A lead-acid battery health assessment method and system based on simulated operating conditions

Abstract

This invention belongs to the field of battery state detection technology. It discloses a method and system for health assessment of lead-acid batteries based on simulated operating conditions. The method includes: collecting simulated operating data of lead-acid batteries and cleaning the data to obtain high-quality battery data; filtering recovery effect data and removing it from the high-quality battery data to obtain disturbance-removed battery data; performing polarization voltage compensation to generate voltage-compensated battery data; performing abnormal impact identification and filtering abnormal data from the voltage-compensated battery data based on the identification results to output reasonable battery data; identifying long-cycle data and performing risk assessment to obtain corrected battery data and overlaying corresponding data from the reasonable battery data to obtain differentiated battery data; and assessing the health status of the differentiated battery data to output lead-acid battery health indicators. This significantly improves the accuracy of assessing the true health level of batteries under simulated operating conditions.

Description

Technical Field

[0001] This invention relates to the field of battery state detection technology, and more specifically, to a method and system for health assessment of lead-acid batteries based on simulated operating conditions. Background Technology

[0002] Currently, lead-acid batteries are widely used in many key industries due to their advantages such as low cost, high reliability, and easy maintenance. In these application scenarios, the battery, as the core of energy storage, has a significant impact on the overall safety of the system due to its operational stability and health status. Therefore, high-precision battery health assessment methods are gradually becoming an important direction for optimizing battery management systems. However, traditional health assessment methods are gradually becoming unable to meet the robustness and accuracy requirements of actual operation. Therefore, there is an urgent need for a health assessment method that is both accurate and versatile.

[0003] Traditional health assessment methods generally fail to consider the recovery effect of batteries after deep discharge. This recovery effect refers to the short-term voltage rebound due to ion diffusion and redistribution. The presence of this recovery effect can easily lead to biases in subsequent health status assessments. For example, in data acquisition from a deep discharge simulation scenario for a certain type of battery pack, a recovery effect occurs after discharge, and the voltage rebound during this period differs from the actual voltage, resulting in assessment errors in subsequent health status evaluations. Furthermore, traditional health assessment methods often use an ideal step current load as the estimation trigger condition. However, in reality, the current behavior during startup of equipment such as UPS systems, IDC data centers, and elevators is often characterized by a ramp load or a non-linear increase. During such loading processes, traditional health assessment methods do not consider the load curvature. Changes in line slope can lead to response delays and amplitude distortions in polarization voltage, further causing misjudgments in assessments. Furthermore, sudden short-term reverse currents or transient oscillations in actual operating conditions are a major blind spot in traditional health assessment methods. For example, a millisecond-level backflow current occurred during a communication base station handover, causing spike errors in the voltage data. Failure to remove such jumps could directly lead to misjudgments of the health status. In addition, for a large number of batteries in float charging mode, prolonged static storage can cause sulfation and degradation of the plates. Traditional health assessment methods often lack consideration for sulfation and degradation, easily mistaking extreme stability for normal phenomena, thus ignoring the possibility of degradation in the battery's basic functions, resulting in significant deviations in the final health status assessment.

[0004] In view of this, the present invention proposes a health assessment method and system for lead-acid batteries based on simulated operating conditions to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a lead-acid battery health assessment method based on simulated operating conditions, comprising:

[0006] S1. Collect simulated operation data of lead-acid batteries and perform data cleaning to obtain high-quality battery data;

[0007] S2. Perform recovery effect analysis on high-quality battery data to obtain recovery effect data and remove it from the high-quality battery data to obtain perturbation-removed battery data;

[0008] S3. Perform polarization voltage compensation on the disturbance-removed battery data to generate voltage-compensated battery data;

[0009] S4. Perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data;

[0010] S5. Identify long-cycle data in reasonable battery data, perform risk assessment on long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data;

[0011] S6. Perform a health status assessment on the differentiated battery data, output the lead-acid battery health indicators, and send them to the preset battery management terminal.

[0012] Furthermore, the methods for conducting the recovery effect analysis include:

[0013] Extract voltage and current data from high-quality battery data and construct voltage and current value sequences respectively; calculate the absolute value of the current difference between any two sampling points in the current value sequence, and identify the time interval where the absolute value of the current difference is less than a preset current amplitude threshold as the non-load interval; calculate the rate of change of each sampling point in the voltage value sequence and construct a voltage rate of change sequence, and identify the non-load voltage time interval corresponding to the non-load interval in the voltage rate of change sequence.

[0014] A dynamic sliding window is constructed to traverse the non-load voltage time interval. If there is a sub-segment within the dynamic sliding window where the slope of consecutive sampling points is greater than a preset rate of change slope threshold and the time length of the sub-segment is greater than a preset time length threshold, then the time interval corresponding to the dynamic sliding window is determined as a candidate recovery effect segment. The standard deviation of the voltage change rate of the candidate recovery effect segment is calculated. If the standard deviation of the voltage change rate is less than a preset stability standard deviation threshold, then the corresponding candidate recovery effect segment is determined as a recovery effect segment. The recovery effect segments are indexed and marked, and corresponding data in high-quality battery data are filtered based on the markings to obtain recovery effect data.

[0015] Furthermore, the method of indexing includes:

[0016] Extend a predetermined number of sampling points towards the starting and ending boundaries of each recovery effect segment to obtain an extended recovery effect segment. Extract the minimum voltage change rate slope in the extended recovery effect segment. If the minimum voltage change rate slope is lower than a preset continuous slope threshold, gradually shrink the extended starting and ending boundaries of the extended recovery effect segment in opposite directions until the minimum voltage change rate slope in the extended recovery effect segment is not lower than the preset continuous slope threshold or the segment length is less than a preset minimum segment length. Mark the sampling points corresponding to the extended starting and ending boundaries of all adjusted extended recovery effect segments. Based on the markings, identify the time segment corresponding to each adjusted extended recovery effect segment as the recovery effect time segment.

[0017] Furthermore, the method for performing polarization voltage compensation includes:

[0018] The time segment in which the absolute value of the current difference in the current value sequence is greater than the preset current amplitude threshold is identified as the original load application segment; the original load application segment is matched with the disturbance removal battery data. If there is no time segment corresponding to the original load application segment in the disturbance removal battery data, no processing is performed. If there is a complete time segment or a partial time segment corresponding to the original load application segment, the original load application segment is determined as the load application segment.

