperature deviation, which indicates the difference between the current temperature and the target temperature.

[0071] Dynamically adjust the voltage according to the verified internal resistance, specifically:

[0072] ;

[0073] Where: V 调整 is the output voltage after internal resistance adjustment, indicating the actual voltage applied to the battery terminal after internal resistance correction, V 输出 is the original output voltage, indicating the output voltage value set by the repair charging strategy. is the internal resistance correction coefficient, which indicates the sensitivity of the internal resistance change to the output voltage adjustment. It is the internal resistance deviation, which indicates the difference between the current internal resistance and the target internal resistance.

[0074] Step S500.2: Detect abnormality in the charging control process.

[0075] When T>60℃ or T<﹣20℃, it will be judged as temperature abnormality. 输出＞ V 最大 or V 输出＜ V 最小 , it will be judged as voltage abnormality. When the internal resistance is abnormal.

[0076] When an abnormality is detected, the repair process is stopped immediately and switches to safe mode, reducing the charging power or completely disconnecting the output.

[0077] Step S500.3: Perform a preliminary evaluation on the restored battery.

[0078] The health status parameters after recovery are compared with the health status parameters before recovery.

[0079] As a preferred solution of the present invention, step S600 is specifically as follows:

[0080] Step S600.1: Perform an extended evaluation on the battery recovery effect.

[0081] Extended assessment includes: life improvement and capacity recovery rate.

[0082] Step S600.2: Generate a repair effect report and output it in an integrated manner.

[0083] The repair effect report includes: the comparison curves of SOH, RUL and internal resistance change curves before and after repair, and the matching degree between the fault type and the repair plan.

[0084] The obtained first report data, second report data and repair effect report are integrated, the first report data is ranked first, the second report data is ranked second, and the repair effect report is ranked third, and they are output to the cloud in the form of word documents and PDF files.

[0085] A charging device for an electric locomotive battery, specifically comprising:

[0086] The parameter monitoring module is used to obtain and convert the battery operating parameters, perform preliminary processing based on the converted battery operating parameters, monitor the battery operating parameters based on the preliminary processed parameters, and obtain health status and life prediction results.

[0087] A cleaning and extraction module is used to perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing, calculate the battery health state based on the obtained health state, obtain health state parameters, calculate the remaining service life based on the obtained life prediction result, obtain life prediction parameters, perform anomaly detection and fault identification based on the obtained health state parameters and life prediction parameters, generate a report based on the obtained health state parameters and life prediction parameters, and obtain first report data.

[0088] A health analysis module is used to perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters, and obtain lithium dendrite data, electrolyte data and active material data; perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters, and obtain repairable data and unrepairable data; define and generate reports on high-risk faults based on the obtained repairable data and unrepairable data, and obtain second report data.

[0089] The life assessment module is used to generate a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and non-repairable data, and optimize the repair parameters based on the generated repair strategy.

[0090] The anomaly detection module is used to perform charging control based on the optimized repair strategy, detect anomalies in the charging control process, and perform preliminary evaluation of the restored battery.

[0091] The charging optimization module is used to perform an extended evaluation of the battery recovery effect, generate a repair effect report, and perform integrated output.

[0092] Compared with the prior art, the present invention has the following beneficial effects:

[0093] 1. Through high-precision sensors combined with filtering and data calibration algorithms, high-frequency noise and system errors in battery operating parameters are effectively eliminated, thereby significantly improving the calculation accuracy of health status parameters and life prediction parameters, ensuring reliability under a variety of complex operating conditions.

[0094] 2. The multi-dimensional parameter integration model is used to realize real-time monitoring of the health status and life prediction of the battery during operation. It provides accurate classification and identification methods for abnormal data and failure modes, which can quickly identify high-risk failure types and generate detailed reports in a timely manner.

[0095] 3. According to the health status parameters and life prediction parameters, a dynamically optimized charging control strategy is designed, which not only improves the charging efficiency, but also effectively avoids battery damage caused by overcharging or undercharging, thereby extending the battery life.

[0096] 4. After detecting a high-risk fault, the system automatically generates a repair strategy based on the abnormal data and optimizes the repair parameters. Finally, the battery repair effect is comprehensively evaluated and reported. This closed-loop maintenance mechanism significantly improves the operational safety and long-term stability of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] Figure 1 A flowchart of a method for charging a locomotive battery provided in an embodiment of the present application.

[0098] Figure 2 A flow chart of a charging device for an electric locomotive battery provided in an embodiment of the present application. DETAILED DESCRIPTION

[0099] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0100] See also Figure 1 , Figure 1 A flowchart of a method for charging a battery of an electric locomotive is provided for an embodiment of the present application.

[0101] In this embodiment, a method and device for charging an electric locomotive battery may include steps S100, S200, S300, S400, S500 and S600;

[0102] Step S100, obtaining and converting battery operating parameters, performing preliminary processing based on the converted battery operating parameters, monitoring based on the preliminary processed battery operating parameters, and obtaining health status and life prediction results.

