Detection method and system of new energy lithium battery management system

Abstract

The invention relates to the technical field of battery detection, in particular to a detection method and system for a new energy lithium battery management system, and the method comprises the following steps: dividing charge and discharge sections based on a multiplying power voltage sampling classification set, classifying data, extracting inversion change points to construct an inflection point feature combination, comparing path trends, removing inconsistent sections, and obtaining a new energy lithium battery management system. And positioning a residual fluctuation edge, marking a state, grouping and associating response behavior segments, and obtaining a multiplying power segment detection feature set. According to the method, the voltage data are grouped and classified through differential multiplying power, the data structure adaptability under the multi-multiplying-power condition is improved, a voltage path structure is constructed through direction reversal and continuous change behaviors, inconsistent paragraphs are eliminated in combination with trend comparison between paths, and the main path range is limited; residual fluctuation edges are extracted, state division boundary sections are marked, affiliation sequence division is completed by comparing voltage changes in paths with response behaviors, and the consistency of behavior recognition logic and data organization in a multi-rate state is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of battery detection technology, and in particular to a detection method and system for a new energy lithium battery management system. Background Art

[0002] The field of battery testing technology encompasses the measurement, evaluation, and fault identification of battery performance. Its core focus is the real-time acquisition and evaluation of parameters such as voltage, current, temperature, capacity, internal resistance, and state of charge, used to determine the battery's health and operating status. This technical field covers a wide range of primary and secondary batteries, from portable power supplies to new energy vehicle power systems. Testing methods include sensor-based voltage, current, and temperature measurement, state determination using electrochemical models, equivalent circuit model parameter extraction, state of charge estimation, and lifespan estimation. Experimental data is often combined to conduct comparative analysis of static and dynamic parameters, providing quantitative reference data to support safe battery operation and system maintenance.

[0003] Among them, the detection method of the new energy lithium battery management system refers to a method of determining the current battery state of charge and health by collecting the terminal voltage, charge, discharge, current, and temperature data of the lithium-ion battery and comparing it with the voltage-time variation curve. This detection method addresses the problem that the accuracy of lithium battery status recognition during operation depends on the stability of external measurements. It usually uses the corresponding interval of the voltage platform measured during the constant current charging phase to determine the charge state by comparing the preset voltage-capacity relationship curve with the open-circuit voltage benchmark data. The attenuation trend is analyzed based on the offset of the discharge curve at different rates. The battery status detection and judgment is completed by comparing the standardized numerical difference result with the set threshold.

[0004] The existing technology for judging the battery status relies on the stability of parameter acquisition and the accuracy of the voltage-capacity relationship setting. When the rate stage fluctuates frequently or changes continuously, it is impossible to form a clear data attribution path. The voltage reversal phenomenon caused by the rate change cannot be timely identified by a single parameter model. The mixed situation of path trends is easily misjudged as effective voltage behavior, resulting in blurred identification of the main path boundary. Residual data is difficult to accurately define abnormal states without trend boundary division. The path response relationship is not behaviorally grouped in the voltage sequence structure, resulting in disordered distribution of response difference fragments. In complex working conditions with multiple rates, there are problems of state recognition delay and confusion in attribution logic. Summary of the Invention

[0005] In order to solve the technical problems existing in the prior art, an embodiment of the present invention provides a detection method for a new energy lithium battery management system, comprising the following steps:

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a detection method for a new energy lithium battery management system, comprising the following steps:

[0007] S1: Based on the rate state data corresponding to the rate voltage sampling classification set, the charge and discharge segments are divided according to the rate change identification points, and the data segments are sorted and classified according to the rate change, and the corresponding data segments are classified into differentiated rate segment sampling groups to obtain the rate voltage sampling classification set;

[0008] S2: Based on the data sequence of the same rate segment in the rate voltage sampling classification set, identify the voltage point with reversed direction and continuous change, splice the previous and next cycle data, mark the start and end ranges and change characteristics, and obtain the rate segment inflection point feature combination set;