[0019] Calculate the rate of change of current in the load application section and construct a current rate of change sequence; identify the section in the current rate of change sequence where the rate of change of current increases continuously as the current increase section; extract the disturbance-free voltage value sequence corresponding to the current increase section and calculate the maximum slope of the disturbance-free voltage value sequence; extract the voltage rate of change sequence corresponding to the section that reaches the standard current steady state after the current increase section to obtain the steady-state voltage sequence; calculate the average rate of change of the steady-state voltage sequence to obtain the reference voltage response rate; calculate the difference between the maximum slope of the steady-state sequence and the reference voltage response rate and take the absolute value to obtain the polarization voltage response deviation;

[0020] A voltage compensation factor is constructed based on the polarization voltage response deviation. The voltage compensation factor is used to weight the de-disturbed voltage value sequence, and the weighted de-disturbed voltage value sequence is used to cover the corresponding data in the de-disturbed battery data to obtain voltage-compensated battery data.

[0021] Furthermore, the method for constructing the voltage compensation factor includes:

[0022] The polarization voltage response deviation is normalized to obtain the normalized slope deviation; the standard deviation of the slope in the undisturbed voltage value sequence is calculated to obtain the stability fluctuation degree; the ratio of the stability fluctuation degree to the reference voltage response rate is calculated to obtain the response stability ratio; the mean slope of the current increase segment is calculated, and the ratio of this mean slope to the maximum slope in the current increase segment is calculated to obtain the dynamic adjustment ratio; the product of the normalized slope deviation, the response stability ratio, and the dynamic adjustment ratio is calculated to obtain the initial compensation factor; the time interval from the last sampling point in the current increase segment to reaching the standard current steady state is measured, and the time adjustment factor is constructed based on this time interval; the product of the initial compensation factor and the time adjustment factor is calculated to obtain the voltage compensation factor.

[0023] Furthermore, the method for performing abnormal impact identification includes:

[0024] The voltage-compensated battery data is divided into windows based on a preset fixed recognition time window to obtain recognition window data; a local voltage value sequence and a local current value sequence belonging to the window are constructed based on the recognition window data; the local fluctuation amplitude of the maximum local voltage value and the minimum local voltage value in the local voltage value sequence is calculated; if the local fluctuation amplitude is higher than the preset peak voltage threshold, the recognition window data corresponding to the local voltage value sequence is determined to be a voltage change risk window.

[0025] Determine the direction of current change within each half-window before and after the local current value sequence, and simultaneously calculate the local current difference between the maximum and minimum local current values ​​within any half-window; if the direction of current change within each half-window is opposite and the local current difference is higher than the preset reverse current judgment value, then the identification window data corresponding to the local current value sequence is determined as a current direction change window.

[0026] Based on the voltage surge risk window, trend fitting is performed on all sampling points to identify trend disturbance windows; if any two types of windows, including the voltage surge risk window, the current direction surge window, and the trend disturbance window, belong to the same window, then the window is determined to be an abnormal impact window.

[0027] Furthermore, the method for performing trend fitting includes:

[0028] A linear fitting trend curve is plotted based on the local voltage value sequence corresponding to the voltage mutation risk window; the voltage deviation of the sampling points of the corresponding window relative to the linear fitting trend curve is calculated, and a fitting deviation sequence is constructed based on the voltage deviation; the variance of the fitting deviation sequence is calculated, and if the variance of the variance is greater than the preset trend disturbance judgment threshold, the window corresponding to the fitting deviation sequence is judged as a suspected trend disturbance window; the segment where the voltage deviation of consecutive sampling points in the suspected trend disturbance window is higher than twice the average voltage deviation is identified, and if the number of consecutive sampling points in the segment is higher than the lower limit of the peak time, the corresponding suspected trend disturbance window is determined as a trend disturbance window.

[0029] Furthermore, the methods for risk assessment include:

[0030] Traverse the time index of all sampling points in the reasonable battery data, extract the data segment whose continuous time length is longer than the preset resting time limit, and take the data segment as long period data; if the current value of the long period data is within the preset current float charging range, and the absolute value of the difference between the voltage value of the long period data and the preset float charging voltage threshold is not higher than the preset difference, then the corresponding long period data is determined as float charging state data; based on the timestamp record of the sampling point in the float charging state data, divide the float charging state data into unequal periods to obtain local resting period float charging data;

[0031] Calculate the voltage standard deviation of any local static period float charge data. If the duration of the local static period float charge data is higher than the preset static period threshold and the voltage standard deviation is lower than the preset fluctuation threshold, then the data segment corresponding to the local static period float charge data is identified as a potential sulfurization risk segment.

[0032] Calculate the float charge voltage amplitude of the potential sulfation risk section and the ratio of this float charge voltage amplitude to the preset standard stable voltage to obtain the float charge voltage stability ratio; calculate the output current variation amplitude and average output current value of the previous data section of the potential sulfation risk section; linearly combine the float charge voltage stability ratio, the output current variation amplitude and average output current value of the previous data section according to a preset ratio to obtain the health weighting coefficient; use the health weighting coefficient to weight the specific data of the potential sulfation risk section to obtain the corrected battery data.

[0033] Furthermore, the methods for conducting the health status assessment include:

[0034] According to the health status assessment requirements, parameters of the corresponding dimensions are extracted from the differentiated battery data, and an assessment input dataset is constructed. The historical assessment template is matched with the assessment input dataset to output the health score of each dimension parameter. The health score is matched with the preset health score range to output the health level of the corresponding dimension parameter. The health levels of all dimension parameters are integrated to obtain the lead-acid battery health index.

[0035] A health assessment system for lead-acid batteries based on simulated operating conditions, used to implement a health assessment method for lead-acid batteries based on simulated operating conditions, characterized by comprising:

[0036] The data acquisition module is used to collect simulated operating data of lead-acid batteries and perform data cleaning to obtain high-quality battery data.

[0037] The effect analysis module is used to perform recovery effect analysis on high-quality battery data, obtain recovery effect data, and remove it from the high-quality battery data to obtain perturbation-removed battery data.