[0103] Step S200: perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing, calculate the battery health status based on the obtained health status, obtain health status parameters, calculate the remaining service life based on the obtained life prediction result, obtain life prediction parameters, perform abnormality detection and fault identification based on the obtained health status parameters and life prediction parameters, generate a report based on the obtained health status parameters and life prediction parameters, and obtain first report data.

[0104] Step S300: perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters to obtain lithium dendrite data, electrolyte data and active material data; perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters to obtain repairable data and unrepairable data; define and generate a report of high-risk faults based on the obtained repairable data and unrepairable data to obtain second report data.

[0105] Step S400: generating a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and unrepairable data, and optimizing repair parameters based on the generated repair strategy.

[0106] Step S500: Perform charging control based on the optimized repair strategy, perform abnormality detection on the charging control process, and perform preliminary evaluation on the restored battery.

[0107] Step S600: perform an extended evaluation on the battery recovery effect, generate a repair effect report, and output the integrated report.

[0108] In some specific implementations, the step S100 is specifically:

[0109] Step S100.1, obtaining battery operating parameters and converting them.

[0110] Battery operating parameters include: voltage, current, temperature and internal resistance.

[0111] The voltage sensor is used to set the error range to ≤±0.01V to obtain voltage parameters, the current sensor is used to set the error range to ≤±0.1A to obtain current, the temperature sensor is used to set the error range to ≤±0.5°C to obtain temperature, and the internal resistance sensor is used to set the frequency response range to 1Hz-10kHz to obtain internal resistance.

[0112] The obtained battery operating parameters are converted into digital signals using an analog-to-digital converter.

[0113] Step S100.2: Perform preliminary processing based on the converted battery operating parameters.

[0114] Preliminary processing includes: filtering and data calibration.

[0115] The high-frequency noise in the battery operating parameters is removed by Kalman filtering, and the battery operating parameters after the high-frequency noise is removed are corrected by a linear compensation model. After correction based on the linear compensation model, a timestamp and a unique identifier are added to the battery operating parameters.

[0116] It should be noted that, since the preliminary processing is a technology well known to those skilled in the art, it will not be described in detail here.

[0117] Step S100.3: Monitor the battery operating parameters based on the preliminary processed battery operating parameters.

[0118] The BMS is used to detect battery types including lithium titanate, lithium iron phosphate and ternary lithium batteries, and the maximum and minimum allowable voltages of lithium titanate, lithium iron phosphate and ternary lithium batteries are extracted from the database. If the current voltage V 当前 Meet V 当前 >V 允许最大 or V 当前＜ V 允许最小 When the alarm is triggered,

[0119] Trigger warning includes: voltage close to V 允许最大 or V 允许最小 When the voltage exceeds V 允许最大 or V 允许最小 When the alarm is triggered, the sound and light alarm is triggered, and the abnormal information is sent to the mobile device. If the alarm persists, the battery connection is disconnected;

[0120] Based on the battery operating parameters after preliminary processing, the integral method is used to calculate the current battery SOC value, obtain the SOC parameter data, and define the SOC safety range for the battery type: 5%≤SOC≤95% for lithium batteries, 10%≤SOC≤90% for lithium iron phosphate batteries, and 20%≤SOC≤80% for lithium titanate batteries;

[0121] When the SOC exceeds the safety range, an early warning is triggered, including: issuing a reminder and triggering an alarm and limiting charging and discharging operations;

[0122] Use the dashboard to monitor battery operating parameters and SOC parameter data, and provide health status and life prediction results.

[0123] In some specific implementations, the step S200 is specifically:

[0124] Step S200.1: Perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing.

[0125] Data cleaning uses statistical methods to remove outliers and interpolates the battery operating parameters after preliminary processing to fill in missing values.

[0126] It should be noted here that since data cleaning is a technology well known to people in this field, it will not be elaborated here.

[0127] Feature extraction includes: voltage change rate, temperature gradient and internal resistance change rate, and voltage change parameters, temperature gradient parameters and internal resistance change parameters are obtained.

[0128] Voltage change rate, specifically:

[0129] ;

[0130] Where: is the voltage change rate, which indicates the change of battery voltage per unit time, V 终点 is the terminal voltage, i.e., the voltage at the time interval The voltage value measured at the end of 起点 is the starting voltage, i.e., the The voltage value measured at the beginning of is the time interval, which represents the length of time it takes for the voltage to change.

[0131] The temperature gradient is:

[0132] ;

[0133] Where: is the temperature gradient, which indicates the rate of change of battery temperature relative to ambient temperature per unit time, T 当前 is the current temperature, that is, the actual temperature of the battery measured at a certain point in time, T 环境 is the ambient temperature, i.e. the temperature of the environment where the battery is located, is the time interval, which represents the length of time it takes for the temperature to change.