[0009] S3: Based on the voltage sequence of the rate segment path in the rate segment inflection point feature combination set, compare the trend changes of the main path and the auxiliary path, remove inconsistent data segments, and obtain the main path detection target set;

[0010] S4: Based on the residual sequence of the rate segment in the main path detection target set, extract the fluctuation edge segment, divide the front and back trend intervals, assign a state identifier to the boundary position, and obtain the rate segment detection range segment;

[0011] S5: Based on the edge segments marked in the rate segment detection range, extract the path segments covering the rate stage, compare the voltage change state and the response content in the path, and group the segments with associated features into the same group to obtain the rate segment detection feature set.

[0012] As a further solution of the present invention, the rate voltage sampling classification set includes rate identification, sorting order, segmentation boundary, and voltage data group; the rate segment inflection point feature combination set includes reversal inflection point, periodic splicing range, turning feature label, and sequence structure; the main path detection target set includes auxiliary path trend, main path trend, deviation segment, and update path; the rate segment detection range segment includes residual fluctuation edge, residual trend direction, start and end state, and fluctuation feature interval; the rate segment detection feature set includes voltage change state, path response feature, matching associated segment, and same group sequence.

[0013] As a further solution of the present invention, the specific steps of S1 are:

[0014] S101: Obtain voltage sampling data corresponding to the rate state in new energy lithium battery management, detect the change point of the rate value in the time series, divide the continuous charge and discharge segments according to the change point, and obtain the rate change segment division result;

[0015] S102: extracting the magnification value of each section based on the magnification change section division result, sorting the section data in the order of magnification change, and obtaining a magnification sorting classification result by matching the sorting result with the voltage data in the section;

[0016] S103: Based on the identification information of adjacent rate segments in the rate sorting and classification result, each segment of voltage data is classified into a sampling group corresponding to the rate value to obtain a rate voltage sampling classification set.

[0017] As a further solution of the present invention, the specific steps of S2 are:

[0018] S201: extracting voltage sampling values ​​and time information based on a data sequence belonging to the same rate segment in the rate voltage sampling classification set, screening change point locations where voltage direction is reversed and exists continuously between adjacent sampling points, and obtaining a direction reversal change point set;

[0019] S202: extracting voltage sampling data of fixed periods before and after the direction reversal change point set based on the position of each change point, calculating the root mean square value of all sampling values ​​in the periodic segment, and obtaining a periodic voltage root mean square sequence;

[0020] S203: Based on the continuously fluctuating segments in the periodic voltage RMS sequence, identifying the voltage path segments of the turning behavior, marking the start and end positions and change characteristics of the associated segments, and obtaining a rate segment inflection point feature combination set.

[0021] As a further solution of the present invention, the calculation formula for the average value of the angle between the voltage change direction of each segment and the main path change direction is specifically:

[0022]

[0023] in, Indicates the root mean square value of all sampling values ​​within the period of the i-th rate segment, u ij Represents the voltage value of the jth voltage sampling point in the i-th rate segment, Represents the arithmetic mean of the voltage values ​​of all sampling points in the i-th rate segment, N i Represents the total number of sampling points in the i-th rate segment, λ i Represents the dynamic adjustment coefficient of voltage offset in the i-th rate cycle, α i Represents the voltage change trend response adjustment factor of the i-th rate cycle, Δt ij represents the sampling time interval of the jth voltage point in the i-th segment, Represents the instantaneous rate of change obtained by dividing the voltage difference before and after the jth sampling point in the i-th segment.