[0038] The voltage compensation module is used to perform polarization voltage compensation on the disturbance removal battery data to generate voltage-compensated battery data.

[0039] The anomaly identification module is used to perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data.

[0040] The risk assessment module is used to identify long-cycle data in reasonable battery data, assess the risk of long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data.

[0041] The health assessment module is used to assess the health status of differentiated battery data, output lead-acid battery health indicators and send them to the preset battery management terminal; the modules are connected to each other via wired and / or wireless means.

[0042] The technical effects and advantages of the health assessment method and system for lead-acid batteries based on simulated operating conditions proposed in this invention are as follows:

[0043] By performing multi-stage, layered processing on the collected simulated operating data of lead-acid batteries and conducting health status assessments based on the processed data, a more accurate health assessment method for lead-acid batteries based on simulated operating conditions was achieved. Compared with existing experience, a recovery effect identification and elimination mechanism was used to screen out voltage rebound phenomena in the original data for analysis, accurately identifying and directly eliminating data with recovery effects, thus avoiding the misinterpretation of recovery effects as normal discharge states. Furthermore, precise load identification was introduced to determine and compensate for the interference intensity in polarization interference sections, ensuring that subsequent health status assessments accurately reflect real load responses. The adaptability of the method indirectly improves the assessment accuracy. By constructing voltage mutation windows, current direction mutation windows, and trend disturbance windows, the method enables real-time identification and cleanup of typical abnormal impact behaviors, avoiding the impact of abnormal operating conditions on the assessment accuracy. By identifying potential sulfation risk sections and constructing health weight coefficients, the method marks and weights these sections, effectively preventing static data from being misjudged as high health values ​​in subsequent health status assessments, thus improving the assessment process's ability to perceive battery aging risks. Therefore, the above-mentioned health status assessment method significantly improves the accuracy of assessing the true health level of the battery under simulated operating conditions. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of a lead-acid battery health assessment method based on simulated working conditions according to the present invention.

[0045] Figure 2 This is a schematic diagram of a lead-acid battery health assessment system based on simulated operating conditions according to the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Example 1

[0048] Please see Figure 1 As shown in this embodiment, a lead-acid battery health assessment method based on simulated operating conditions includes:

[0049] S1. Collect simulated operation data of lead-acid batteries and perform data cleaning to obtain high-quality battery data;

[0050] S2. Perform recovery effect analysis on high-quality battery data to obtain recovery effect data and remove it from the high-quality battery data to obtain perturbation-removed battery data;

[0051] S3. Perform polarization voltage compensation on the disturbance-removed battery data to generate voltage-compensated battery data;

[0052] S4. Perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data;

[0053] S5. Identify long-cycle data in reasonable battery data, perform risk assessment on long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data;

[0054] S6. Perform a health status assessment on the differentiated battery data, output the lead-acid battery health indicators, and send them to the preset battery management terminal.

[0055] The simulated operation data of lead-acid batteries includes data such as voltage, current and corresponding timestamps that can represent the operating status of the battery in the simulated scenario. By performing missing value completion and data filtering on the simulated operation data of lead-acid batteries, data cleaning is achieved to obtain higher quality battery data.

[0056] Methods for conducting recovery effect analysis include:

[0057] Voltage and current data are extracted from high-quality battery data, and voltage and current value sequences are constructed respectively. The voltage and current value sequences are sorted according to the timestamp order of the corresponding sampling points to obtain a continuous and ordered dataset, which ensures time consistency and point-to-point correspondence.

[0058] The absolute value of the current difference between any two adjacent sampling points in the current value sequence is calculated. Time intervals in which the absolute value of the current difference is less than a preset current amplitude threshold are identified as non-load intervals. The preset current amplitude threshold is set based on historical simulation experience. If the absolute value of the current difference is less than the preset current amplitude threshold, it means that the lead-acid battery is not in a significant loading or unloading process in the time interval corresponding to these continuous sampling points. Therefore, the corresponding time interval is determined to be a non-load interval. In this embodiment, "continuous" means that at least 5 sampling points are required to form an interval.

[0059] The rate of change of each sampling point in the voltage value sequence is calculated and a voltage rate of change sequence is constructed. The non-load voltage time segment corresponding to the non-load segment in the voltage rate of change sequence is identified. The rate of change refers to the slope between any two adjacent sampling points in the voltage value sequence. The slope is the ratio of the difference in voltage value between adjacent sampling points to the time length. The slope of each sampling point is sorted according to the corresponding timestamp order to obtain the voltage rate of change sequence. At the same time, the non-load voltage time segment corresponding to the non-load segment is identified in the voltage rate of change sequence.

[0060] A dynamic sliding window is constructed to traverse the non-load voltage time interval. If there is a sub-segment within the dynamic sliding window where the slope of continuous sampling points is greater than a preset rate of change slope threshold and the time length of the sub-segment is greater than a preset time length threshold, then the time interval corresponding to the dynamic sliding window is determined as a candidate recovery effect segment. It should be noted that the slope here is calculated in the same way as the rate of change, and is only used to represent the slope of continuous data in mathematical terms. The rate of change mentioned above is not named after the slope in order to reflect that the value is used to represent the change of voltage value.

[0061] Based on historical simulation experience, preset thresholds for the rate of change slope and time length are set. If five or more consecutive sampling points in the dynamic sliding window have a slope higher than the preset rate of change slope threshold, it indicates that the rate of change in the sub-segment formed by the corresponding consecutive sampling points is relatively fast. At the same time, the time length of the sub-segment is higher than the preset time length threshold. If both conditions are met, the time segment in which the dynamic sliding window is located can be determined to be a candidate segment where the voltage shows a recovery rise, and therefore it is determined to be a candidate recovery effect segment. In this embodiment, the window size of the dynamic sliding window can be dynamically adjusted based on historical simulation experience or specific simulation conditions.

[0062] The standard deviation of the voltage change rate of candidate recovery effect segments is calculated. If the standard deviation of the voltage change rate is less than the preset stability standard deviation threshold, the corresponding candidate recovery effect segment is determined as a recovery effect segment. The standard deviation of the voltage change rate is used to determine the recovery effect segment. When the standard deviation of the voltage change rate is less than the preset stability standard deviation threshold set based on historical simulation experience, it indicates that the voltage rise trend is stable and there is no high-frequency disturbance. Therefore, it can be officially confirmed as a segment where a recovery effect has occurred. Conversely, if the standard deviation of the voltage change rate is large, it may be a false identification caused by invalid voltage fluctuations, thus improving the reliability of the screening of recovery effect segments.