[0134] The internal resistance change rate is:

[0135] ;

[0136] Where: is the internal resistance change rate, which indicates the percentage change of the current internal resistance relative to the initial internal resistance, R 当前 is the current internal resistance, that is, the internal resistance value measured when the battery is currently running, R 初始 It is the initial internal resistance, that is, the internal resistance value measured when the battery is in a new state.

[0137] Step S200.2: Calculate the battery health state based on the obtained health state.

[0138] Through the health status definition, the health status parameters are obtained.

[0139] Health status definition, specifically:

[0140] ;

[0141] Where: SOH is the battery health state, which indicates the percentage of the current battery performance relative to the new state, C 当前 is the current capacity, i.e. the actual available capacity of the battery in its current state, C 额定 It is the rated capacity, which is the theoretical maximum capacity of the battery when it is designed.

[0142] Step S200.3: Calculate the remaining service life based on the obtained life prediction result.

[0143] The life prediction parameters are obtained through the definition of life prediction results.

[0144] The definition of life prediction results is as follows:

[0145] ;

[0146] Where: RUL is the remaining service life, which means the time the battery can still operate normally in the current state, C 当前 is the current capacity, that is, the actual capacity of the battery in its current state, C 阈值 The capacity threshold is the minimum capacity available for the battery. If the capacity is lower than this value, the battery cannot be used normally. It is the capacity decay rate, which indicates the reduction of capacity per unit time.

[0147] Step S200.4: perform abnormality detection and fault identification based on the obtained health status parameters and life prediction parameters.

[0148] Statistical methods were used to detect abnormalities in health status parameters and life prediction parameters.

[0149] When T>60℃ or T<﹣20℃, it is abnormal temperature. h >2×R 初始 When , the internal resistance is too large, specifically:

[0150] ;

[0151] Where: E is the degree of abnormality, indicating the degree to which the current data deviates from the normal range; X is the current data point value, that is, the specific data currently measured. is the data mean, that is, the average value of historical data, is the standard deviation of the data, which indicates the range of data fluctuation.

[0152] The fault type is identified through characteristic pattern matching. When the temperature rises abnormally, there is a risk of thermal runaway. When the internal resistance increases abnormally, it indicates active material depletion.

[0153] Step S200.5: Generate a report based on the obtained health status parameters and life prediction parameters to obtain report data.

[0154] The report content includes: current values ​​and predicted trends of health status parameters and life prediction parameters, SOC parameter data and detected anomalies and fault types.

[0155] The time series curves of health status parameters and life prediction parameters and abnormal event markers are displayed in graphs.

[0156] In some specific implementations, the step S300 is specifically:

[0157] Step S300.1, perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters.

[0158] Fault diagnosis includes: detection of lithium dendrite growth, detection of electrolyte aging and detection of active material consumption, and obtaining lithium dendrite data, electrolyte data and active material data.

[0159] Detection of lithium dendrite growth, specifically:

[0160] ;

[0161] Where: R 增量 is the internal resistance increment, which indicates the change of the current internal resistance value relative to the internal resistance value at the previous time point, R 当前 is the current internal resistance value, that is, the battery internal resistance measured at the current time point, R 前一时刻 It is the internal resistance value at the previous moment, that is, the internal resistance data recorded at the previous time point.

[0162] The detection of electrolyte aging is as follows:

[0163] ;

[0164] Where: t 恢复 is the voltage recovery time, which indicates the time it takes for the battery voltage to recover from a low value to a steady-state value, t 终点 is the time point when the voltage returns to steady state, t 起点 This is the point in time when the voltage starts to recover.

[0165] Active material depletion detection, specifically:

[0166] ;

[0167] Where: SOH 变化率 SOH is the rate of change of the health status, which indicates the change of the battery health status per unit time. 当前 Is the battery health status at the current time, SOH 前一时刻 is the battery health status at the previous point in time, It is the time interval, that is, the time difference between the current time point and the previous time point.

[0168] Step S300.2: Perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters.

[0169] Repairable means that according to the lithium dendrite data, it is repaired by high-frequency pulse current, and according to the electrolyte data, the electrolyte is regenerated by long-term charging at low current, and the repairable data is obtained, specifically:

[0170] ;

[0171] In the formula: SOH is the battery health state, which indicates the percentage of the current battery performance relative to the new state; RUL is the remaining service life, which indicates the number of charge and discharge cycles that the battery can still operate normally in the current state; 40% is the SOH threshold. A value below this value indicates that the battery performance is seriously degraded; 20 cycles is the RUL threshold. A value below this value indicates that the battery is about to reach the limit of its service life.

[0172] Unrepairable means that the active material is worn out over a large area and the internal short circuit leads to thermal runaway, resulting in unrepairable data, specifically:

[0173] ;

[0174] In the formula: SOH≤40% means the battery health status is lower than 40%, indicating that the battery performance is severely degraded and cannot be restored to a usable state through repair; RUL≤20 cycles means the remaining service life is less than or equal to 20 cycles, indicating that the battery is approaching the end of its service life.