[0024] As a further solution of the present invention, the specific steps of S3 are:

[0025] S301: Based on the path voltage sampling sequence corresponding to each rate segment in the rate segment inflection point feature combination set, extract the voltage change trend of the main path and the auxiliary path in the same rate segment, compare the change direction and continuity in the path according to the sampling time sequence, identify the data segments with direction deviation and lack of trend continuation, and obtain the path deviation identification sequence;

[0026] S302: Based on the marked data segments in the path deviation identification sequence, extract the voltage sampling sequence within the corresponding rate segment, calculate the average angle between the voltage change direction of each segment and the main path change direction, analyze the degree of deviation based on the difference in the angle change amplitude, filter out segments with inconsistent directions, and obtain a direction deviation angle distribution sequence;

[0027] S303: Based on the data segment positions of the direction differences in the direction deviation angle distribution sequence, the data interval covered by the corresponding deviation segment is removed, and the current path range is defined according to the continuous distribution relationship of the remaining path segments in the magnification stage to obtain the main path detection target set.

[0028] As a further solution of the present invention, the average value of the angle between the voltage change direction of each segment and the change direction of the main path is calculated specifically as follows:

[0029]

[0030] in, Represents the average value of the angle within the magnification segment e, Represents the voltage change at sampling point y in the auxiliary path at the rate segment e. Represents the time interval corresponding to the voltage change of the auxiliary path sampling point y, Represents the voltage change at sampling point y in the main path, represents the time interval corresponding to the voltage change of sampling point y on the main path, and n represents the number of sampling points in the rate segment e.

[0031] As a further solution of the present invention, the specific steps of S4 are:

[0032] S401: Based on the residual sequence of the rate segment in the main path detection target set, extract the starting and ending positions of the residual data segment, analyze the change trend of the residual value of the sampling point on the time axis, identify the edge segments with sudden changes and amplitude jumps at both ends, and obtain the fluctuation edge positioning interval;

[0033] S402: Based on the edge segments included in the fluctuation edge positioning interval, sequentially segment the front and back time points of each segment, and divide them into a leading edge area and a trailing edge area according to the positional relationship. According to the direction of the residual trend change in the segment, the start and end states are marked to obtain a residual trend identification sequence;

[0034] S403: Based on the state information and fluctuation direction marked in the residual trend identification sequence, extract the coverage of the edge segment with continuous state in the rate stage, merge the start and end boundaries, and obtain the rate segment detection range segment.

[0035] As a further solution of the present invention, the specific steps of S5 are:

[0036] S501: Based on the marked edge segments in the rate detection range, extract the path data content covered by the edge segments within the rate stage, intercept the voltage sampling sequence and rate state information within the corresponding time range, and obtain a set of edge path data segments;

[0037] S502: Calling the voltage sampling sequence in the edge path data segment set, comparing the voltage change state in each segment of data with the response characteristics of the path markers one by one, analyzing the correspondence between the change direction and the response behavior, extracting segments with associated characteristics, and obtaining a voltage response associated segment group;

[0038] S503: Based on the change trend and response performance in the voltage response associated segment group, the associated segments are grouped and labeled according to sequence distribution, and classified into a path structure of a unified rate stage to obtain a rate segment detection feature set.

[0039] A detection system for a new energy lithium battery management system, comprising:

[0040] The rate recognition module obtains the voltage sampling data and corresponding rate status of the new energy lithium battery in the differentiated charging and discharging stages, identifies the rate status change points and divides them into continuous segments, arranges the data segments according to the time sequence of the rate change, and assigns the voltage sampling data corresponding to adjacent rate segments to the data group under each rate value. The voltage sample sequence and label content are divided by rate to obtain the rate voltage sampling classification set;

[0041] The inflection point extraction module identifies the points where the direction of voltage change reverses and has continuity based on the data sequence of the same rate segment in the rate voltage sampling classification set, connects the data range of adjacent periods before and after the change point and marks the start and end positions, filters the data segments with turning points and classifies them into a sequence form to obtain a rate segment inflection point feature combination set;