[0063] The recovery effect segment is indexed and marked, and the corresponding data in the high-quality battery data is filtered based on the marking to obtain the recovery effect data. The data at the time position corresponding to the recovery effect time segment is determined to be the recovery effect data. The specific parameters and timestamps corresponding to the recovery effect data are removed from the high-quality battery data. The adjusted high-quality battery data is used as the perturbation to remove battery data.

[0064] The methods for indexing include:

[0065] By extending a predetermined number of sampling points in the starting and ending boundary directions of each recovery effect segment, an extended recovery effect segment is obtained. In this embodiment, extending a predetermined number of sampling points in the starting and ending boundary directions of the recovery effect segment is used to cover the premature rise and delayed fall phenomena that may occur during battery operation. Since the electrochemical recovery process is not instantaneous initiation and termination, there may be partial recovery features outside the judgment range. By extending the predetermined number of sampling points based on historical simulation experience, the accuracy of recovery effect identification is improved.

[0066] Extract the slope of the minimum voltage change rate in the extended recovery effect segment. If the slope of the minimum voltage change rate is lower than the preset continuous slope threshold, then gradually shrink the extension start boundary and extension termination boundary of the extended recovery effect segment in opposite directions until the slope of the minimum voltage change rate in the extended recovery effect segment is not lower than the preset continuous slope threshold or the segment length is less than the preset minimum segment length.

[0067] The minimum voltage change rate slope refers to the slope of the minimum voltage change rate in the extended recovery effect section. To ensure the continuity and stability of the voltage rise in the extended recovery effect section, a preset continuous slope threshold is set based on historical simulation experience. If the minimum voltage change rate slope is less than the preset continuous slope threshold, it indicates that there are voltage suppression or non-recovery disturbances in this section, which may lead to discontinuity and make it impossible to completely determine whether the recovery effect has occurred normally. Therefore, the corresponding extended recovery effect section needs to be adjusted.

[0068] In this embodiment, the expansion start boundary is shrunk towards the expansion end boundary, and the expansion end boundary is shrunk towards the expansion start boundary, until the minimum voltage change rate slope is not lower than a preset continuous slope threshold or the segment length is less than a preset minimum segment length set based on historical simulation experience. This avoids including invalid data for judging the recovery effect into the data that needs to be processed. It should be noted that the expansion recovery effect segment with a minimum voltage change rate slope greater than or equal to the preset continuous slope threshold is not processed and can be directly marked.

[0069] The sampling points corresponding to the expansion start boundary and expansion end boundary of all adjusted extended recovery effect segments are marked. Based on the marking, the time segment corresponding to each adjusted extended recovery effect segment is identified as the recovery effect time segment. Specifically, by marking the sampling points corresponding to the expansion start boundary and expansion end boundary of the adjusted extended recovery effect segment, the time segment corresponding to the extended recovery effect segment is detected based on the marking, and these time segments are used as the recovery effect time segments.

[0070] Methods for polarization voltage compensation include:

[0071] The time intervals in the current value sequence where the absolute value of the current difference is greater than the preset current amplitude threshold are identified as the original load application intervals. This is achieved by calculating the absolute value of the current difference between two adjacent sampling points in the current value sequence point by point, and filtering out continuous time intervals where the absolute value of the current difference is greater than the preset current amplitude threshold set based on historical simulation experience. Since the current in these time intervals shows a continuous and rapid increase, they can be identified as the original load application intervals.

[0072] The original load application segment is matched with the disturbance removal battery data. If the disturbance removal battery data does not contain a time segment corresponding to the original load application segment, no processing is performed. If a complete or partial time segment corresponding to the original load application segment exists, the original load application segment is determined to be the load application segment.

[0073] The process involves matching the original load application segment with the time index of the disturbance-removed battery data to determine if it exists in the disturbance-removed battery data. Since some data was removed from the high-quality battery data in step S2, and the current value sequence is constructed based on the high-quality battery data, the time segment corresponding to the original load application segment may not exist in the disturbance-removed battery data. If a complete or partial time segment corresponding to the original load application segment exists, the original load application segment can be identified as a load application segment. The specific time length threshold for the "partial time segment" is set based on historical records. If the time length is lower than the specific time length threshold for the "partial time segment", no processing is performed.

[0074] The current change rate of the load application section is calculated and a current change rate sequence is constructed. Since there may only be partial data for the load application section, the corresponding parameters are extracted from the current value sequence of high-quality battery data and temporarily filled during the polarization voltage compensation process. The current change rate is sorted according to the timestamp of the corresponding sampling point to obtain the current change rate sequence.

[0075] The segment in the current rate of change sequence that continuously increases in current rate of change is identified as the current increase segment. If the current rate of change of consecutive sampling points in the current rate of change sequence continuously increases and shows positive growth, the corresponding segment is identified as the current increase segment.

[0076] Extract the perturbation voltage value sequence corresponding to the current increase segment, and calculate the maximum change slope of the perturbation voltage value sequence. The perturbation voltage value sequence refers to the sequence of voltage values ​​constructed in the current increase segment, and the maximum change slope refers to the maximum slope in the perturbation voltage value sequence, which reflects the degree of voltage change of the battery when facing rapid load changes.

[0077] The voltage change rate sequence corresponding to the segment that reaches the standard current steady state after the current increases is extracted to obtain the steady-state voltage sequence. The standard current steady state is a current stability target value set based on existing theoretical knowledge and historical experience. When the current value in the current increase segment reaches the standard current steady state and begins to be maintained, the sampling point that first reaches the standard current steady state is taken as the starting point, and the corresponding sampling points are extracted point by point until the current value is lower than the standard current steady state. Based on the segment formed by these sampling points, the voltage change rate sequence corresponding to this segment is constructed. This sequence is the steady-state voltage sequence.