[0175] Step S300.3: Define high-risk faults and generate reports based on the obtained repairable data and unrepairable data.

[0176] When the temperature rises abnormally, the internal resistance increases sharply, or the remaining life approaches zero, it is defined as a high-risk fault and is handled by stopping the charging or discharging process or triggering an emergency alarm.

[0177] The report generated includes: fault type, related parameters and unrepairable fault report.

[0178] Fault types include thermal runaway and internal short circuit. Related parameters include health status parameters, life prediction parameters, temperature, internal resistance and diagnostic timestamp. Unrepairable fault reports include current status parameters and recommended solutions.

[0179] In some specific implementations, the step S400 is specifically:

[0180] Step S400.1, generating a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and unrepairable data.

[0181] Repair strategies include: lithium dendrite repair, electrolyte aging repair and material wear repair.

[0182] Lithium dendrite repair, dissolving early lithium dendrites through high-frequency pulse current while limiting heat generation, specifically:

[0183] ;

[0184] Where: f 脉冲 is the frequency of the pulse current, which indicates the number of oscillations of the pulse current per second, T 脉冲 is the pulse cycle time, i.e. the length of time of a complete pulse.

[0185] ;

[0186] Where: I 脉冲 is the amplitude of the pulse current, indicating the magnitude of the high-frequency pulse current. It is the pulse current correction coefficient, which is used to adjust the pulse current amplitude. The value range is usually 0.2≤ ≤0.5, I 额定 It is the rated current of the battery, which indicates the maximum operating current for which the battery is designed.

[0187] Electrolyte aging repair uses low current and long-term constant voltage charging to restore the chemical activity of the electrolyte. Specifically:

[0188] ;

[0189] Where: I 修复 is the repair current, which indicates the current used in the low current constant voltage repair process. It is the repair current correction coefficient, which is used to adjust the repair current amplitude. The value range is usually 0.1≤ ≤0.3, I 额定 is the rated current of the battery.

[0190] ;

[0191] Where: V 修复 is the repair voltage, which indicates the constant voltage value during the repair process, V 额定 is the rated voltage of the battery, that is, the standard voltage value when it is designed. It is the repair voltage correction value, which is used to adjust the constant voltage value. The value range is .

[0192] Active material data, activated by step current charging in the edge region, specifically:

[0193] ;

[0194] Where: I 阶梯 It is a step-by-step current value, indicating the current magnitude at each stage of the repair process, I 额定 is the rated current of the battery, is the current step number, indicating the current step number, N 总 is the total number of steps, indicating the total number of steps included in the repair process.

[0195] ;

[0196] Where: t 阶梯 is the duration of each step, indicating the length of time for each repair phase, t 总 is the total time of the repair process, indicating the time to complete the entire repair operation, N 总 is the total number of steps.

[0197] During the repair charging process, the SOC parameter data is monitored through the instrument panel. If the SOC is close to the upper or lower limit, the charging strategy is dynamically adjusted, including: when the SOC is close to the upper limit, the charging current is reduced, and when the SOC is close to the lower limit, the discharge is limited.

[0198] Step S400.2: Optimize repair parameters based on the generated repair strategy.

[0199] Genetic algorithm is used to optimize the repair parameters, specifically:

[0200] ;

[0201] Where: F 优化 is the optimization objective function, which means maximizing the improvement of health status per unit time while minimizing power loss. is the improvement in health status, indicating the improvement in health status after repair, t 修复 is the repair time, which indicates the length of time it takes to complete the charging repair. 损耗 is the power loss, which indicates the power consumed during the repair process, is the power loss weight factor used to balance the relationship between health status improvement and power consumption.

[0202] In some specific implementations, the step S500 is specifically:

[0203] Step S500.1: Perform charging control based on the optimized repair strategy.

[0204] A DC / DC converter is used to control current and voltage. The DC / DC converter control includes pulse mode, constant voltage low current mode and step current mode.

[0205] The pulse mode is used to repair lithium dendrites, the constant voltage and low current mode is used to repair electrolyte aging, and the step current mode is used to repair material loss.

[0206] When in the charging state, the changes in battery performance are monitored, the battery temperature is detected, and the reasonable range of internal resistance is verified.

[0207] The current is adjusted dynamically according to the detected battery temperature, specifically:

[0208] ;

[0209] Where: I 调整 is the output current after temperature adjustment, indicating the actual output current value after temperature correction, I 输出 is the original output current, indicating the output current value set by the repair charging strategy. is the temperature correction coefficient, which indicates the sensitivity of temperature to current adjustment. is the temperature deviation, which indicates the difference between the current temperature and the target temperature.