[0042] The main path identification module compares the voltage change directions of the main path and the auxiliary path under the same rate conditions based on the path voltage sequence associated with the rate segment in the rate segment inflection point feature combination set, identifies the segments with direction deviation but no continuous trend response, eliminates data segments unrelated to the trend, and adjusts the corresponding interval of the main path to obtain the main path detection target set;

[0043] The residual positioning module locates the fluctuation edges at both ends of the data segment based on the residual sequence of the rate segment in the main path detection target set, divides the residual trend start and end states and extracts the edge trend direction, extracts the fluctuation performance area in combination with the rate segment, and obtains the rate segment detection range segment;

[0044] The judgment construction module extracts the path data content under the covered rate stage based on the edge segments marked in the rate segment detection range segment, compares the relationship characteristics between the voltage state change and the response path, and classifies the fragments showing correlation into the same group to obtain the rate segment detection feature set.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are:

[0046] In the present invention, voltage data is classified by differentiated rate grouping, which improves the adaptability of data structure under multi-rate conditions, constructs voltage path structure through direction reversal and continuous change behavior, and eliminates inconsistent sections by combining trend comparison between paths to limit the main path range. By extracting residual fluctuation edges and marking state division boundary segments, the attribution sequence division is completed by comparing voltage changes and response behaviors in the path, which enhances the behavior recognition logic and data organization consistency under multi-rate states. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 Schematic diagram of the steps of the present invention;

[0049] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0051] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0052] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0053] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0054] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0055] See also Figure 1 , an embodiment of the present invention provides a detection method for a new energy lithium battery management system, comprising the following steps:

[0056] S1: Obtain voltage sampling data corresponding to the rate state in new energy lithium battery management, separate the continuous charge and discharge segments according to the rate change identification point, sort the data segments according to the order of rate value change, and classify the data content corresponding to adjacent rate segments into the voltage sampling group associated with each rate segment to obtain the rate voltage sampling classification set;

[0057] S2: Based on the data sequences in the rate voltage sampling classification set that belong to the same rate segment, identify the voltage change points in the sequence that reverse direction and have continuity, sequentially splice the periods before and after each change point and mark the start and end ranges, screen the segments with turning features and summarize them into a sequence structure, and annotate them according to the change behavior to obtain the rate segment inflection point feature combination set;

[0058] S3: Based on the voltage sampling sequence of the path corresponding to each rate segment in the rate segment inflection point feature combination set, the voltage change trend between the auxiliary path and the main path in the same rate segment is compared, and data segments with direction deviation and no obvious continuous trend matching are identified. Inconsistent segments are removed, and the current path range is updated to obtain the main path detection target set;

[0059] S4: Based on the residual sequence of the rate segment in the main path detection target set, locate the fluctuation edge segments before and after the data segment, divide the residual trend at both ends into segments and record the start and end states, extract the fluctuation feature range of the interval within the rate stage, and obtain the rate segment detection range segment;

[0060] S5: Based on the marked edge segments in the rate segment detection range, obtain the path data segments within the rate stage covered by the edge segments, compare the voltage change states and path response characteristics in the segments, and classify the parts with related performance into the same sequence group to obtain the rate segment detection feature set.

[0061] The rate voltage sampling classification set includes rate identification, sorting order, segment boundary, and voltage data group. The rate segment inflection point feature combination set includes reversal inflection point, period splicing range, turning feature label, and sequence structure. The main path detection target set includes auxiliary path trend, main path trend, deviation segment, and update path. The rate segment detection range segment includes residual fluctuation edge, residual trend direction, start and end state, and fluctuation feature interval. The rate segment detection feature set includes voltage change state, path response feature, matching associated segment, and same group sequence.