[0078] The average rate of change of the steady-state voltage sequence is calculated to obtain the reference voltage response rate. The difference between the maximum slope of the steady-state sequence and the reference voltage response rate is calculated and its absolute value is taken to obtain the polarization voltage response deviation. The maximum slope of the steady-state sequence refers to the maximum slope in the steady-state voltage sequence. The absolute value of the difference between the maximum slope of the steady-state sequence and the reference voltage response rate reflects the disturbance to the voltage during the load loading process.

[0079] A voltage compensation factor is constructed based on the polarization voltage response deviation. The voltage compensation factor is used to multiply each specific value in the de-disturbed voltage value sequence to achieve weighting. The weighted de-disturbed voltage value sequence is then used to cover the corresponding data in the perturbed battery data to obtain voltage-compensated battery data. The weighted de-disturbed voltage value sequence is used to cover the voltage values ​​in the perturbed battery data. If the voltage value of a sampling point that has already been removed is removed here as well.

[0080] Methods for constructing voltage compensation factors include:

[0081] The polarization voltage response deviation is normalized to obtain the normalized slope deviation. This is achieved by calculating the ratio of the polarization voltage response deviation to a reference deviation threshold set based on historical simulation experience, and then converting the result into a value within the normalized range.

[0082] The standard deviation of the slope in the perturbation voltage value sequence is calculated to obtain the stability fluctuation degree. The stability fluctuation degree quantifies the degree of voltage response jitter during the load stage. The larger the value, the greater the influence of other factors on the polarization voltage.

[0083] The response stability ratio is obtained by calculating the ratio of the stability fluctuation degree to the reference voltage response rate. The response stability ratio quantifies the actual instability degree in the current voltage polarization process. The larger the value, the more the response state deviates from the ideal state.

[0084] The average slope of the current-increasing section is calculated, and the ratio of this average slope to the maximum slope within the current-increasing section is calculated to obtain the dynamic adjustment ratio. The average slope is used to reflect the typical speed of load application, while the ratio of the average slope to the maximum slope within the current-increasing section reflects the degree of abrupt change in the load ramp-up process, which is used to quantify the influence of load intensity.

[0085] The initial compensation factor is obtained by multiplying the normalized slope deviation, the response stability ratio, and the dynamic adjustment ratio. The initial compensation factor reflects the combined influence of the three factors: offset, unstable response, and rapid loading.

[0086] The time interval from the last sampling point in the current increase section to the attainment of the standard current steady state is measured. A time adjustment factor is constructed based on this time interval, and the formula for calculating the time adjustment factor is as follows: ;in, Indicates the time adjustment factor; This indicates selecting the minimum value of the variable within the parentheses; This represents the time interval from the last sampling point in the current-increasing section to the point where the standard current steady state is reached; This represents the maximum allowable response time set based on historical simulation experience. If the time interval from the last sampling point in the current increase section to reaching the standard current steady state is greater than the maximum allowable response time, then a constant of 1 is directly selected as the value of the time adjustment factor to avoid over-control. The voltage compensation factor is obtained by calculating the product of the initial compensation factor and the time adjustment factor.

[0087] Methods for performing abnormal impact identification include:

[0088] The voltage compensation battery data is divided into windows based on a preset fixed recognition time length window to obtain recognition window data. The window size of the preset fixed recognition time length window is set based on historical simulation experience and can be adjusted according to specific working conditions. Using this preset fixed recognition time length window, the voltage compensation battery data is divided into several data segments of equal length according to the time length, and the data in any one of these windows is the recognition window data.

[0089] Based on the identification window data, a local voltage value sequence and a local current value sequence belonging to the window are constructed. Specifically, the local voltage value sequence and the local current value sequence belonging to the window are constructed by extracting the current value, voltage value and corresponding timestamp from each identification window data, respectively, in order of time. This is used to more accurately identify data fluctuations within a local time period.

[0090] The local fluctuation amplitude of the maximum and minimum local voltage values ​​in the local voltage value sequence is calculated. If the local fluctuation amplitude is higher than the preset peak voltage threshold, the identification window data corresponding to the local voltage value sequence is determined to be a voltage surge risk window. The preset peak voltage threshold is set based on existing electrical theory knowledge. By calculating the local fluctuation amplitude and comparing it with the preset peak voltage threshold, if the amplitude is higher than the preset peak voltage threshold, it indicates that the window where the corresponding identification window data is located belongs to the voltage surge risk window, and there may be instantaneous voltage surge or sag.

[0091] The system determines the direction of current change within each half-window of the local current value sequence, and simultaneously calculates the local current difference between the maximum and minimum local current values ​​within any half-window. Each local current value sequence is divided into two sub-windows: the first half and the second half. The system identifies the number of times the current value increases and decreases within each sub-window; if the number of increases exceeds the number of decreases, the current change is considered positive; otherwise, it is considered negative. The local current difference is calculated to reflect the magnitude of current change over a short period.

[0092] If the current change direction is opposite in each half-window and the local current difference is higher than the preset reverse current judgment value, then the identification window data corresponding to the local current value sequence is judged as a current direction change window. In this embodiment, the preset reverse current judgment value is set based on existing electrical theory knowledge and historical simulation experience. If the current change direction is opposite in each half-window and the local current difference is higher than the preset reverse current judgment value, it indicates that a sudden change in load direction may occur at this time. The window to which the corresponding identification window data belongs is judged as a current direction change window.

[0093] Based on the voltage surge risk window, trend fitting is performed on all sampling points to identify trend disturbance windows. If any two of the voltage surge risk window, current direction surge window, and trend disturbance window belong to the same window, then the window is identified as an abnormal impact window. As long as any two of the above three conditions are met, the window that meets the conditions can be identified as an abnormal impact window, avoiding misjudgment that may be caused by a single condition. The data corresponding to the abnormal impact window is removed from the voltage compensation battery data to obtain reasonable battery data.

[0094] Methods for trend fitting include:

[0095] A linear fitting trend curve is plotted based on the local voltage value sequence corresponding to the voltage mutation risk window. In this embodiment, the least squares linear fitting is used to plot all the sampling point data of the local voltage value sequence corresponding to the voltage mutation risk window as a linear fitting trend curve. This curve shows the overall trend of voltage in the above local voltage value sequence.