[0210] Dynamically adjust the voltage according to the verified internal resistance, specifically:

[0211] ;

[0212] Where: V 调整 is the output voltage after internal resistance adjustment, indicating the actual voltage applied to the battery terminal after internal resistance correction, V 输出 is the original output voltage, indicating the output voltage value set by the repair charging strategy. is the internal resistance correction coefficient, which indicates the sensitivity of the internal resistance change to the output voltage adjustment. It is the internal resistance deviation, which indicates the difference between the current internal resistance and the target internal resistance.

[0213] The battery operating parameters are obtained every 30 minutes, the target voltage and current are calculated according to the repair charging curve, the charging parameters are adjusted according to the deviation of the reasonable range of battery temperature and internal resistance, and the adjusted voltage and current instructions are output to the DC / DC converter.

[0214] Step S500.2: Detect abnormality in the charging control process.

[0215] When T>60℃ or T<﹣20℃, it will be judged as temperature abnormality. 输出＞ V 最大 or V 输出＜ V 最小, it will be judged as voltage abnormality. When the internal resistance is abnormal.

[0216] When an abnormality is detected, the repair process is stopped immediately and the system switches to safe mode, reducing the charging power or completely disconnecting the output. The specific safe mode is:

[0217] ;

[0218] Where: I 安全 It is the reduced power output current in safety mode, indicating the safe current value output under abnormal conditions, I 输出 is the original output current, which indicates the output current value set by the repair charging strategy.

[0219] Step S500.3: Perform a preliminary evaluation on the restored battery.

[0220] Compare the health status parameters after recovery with those before recovery, specifically:

[0221] ;

[0222] Where: It is the improvement of health status, indicating the improvement of battery health status after repair and charging. 修复后 It is the health status measured after repair, indicating the SOH measured after the repair and charging are completed. 修复前 It is the health state measured before repair, indicating the SOH measured before repair charging.

[0223] Evaluate the improvement of internal resistance, specifically:

[0224] ;

[0225] Where: is the internal resistance improvement, which indicates the reduction in the internal resistance of the battery after repair and charging. 修复前 is the internal resistance measured before repair, indicating the internal resistance value before the repair charging begins, R 修复后 It is the internal resistance measured after repair, indicating the internal resistance value after the repair and charging are completed.

[0226] In some specific implementations, step S600 is specifically:

[0227] Step S600.1: Perform an extended evaluation on the battery recovery effect.

[0228] Extended assessment includes: life improvement and capacity recovery rate.

[0229] The lifespan improvement is as follows:

[0230] ;

[0231] Where: The improvement in the remaining service life indicates the increase in the remaining service life of the battery after repair compared to that before repair. 修复后 It is the remaining service life after repair, which means the remaining service life of the battery after the repair is completed. RUL 修复前 It is the remaining service life before repair, which means the remaining service life of the battery measured before repair.

[0232] Capacity recovery rate, specifically:

[0233] ;

[0234] Where: is the capacity recovery rate, which indicates the recovery ratio of the battery capacity after repair relative to the rated capacity, C 修复后 It is the actual capacity after repair, which means the actual capacity of the battery measured after the repair is completed. C 修复前 It is the actual capacity before repair, which means the actual capacity of the battery measured before repair. C rated is the rated capacity of the battery, which means the theoretical maximum capacity when the battery is designed.

[0235] Step S600.2: Generate a repair effect report and output it in an integrated manner.

[0236] The repair effect report includes: the comparison curves of SOH, RUL and internal resistance change curves before and after repair, and the matching degree between the fault type and the repair plan.

[0237] The obtained first report data, second report data and repair effect report are integrated, the first report data is ranked first, the second report data is ranked second, and the repair effect report is ranked third, and they are output to the cloud in the form of word documents and PDF files.

[0238] See also Figure 2 , Figure 2 A flow chart of a charging device for an electric locomotive battery is provided for an embodiment of the present application.

[0239] A charging device for an electric locomotive battery, specifically comprising:

[0240] The parameter monitoring module is used to obtain and convert the battery operating parameters, perform preliminary processing based on the converted battery operating parameters, monitor the battery operating parameters based on the preliminary processed parameters, and obtain health status and life prediction results.

[0241] A cleaning and extraction module is used to perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing, calculate the battery health state based on the obtained health state, obtain health state parameters, calculate the remaining service life based on the obtained life prediction result, obtain life prediction parameters, perform anomaly detection and fault identification based on the obtained health state parameters and life prediction parameters, generate a report based on the obtained health state parameters and life prediction parameters, and obtain first report data.

[0242] A health analysis module is used to perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters, and obtain lithium dendrite data, electrolyte data and active material data; perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters, and obtain repairable data and unrepairable data; define and generate reports on high-risk faults based on the obtained repairable data and unrepairable data, and obtain second report data.

[0243] The life assessment module is used to generate a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and non-repairable data, and optimize the repair parameters based on the generated repair strategy.

[0244] The anomaly detection module is used to perform charging control based on the optimized repair strategy, detect anomalies in the charging control process, and perform preliminary evaluation of the restored battery.