[0096] The voltage deviation of the sampling points in the corresponding window relative to the linear fitting trend curve is calculated. Based on the voltage deviation, a fitting deviation sequence is constructed. The deviation between the predicted voltage value in the linear fitting trend curve and the actual voltage value of the corresponding sampling point is calculated to obtain the voltage deviation value. The voltage deviation values ​​are sorted according to the timestamp of the corresponding sampling point to obtain the fitting deviation sequence, which quantifies the degree of fluctuation of the actual voltage around the linear fitting trend curve within the window.

[0097] The variance of the fitted deviation sequence is calculated. If the variance of ...

[0098] Identify segments within a suspected trend disturbance window where the voltage deviation of consecutive sampling points exceeds twice the average voltage deviation. If the number of consecutive sampling points in such a segment exceeds the lower limit of the peak time, the corresponding suspected trend disturbance window is determined as a trend disturbance window. Further iterate through all sampling points within the suspected trend disturbance window. If there are segments where the voltage deviation of consecutive sampling points exceeds twice the average voltage deviation, and the time length corresponding to the number of consecutive sampling points in such segments exceeds the lower limit of the peak time set based on historical simulation experience, then it can be determined that the disturbance level of the current suspected trend disturbance window has sufficient instability, and therefore, the suspected trend disturbance window is determined as a trend disturbance window.

[0099] Methods for risk assessment include:

[0100] Traverse the time index of all sampling points in the reasonable battery data, extract the data segment whose continuous time length is higher than the preset resting time limit, and regard the data segment as long-cycle data. The continuous time length refers to the time length of the time segment composed of consecutive sampling points. The preset resting time limit is set based on historical simulation experience and specific load conditions. If the continuous time length is higher than the preset resting time limit, it means that the duration is too long, and the corresponding data segment is determined to be long-cycle data.

[0101] If the current values ​​of the long-cycle data are all within the preset current float charging range, and the absolute value of the difference between the voltage value of the long-cycle data and the preset float charging voltage threshold is not higher than the preset difference, then the corresponding long-cycle data is determined to be float charging state data. In this embodiment, the preset current float charging range and the preset float charging voltage threshold are set based on existing electrical theory knowledge, and the preset difference is set based on historical simulation experience. At the same time, it is determined whether the current value of the long-cycle data is within the preset current float charging range and whether the tolerance between the voltage value and the preset float charging voltage threshold is less than or equal to the preset difference. If both conditions are met, it means that in the state of the corresponding long-cycle data, the battery has not been charged or discharged for a long time and only maintains charge balance with extremely low current. Therefore, it is determined to be float charging state data.

[0102] Based on the timestamp records of the sampling points in the float charge state data, the float charge state data is divided into unequal periods to obtain local static period float charge data. The unequal period division means that, based on the differences in the timestamps in the float charge state data, the segments with time intervals that are significantly higher than the upper limit of the float charge time interval set based on historical simulation experience are independently divided into several sub-data segments, which are the local static period float charge data.

[0103] Calculate the voltage standard deviation of any local static period float charge data. If the duration of the local static period float charge data is higher than the preset static period threshold and the voltage standard deviation is lower than the preset fluctuation threshold, then the data segment corresponding to the local static period float charge data is identified as a potential sulfidation risk segment. The preset static period threshold and preset fluctuation threshold are set based on historical experience. Anomalies are further screened based on the local static period float charge data. If the local static period float charge data simultaneously meets the conditions that the corresponding duration is higher than the preset static period threshold and the voltage standard deviation is lower than the preset fluctuation threshold, then the electrochemical reaction in the data segment corresponding to the current local static period float charge data is considered to be slow, and therefore it is identified as a potential sulfidation risk segment.

[0104] The float charge voltage amplitude of the potential sulfidation risk zone is calculated, and the ratio of this float charge voltage amplitude to the preset standard stable voltage is calculated to obtain the float charge voltage stability ratio. The float charge voltage amplitude refers to the difference between the maximum and minimum voltage values ​​in the potential sulfidation risk zone. At the same time, a preset standard stable voltage is set based on existing electrical theory knowledge, and the ratio of the float charge voltage amplitude to the preset standard stable voltage is calculated. This value reflects the relative fluctuation intensity of the float charge state of the current potential sulfidation risk zone under the ideal standard float charge state.

[0105] Calculate the output current variation range and average output current value of the data segment preceding the potential sulfurization risk segment. The output current variation range refers to the difference between the maximum and minimum current values ​​in the data segment preceding the potential sulfurization risk segment, and the average output current value refers to the average value of all current values ​​in the aforementioned data segment.

[0106] The float charge voltage stability ratio, the output current variation amplitude of the previous data segment, and the average output current value are linearly combined according to a preset ratio to obtain the health weight coefficient. The health weight coefficient is obtained by weighting and summing the above three indicators according to a preset ratio based on historical simulation experience. This health weight coefficient is used to reflect the degree of potential risk.

[0107] By multiplying each specific data point in the potential sulfidation risk zone using a health weighting coefficient, corrected battery data is obtained. By weighting the specific data in the potential sulfidation risk zone using the health weighting coefficient, the contribution of this type of data in the health status assessment process is adjusted, indirectly reducing the misjudgment rate of health status assessment.

[0108] Methods for conducting health status assessments include:

[0109] According to the health status assessment requirements, parameters of the corresponding dimensions in the differentiated battery data are extracted, and the assessment input dataset is constructed. The health status assessment requirements refer to the health status assessment standards set based on historical assessment records. The parameters of the corresponding dimensions required for health status assessment are extracted from the differentiated battery data, and these parameters are used to form the assessment input dataset.

[0110] By matching the historical evaluation template with the evaluation input dataset, a health score for each dimension parameter is output. The historical evaluation template refers to a parameter mapping template composed of a large number of simulated data of healthy batteries under different operating conditions in the historical record. In this embodiment, a regression model trained on the historical evaluation template is used to receive the evaluation input dataset, and the parameters of each dimension are mapped to the historical evaluation template to output a health score for each dimension parameter.

[0111] The health score is matched with a preset health score range to output the health level of the corresponding dimension parameter. The health levels of all dimension parameters are integrated to obtain the lead-acid battery health index. The preset health score range is set based on a large amount of historical simulation experience. The health score is mapped to the preset health score range to determine the health level of the corresponding dimension parameter. The health levels of all parameters used for health status assessment are integrated into the lead-acid battery health index.

[0112] This embodiment achieves a more accurate health assessment method for lead-acid batteries based on simulated operating conditions by performing multi-stage, layered processing on the collected simulated operating data and conducting a health status assessment based on the processed data. Compared with existing experience, it uses a recovery effect identification and elimination mechanism to screen out voltage rebound phenomena in the original data for analysis, accurately identifying and directly eliminating data with recovery effects, avoiding the misinterpretation of recovery effects as normal discharge states. Furthermore, it introduces precise load identification, determines the interference intensity of polarization interference sections, and compensates for it, ensuring that subsequent health status assessments accurately reflect the actual load conditions. The adaptability of the load response indirectly improves the assessment accuracy; by constructing voltage mutation windows, current direction mutation windows, and trend disturbance windows, real-time identification and cleaning of typical abnormal impact behaviors are achieved, avoiding the impact of abnormal operating condition interference on the assessment accuracy; by identifying potential sulfation risk sections and constructing health weight coefficients, potential sulfation risk sections are marked and weighted to effectively avoid static data being misjudged as high health values ​​in subsequent health status assessment processes, thus improving the assessment process's ability to perceive battery aging risks; therefore, the above-mentioned health status assessment method significantly improves the accuracy of assessing the true health level of the battery under simulated operating conditions.

[0113] Example 2

[0114] Please see Figure 2 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A lead-acid battery health assessment system based on simulated operating conditions is provided, including:

[0115] The data acquisition module is used to collect simulated operating data of lead-acid batteries and perform data cleaning to obtain high-quality battery data.

[0116] The effect analysis module is used to perform recovery effect analysis on high-quality battery data, obtain recovery effect data, and remove it from the high-quality battery data to obtain perturbation-removed battery data.

[0117] The voltage compensation module is used to perform polarization voltage compensation on the disturbance removal battery data to generate voltage-compensated battery data.

[0118] The anomaly identification module is used to perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data.

[0119] The risk assessment module is used to identify long-cycle data in reasonable battery data, assess the risk of long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data.

[0120] The health assessment module is used to assess the health status of differentiated battery data, output lead-acid battery health indicators and send them to the preset battery management terminal; the modules are connected to each other via wired and / or wireless means.

[0121] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0122] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0123] In the description of this invention, it should be understood that the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0124] In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0125] In the description of this invention, "several" means one or more, and "a large number" means two or more.

[0126] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0127] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0128] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A health assessment method for lead-acid batteries based on simulated operating conditions, characterized in that, include: S1. Collect simulated operation data of lead-acid batteries and perform data cleaning to obtain high-quality battery data; S2. Perform recovery effect analysis on high-quality battery data to obtain recovery effect data and remove it from the high-quality battery data to obtain perturbation-removed battery data; S3. Perform polarization voltage compensation on the disturbance-removed battery data to generate voltage-compensated battery data; S4. Perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data; S5. Identify long-cycle data in reasonable battery data, perform risk assessment on long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data; S6. Perform a health status assessment on the differentiated battery data, output the lead-acid battery health indicators, and send them to the preset battery management terminal.

2. The method for health assessment of lead-acid batteries based on simulated operating conditions according to claim 1, characterized in that, The methods for performing recovery effect analysis include: Extract voltage and current data from high-quality battery data and construct voltage and current value sequences respectively; calculate the absolute value of the current difference between any two adjacent sampling points in the current value sequence, and identify the time segment where the absolute value of the current difference is less than a preset current amplitude threshold as the non-load segment; calculate the rate of change of each sampling point in the voltage value sequence and construct a voltage rate of change sequence, and identify the non-load voltage time segment corresponding to the non-load segment in the voltage rate of change sequence. A dynamic sliding window is constructed to traverse the non-load voltage time interval. If there is a sub-segment within the dynamic sliding window where the slope of consecutive sampling points is greater than a preset rate of change slope threshold and the time length of the sub-segment is greater than a preset time length threshold, then the time interval corresponding to the dynamic sliding window is determined as a candidate recovery effect segment. The standard deviation of the voltage change rate of the candidate recovery effect segment is calculated. If the standard deviation of the voltage change rate is less than a preset stability standard deviation threshold, then the corresponding candidate recovery effect segment is determined as a recovery effect segment. The recovery effect segments are indexed and marked, and corresponding data in high-quality battery data are filtered based on the markings to obtain recovery effect data.

3. The health assessment method for lead-acid batteries based on simulated operating conditions according to claim 2, characterized in that, The methods for performing index marking include: Extend a predetermined number of sampling points towards the starting and ending boundaries of each recovery effect segment to obtain an extended recovery effect segment. Extract the minimum voltage change rate slope in the extended recovery effect segment. If the minimum voltage change rate slope is lower than a preset continuous slope threshold, gradually shrink the extended starting and ending boundaries of the extended recovery effect segment in opposite directions until the minimum voltage change rate slope in the extended recovery effect segment is not lower than the preset continuous slope threshold or the segment length is less than a preset minimum segment length. Mark the sampling points corresponding to the extended starting and ending boundaries of all adjusted extended recovery effect segments. Based on the markings, identify the time segment corresponding to each adjusted extended recovery effect segment as the recovery effect time segment.

4. The health assessment method for lead-acid batteries based on simulated operating conditions according to claim 3, characterized in that, The methods for performing polarization voltage compensation include: The time segment in which the absolute value of the current difference in the current value sequence is greater than the preset current amplitude threshold is identified as the original load application segment; the original load application segment is matched with the disturbance removal battery data. If there is no time segment corresponding to the original load application segment in the disturbance removal battery data, no processing is performed. If there is a complete time segment or a partial time segment corresponding to the original load application segment, the original load application segment is determined as the load application segment. Calculate the rate of change of current in the load application section and construct a current rate of change sequence; identify the section in the current rate of change sequence where the rate of change of current increases continuously as the current increase section; extract the disturbance-free voltage value sequence corresponding to the current increase section and calculate the maximum slope of the disturbance-free voltage value sequence; extract the voltage rate of change sequence corresponding to the section that reaches the standard current steady state after the current increase section to obtain the steady-state voltage sequence; calculate the average rate of change of the steady-state voltage sequence to obtain the reference voltage response rate; calculate the difference between the maximum slope of the steady-state sequence and the reference voltage response rate and take the absolute value to obtain the polarization voltage response deviation; A voltage compensation factor is constructed based on the polarization voltage response deviation. The voltage compensation factor is used to weight the de-disturbed voltage value sequence, and the weighted de-disturbed voltage value sequence is used to cover the corresponding data in the de-disturbed battery data to obtain voltage-compensated battery data.