[0245] The charging optimization module is used to perform an extended evaluation of the battery recovery effect, generate a repair effect report, and perform integrated output.

[0246] In the above content, in actual application, first, the battery operating parameters are collected through high-precision sensors, including voltage, current, temperature and internal resistance. The error range of the sensor is voltage ≤±0.01V, current ≤±0.1A, temperature ≤±0.5°C, and the internal resistance frequency response range is 1Hz-10kHz. The collected analog signal is converted into a digital signal through an analog-to-digital converter to ensure the accuracy of the data. The converted data is filtered and corrected to remove noise and compensate for system errors. To ensure the traceability of the data, all operating parameters are attached with timestamps and unique identifiers.

[0247] Based on the processed operating parameters, data cleaning and feature extraction are performed. Data cleaning uses statistical methods to remove outliers and fill missing values. The feature extraction stage extracts key features such as voltage change rate, temperature gradient, and internal resistance change rate from the operating parameters to provide high-quality input for subsequent analysis. All feature extraction results are stored in the form of structured data to ensure the efficiency and consistency of the calculation.

[0248] The health status and life prediction are performed based on the operating parameters. The health status parameters are obtained by analyzing the ratio of the current capacity to the rated capacity. The life prediction estimates the remaining service life based on the capacity attenuation rate. The health status and life prediction results are verified through model verification to ensure accuracy, and their reliability is verified by combining multiple experimental data. All prediction results are presented in charts for intuitive analysis.

[0249] Anomaly detection and fault identification are performed based on health status parameters and life prediction parameters. Anomaly detection accurately identifies abnormal fluctuations in voltage, temperature and internal resistance by comparing historical data with current data. The fault identification link locates possible sources of faults, such as lithium dendrite growth, electrolyte failure or active material depletion, through multi-dimensional feature matching technology. The detection results are recorded in real time and notified in the form of an alarm.

[0250] The repair strategy is generated based on the detection results. After anomaly detection, an optimization strategy is formulated according to the adjustable range of the repair parameters, and the feasibility and effectiveness of the repair plan are verified through simulation. The repair process is guided by the optimized parameters, and the best effect is ensured through gradual adjustment. All repair plans are output in the form of operation instructions, with detailed verification data.

[0251] After the battery repair is completed, the effect evaluation and extended verification are carried out. The repair effect is comprehensively analyzed through the degree of improvement of health status parameters and life prediction parameters. After the evaluation is completed, an integrated report is generated, including comparative data before and after repair, abnormal repair records and extended application suggestions. The repaired battery undergoes multiple rounds of cycle charge and discharge tests to verify the consistency and stability of its performance improvement.

[0252] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for charging a battery of an electric locomotive, characterized in that: The steps include: S100, obtaining and converting battery operating parameters, performing preliminary processing based on the converted battery operating parameters, monitoring based on the preliminary processed battery operating parameters, and obtaining health status and life prediction results; S200, performing data cleaning and feature extraction based on the battery operation parameters after preliminary processing, obtaining voltage change parameters, temperature gradient parameters and internal resistance change parameters, calculating the battery health state based on the obtained health state, obtaining health state parameters, calculating the remaining service life based on the obtained life prediction result, obtaining life prediction parameters, performing abnormality detection and fault identification based on the obtained health state parameters and life prediction parameters, and generating a report based on the obtained health state parameters and life prediction parameters to obtain first report data; S300, performing fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters, obtaining lithium dendrite data, electrolyte data and active material data, and performing repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters, and obtaining repairable data and non-repairable data; Repairable means: repairing by high-frequency pulse current according to lithium dendrite data, and regenerating electrolyte by long-term charging with low current according to electrolyte data, to obtain repairable data; Irreparable means: large-scale consumption of active materials and internal short circuits lead to thermal runaway, resulting in irreparable data; Defining and reporting high-risk failures based on the obtained repairable data and unrepairable data to obtain second report data; S400, generating a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and non-repairable data, the repair strategy including: lithium dendrite repair, electrolyte aging repair and material depletion repair; Lithium dendrite repair: Dissolve early lithium dendrites through high-frequency pulsed current while limiting heat generation; Electrolyte aging repair: Use low current and long-term constant voltage charging to restore the chemical activity of the electrolyte; Material depletion repair: Activate active materials in edge areas through step-by-step current charging; The SOC parameter data is monitored through the instrument panel during the repair charging process. If the SOC is close to the upper or lower limit, the charging strategy is dynamically adjusted, including: when the SOC is close to the upper limit, the charging current is reduced, and when the SOC is close to the lower limit, the discharge is limited; Optimizing repair parameters based on the generated repair strategy; Genetic algorithm is used to optimize the repair parameters, specifically: ； Where: is the optimization objective function, which means maximizing the improvement of health status per unit time while minimizing power loss. is the improvement in health status, indicating the improvement in health status after repair. is the repair time, which indicates the length of time it takes to complete the charging repair. is the power loss, which indicates the power consumed during the repair process, is the power loss weight factor, which is used to balance the relationship between health status improvement and power consumption; S500, performing charging control based on the optimized repair strategy, performing abnormality detection on the charging control process, and performing preliminary evaluation on the restored battery; S600, perform an extended evaluation on the battery recovery effect, generate a repair effect report, and perform integrated output.