5. The health assessment method for lead-acid batteries based on simulated operating conditions according to claim 4, characterized in that, The methods for constructing the voltage compensation factor include: The polarization voltage response deviation is normalized to obtain the normalized slope deviation; the standard deviation of the slope in the undisturbed voltage value sequence is calculated to obtain the stability fluctuation degree; the ratio of the stability fluctuation degree to the reference voltage response rate is calculated to obtain the response stability ratio; the mean slope of the current increase segment is calculated, and the ratio of this mean slope to the maximum slope in the current increase segment is calculated to obtain the dynamic adjustment ratio; the product of the normalized slope deviation, the response stability ratio, and the dynamic adjustment ratio is calculated to obtain the initial compensation factor; the time interval from the last sampling point in the current increase segment to reaching the standard current steady state is measured, and the time adjustment factor is constructed based on this time interval; the product of the initial compensation factor and the time adjustment factor is calculated to obtain the voltage compensation factor.

6. The health assessment method for lead-acid batteries based on simulated operating conditions according to claim 5, characterized in that, The methods for performing abnormal impact identification include: The voltage-compensated battery data is divided into windows based on a preset fixed recognition time window to obtain recognition window data; a local voltage value sequence and a local current value sequence belonging to the window are constructed based on the recognition window data; the local fluctuation amplitude of the maximum local voltage value and the minimum local voltage value in the local voltage value sequence is calculated; if the local fluctuation amplitude is higher than the preset peak voltage threshold, the recognition window data corresponding to the local voltage value sequence is determined to be a voltage change risk window. Determine the direction of current change within each half-window before and after the local current value sequence, and simultaneously calculate the local current difference between the maximum and minimum local current values ​​within any half-window; if the direction of current change within each half-window is opposite and the local current difference is higher than the preset reverse current judgment value, then the identification window data corresponding to the local current value sequence is determined as a current direction change window. Based on the voltage surge risk window, trend fitting is performed on all sampling points to identify trend disturbance windows; if any two types of windows, including the voltage surge risk window, the current direction surge window, and the trend disturbance window, belong to the same window, then the window is determined to be an abnormal impact window.

7. The health assessment method for lead-acid batteries based on simulated operating conditions according to claim 6, characterized in that, The methods for performing trend fitting include: A linear fitting trend curve is plotted based on the local voltage value sequence corresponding to the voltage mutation risk window; the voltage deviation of the sampling points of the corresponding window relative to the linear fitting trend curve is calculated, and a fitting deviation sequence is constructed based on the voltage deviation; the variance of the fitting deviation sequence is calculated, and if the variance of the variance is greater than the preset trend disturbance judgment threshold, the window corresponding to the fitting deviation sequence is judged as a suspected trend disturbance window; the segment where the voltage deviation of consecutive sampling points in the suspected trend disturbance window is higher than twice the average voltage deviation is identified, and if the number of consecutive sampling points in the segment is higher than the lower limit of the peak time, the corresponding suspected trend disturbance window is determined as a trend disturbance window.

8. The method for health assessment of lead-acid batteries based on simulated operating conditions according to claim 7, characterized in that, The methods for risk assessment include: Traverse the time index of all sampling points in the reasonable battery data, extract the data segment whose continuous time length is longer than the preset resting time limit, and take the data segment as long period data; if the current value of the long period data is within the preset current float charging range, and the absolute value of the difference between the voltage value of the long period data and the preset float charging voltage threshold is not higher than the preset difference, then the corresponding long period data is determined as float charging state data; based on the timestamp record of the sampling point in the float charging state data, divide the float charging state data into unequal periods to obtain local resting period float charging data; Calculate the voltage standard deviation of any local static period float charge data. If the duration of the local static period float charge data is higher than the preset static period threshold and the voltage standard deviation is lower than the preset fluctuation threshold, then the data segment corresponding to the local static period float charge data is identified as a potential sulfidation risk segment. Calculate the float charge voltage amplitude of the potential sulfation risk section and the ratio of this float charge voltage amplitude to the preset standard stable voltage to obtain the float charge voltage stability ratio; calculate the output current variation amplitude and average output current value of the previous data section of the potential sulfation risk section; linearly combine the float charge voltage stability ratio, the output current variation amplitude and average output current value of the previous data section according to a preset ratio to obtain the health weighting coefficient; use the health weighting coefficient to weight the specific data of the potential sulfation risk section to obtain the corrected battery data.

9. A method for health assessment of lead-acid batteries based on simulated operating conditions according to claim 8, characterized in that, The methods for conducting health status assessments include: According to the health status assessment requirements, parameters of the corresponding dimensions are extracted from the differentiated battery data, and an assessment input dataset is constructed. The historical assessment template is matched with the assessment input dataset to output the health score of each dimension parameter. The health score is matched with the preset health score range to output the health level of the corresponding dimension parameter. The health levels of all dimension parameters are integrated to obtain the lead-acid battery health index.

10. A lead-acid battery health assessment system based on simulated operating conditions, used to implement the lead-acid battery health assessment method based on simulated operating conditions according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to collect simulated operating data of lead-acid batteries and perform data cleaning to obtain high-quality battery data. The effect analysis module is used to perform recovery effect analysis on high-quality battery data, obtain recovery effect data, and remove it from the high-quality battery data to obtain perturbation-removed battery data. The voltage compensation module is used to perform polarization voltage compensation on the disturbance removal battery data to generate voltage-compensated battery data. The anomaly identification module is used to perform abnormal impact identification on the voltage-compensated battery data, and filter abnormal data based on the identification results to output reasonable battery data. The risk assessment module is used to identify long-cycle data in reasonable battery data, assess the risk of long-cycle data, obtain corrected battery data and cover the corresponding data in reasonable battery data to obtain differentiated battery data. The health assessment module is used to assess the health status of differentiated battery data, output lead-acid battery health indicators and send them to the preset battery management terminal; the modules are connected to each other via wired and / or wireless means.

Images