2. A method for charging an electric locomotive battery according to claim 1, characterized in that: The S100 is specifically: S100.

1. Obtain battery operating parameters and convert them; Battery operating parameters include: voltage, current, temperature and internal resistance; An analog-to-digital converter is used to convert the obtained battery operating parameters into digital signals; S100.

2. Perform preliminary processing based on the converted battery operating parameters; The preliminary processing includes: filtering and data calibration; Removing high-frequency noise from battery operating parameters by using Kalman filtering, correcting the battery operating parameters after removing the high-frequency noise by using a linear compensation model, and adding a timestamp and a unique identifier to the battery operating parameters after correction based on the linear compensation model; S100.3, monitoring the battery operating parameters based on the preliminary processing; The BMS is used to detect battery types including lithium titanate, lithium iron phosphate and ternary lithium batteries, and the maximum and minimum allowable voltages of lithium titanate, lithium iron phosphate and ternary lithium batteries are extracted from the database. If the current voltage satisfy or When the alarm is triggered, Trigger warning includes: voltage approaching or When the voltage exceeds or When the alarm is triggered, the sound and light alarm is triggered, and the abnormal information is sent to the mobile device. If the alarm persists, the battery connection is disconnected; Based on the battery operating parameters after preliminary processing, the integral method is used to calculate the current battery SOC value, obtain the SOC parameter data, and define the SOC safety range for the battery type: 5%≤SOC≤95% for lithium batteries, 10%≤SOC≤90% for lithium iron phosphate batteries, and 20%≤SOC≤80% for lithium titanate batteries; When the SOC exceeds the safe range, an early warning is triggered, including: issuing a reminder and triggering an alarm and limiting charging and discharging operations; Use the dashboard to monitor battery operating parameters and SOC parameter data, and provide health status and life prediction results.

3. A method for charging a locomotive battery according to claim 1, characterized in that: The S200 is specifically: S200.

1. Perform data cleaning and feature extraction based on the battery operating parameters after preliminary processing; Data cleaning uses statistical methods to remove outliers and interpolate the battery operating parameters after preliminary processing to fill in missing values; Feature extraction includes: voltage change rate, temperature gradient and internal resistance change rate, and voltage change parameters, temperature gradient parameters and internal resistance change parameters are obtained; S200.

2. Calculate the battery health state based on the obtained health state; Through the definition of health status, the health status parameters are obtained; S200.

3. Calculate the remaining service life based on the obtained life prediction result; Through the definition of life prediction results, life prediction parameters are obtained; S200.

4. Perform anomaly detection and fault identification based on the obtained health status parameters and life prediction parameters; Statistical methods are used to detect abnormalities in health status parameters and life prediction parameters; Identify fault types through feature pattern matching. When the temperature rises abnormally, it indicates thermal runaway risk. When the internal resistance increases abnormally, it indicates active material depletion. S200.

5. Generate a report based on the obtained health status parameters and life prediction parameters to obtain report data; The report content includes: current values ​​and predicted trends of health status parameters and life prediction parameters, SOC parameter data and detected abnormalities and fault types; The time series curves of health status parameters and life prediction parameters and abnormal event markers are displayed in graphs.

4. A method for charging a battery of an electric locomotive according to claim 1, characterized in that: The S300 is specifically: S300.

1. Perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters; Fault diagnosis includes: detection of lithium dendrite growth, electrolyte aging and active material consumption, and obtaining lithium dendrite data, electrolyte data and active material data; S300.2, performing repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters; S300.

3. Define and generate reports of high-risk failures based on the obtained repairable data and unrepairable data; When the temperature rises abnormally, the internal resistance increases sharply, or the remaining life is close to zero, it is defined as a high-risk fault and is handled by stopping the charging or discharging process or triggering an emergency alarm; The report generated includes: fault type, related parameters and unrepairable fault report; Fault types include: thermal runaway and internal short circuit. Related parameters include: health status parameters, life prediction parameters, temperature, internal resistance and diagnostic timestamp. Unrepairable fault reports include: current status parameters and recommended solutions.

5. A method for charging a locomotive battery according to claim 1, characterized in that: The S500 is specifically: S500.

1. Perform charging control based on the optimized repair strategy; The DC / DC converter is used to control the current and voltage. The DC / DC converter control includes: pulse mode, constant voltage low current mode and step current mode; The pulse mode is used to repair lithium dendrites, the constant voltage and low current mode is used to repair electrolyte aging, and the step current mode is used to repair material loss. When in charging state, monitor the changes in battery performance, detect battery temperature, and verify the reasonable range of internal resistance; The current is adjusted dynamically according to the detected battery temperature, specifically: ； Where: is the output current after temperature adjustment, indicating the actual output current value after temperature correction. is the original output current, indicating the output current value set by the repair charging strategy. is the temperature correction coefficient, which indicates the sensitivity of temperature to current adjustment. is the temperature deviation, which indicates the difference between the current temperature and the target temperature; Dynamically adjust the voltage according to the verified internal resistance, specifically: ； Where: It is the output voltage after internal resistance adjustment, indicating the actual voltage applied to the battery terminal after correction based on internal resistance. is the original output voltage, indicating the output voltage value set by the repair charging strategy. is the internal resistance correction coefficient, which indicates the sensitivity of the internal resistance change to the output voltage adjustment. is the internal resistance deviation, which indicates the difference between the current internal resistance and the target internal resistance; S500.

2. Detect abnormality in the charging control process; When an abnormality is detected, the repair process is stopped immediately and the system switches to safe mode, reducing the charging power or completely disconnecting the output. S500.

3. Conduct a preliminary assessment of the restored battery; The health status parameters after recovery are compared with the health status parameters before recovery.

6. A method for charging an electric locomotive battery according to claim 1, characterized in that: The S600 is specifically: S600.

1. Extended evaluation of battery recovery effects; The extended assessment includes: life improvement and capacity recovery rate; S600.

2. Generate a repair effect report and output it in an integrated manner; The repair effect report includes: the comparison of SOH, RUL and internal resistance change curves before and after repair, and the matching degree between the fault type and the repair solution; The obtained first report data, second report data and repair effect report are integrated, the first report data is ranked first, the second report data is ranked second, and the repair effect report is ranked third, and they are output to the cloud in the form of word documents and PDF files.

7. A charging device for an electric locomotive battery, using a charging method for an electric locomotive battery as claimed in any one of claims 1 to 6, characterized in that: The modules include: A parameter monitoring module is used to obtain and convert battery operating parameters, perform preliminary processing based on the converted battery operating parameters, monitor the battery operating parameters based on the preliminary processed parameters, and obtain health status and life prediction results; a cleaning and extraction module, used for performing data cleaning and feature extraction based on the battery operation parameters after preliminary processing, obtaining voltage change parameters, temperature gradient parameters and internal resistance change parameters, calculating the battery health state based on the obtained health state, obtaining health state parameters, calculating the remaining service life based on the obtained life prediction result, obtaining life prediction parameters, performing abnormality detection and fault identification based on the obtained health state parameters and life prediction parameters, and generating a report based on the obtained health state parameters and life prediction parameters to obtain first report data; A health analysis module is used to perform fault diagnosis based on the obtained health status parameters, life prediction parameters, voltage change parameters, temperature gradient parameters and internal resistance change parameters, obtain lithium dendrite data, electrolyte data and active material data, and perform repairability classification based on the obtained lithium dendrite data, electrolyte data, active material data, health status parameters and life prediction parameters to obtain repairable data and non-repairable data; Repairable means: repairing by high-frequency pulse current according to lithium dendrite data, and regenerating electrolyte by long-term charging with low current according to electrolyte data, to obtain repairable data; Irreparable means: large-scale consumption of active materials and internal short circuits lead to thermal runaway, resulting in irreparable data; Defining and reporting high-risk failures based on the obtained repairable data and unrepairable data to obtain second report data; A life assessment module is used to generate a repair strategy based on the obtained lithium dendrite data, electrolyte data, active material data, repairable data and non-repairable data. The repair strategy includes: lithium dendrite repair, electrolyte aging repair and material wear repair; Lithium dendrite repair: Dissolve early lithium dendrites through high-frequency pulsed current while limiting heat generation; Electrolyte aging repair: Use low current and long-term constant voltage charging to restore the chemical activity of the electrolyte; Material depletion repair: Activate active materials in edge areas through step-by-step current charging; The SOC parameter data is monitored through the instrument panel during the repair charging process. If the SOC is close to the upper or lower limit, the charging strategy is dynamically adjusted, including: when the SOC is close to the upper limit, the charging current is reduced, and when the SOC is close to the lower limit, the discharge is limited; Optimizing repair parameters based on the generated repair strategy; Genetic algorithm is used to optimize the repair parameters, specifically: ； Where: is the optimization objective function, which means maximizing the improvement of health status per unit time while minimizing power loss. is the improvement in health status, indicating the improvement in health status after repair. is the repair time, which indicates the length of time it takes to complete the charging repair. is the power loss, which indicates the power consumed during the repair process, is the power loss weight factor, which is used to balance the relationship between health status improvement and power consumption; Anomaly detection module, which is used to control charging based on the optimized repair strategy, detect anomalies in the charging control process, and conduct preliminary evaluation of the recovered battery; The charging optimization module is used to perform an extended evaluation of the battery recovery effect, generate a repair effect report, and perform integrated output.

Images