[0062] The specific steps of S1 are:

[0063] S101: Obtain voltage sampling data corresponding to the rate state in new energy lithium battery management, detect the change point of the rate value in the time series, divide the continuous charge and discharge segments according to the change point, and obtain the rate change segment division result;

[0064] First, it is necessary to collect a synchronous data set of voltage and rate status from the charge and discharge control unit. The sampling interval should be clearly defined during the collection process, such as sampling once per second, and the three parameters of the timestamp, voltage value, and current rate label corresponding to each data should be recorded. For example, at the 100th second, the voltage is 3.75V and the rate is 1C, at the 101st second, the voltage is 3.72V and the rate is 1C, and at the 102nd second, the voltage is 3.68V and the rate becomes 2C. Therefore, the 101st to 102nd seconds can be marked as the rate change point. When detecting the change point of the rate value in the time series, it is necessary to traverse the adjacent sampling points in the complete data sequence to see if there is a difference in the rate value, extract the time point with change and use it as the division node. If the previous rate value is 1C and the next rate value is 2C, the latter time point is used as the starting node, and the position of each change point is recorded in turn to form a change node index sequence, and then the change node index sequence is formed according to each two phases. The time period between adjacent change points is divided into a rate stable segment, and all voltage sampling values ​​and rate status labels within the time period are extracted respectively. Then, the segments are sorted according to the chronological order. For example, the period from 102 seconds to 160 seconds is identified as a stable segment with a rate of 2C, and its previous segment is the 1C segment, thereby completing the identification of the rate change boundary. During this period, it is also necessary to eliminate the single-point jitter phenomenon caused by misidentification. For example, the rate fluctuates once within two seconds and then returns to the original rate. At this time, it is necessary to judge whether the fluctuation duration is less than the set minimum valid segment time. If it is less than 5 seconds, it is merged into the same rate segment. Finally, the effective rate change node after eliminating the abnormality is used as the boundary, and all stable rate segments are output in chronological order. The start and end time, number of voltage data points, average voltage, rate label and other information of each segment are organized into a structured output to obtain the rate change segment division result.

[0065] S102: extracting the magnification value of each section based on the magnification change section division result, sorting the section data in the order of magnification change, and matching the sorting result with the voltage data in the section to obtain the magnification sorting classification result;

[0066] First, obtain the start and end time and corresponding rate label of each charge and discharge segment from the division result. For example, the start and end time of segment A is 0-120 seconds, the rate is 1C, segment B is 120-210 seconds, the rate is 2C, segment C is 210-300 seconds, and the rate is 0.5C. At this time, the rate value of each segment is extracted and sorted into a triplet corresponding to the time period. In the subsequent sorting operation, a joint judgment is made according to the size of the rate value and the time sequence. When the rate value is repeated but the order is different, the segment with the earlier start time is prioritized, that is, sorted by rate first, and then sorted by time sequence to form a sorting index list. The sorting result is reversely mapped to the original segment index, and the voltage sampling sequence and time label corresponding to each segment are retained. If segment A is If the voltage of segment A is 3.7V, 3.68V, and 3.65V, and segment B is 3.6V, 3.55V, and 3.5V, the segment labels and voltage sampling data are aligned in sorting order. Here, it is still necessary to determine the uniqueness of the multiplier value in the sorted segment. If there are multiple segments corresponding to the same multiplier but with staggered time, they are processed according to the principle of multiplier first and time last, and the multiplier unit conversion is unified (such as 0.5C, 1C, and 2C are converted to numerical values ​​0.5, 1, and 2 for sorting) to avoid sorting deviation affecting subsequent classification results. Each segment of voltage data retains its corresponding multiplier label for clustering by multiplier in the subsequent feature extraction stage. After all segments are sorted, the sorting number, corresponding multiplier value, and voltage sequence are organized into structured data one by one to obtain the multiplier sorting classification result.

[0067] S103: Based on the identification information of adjacent magnification segments in the magnification sorting and classification result, each segment of voltage data is classified into a sampling group corresponding to the magnification value to obtain a magnification voltage sampling classification set;

[0068] First, extract the magnification value and its start and end time corresponding to each segment after sorting, and establish a mapping table between the magnification value and the segment number. During the processing, it is necessary to classify according to the literal value of the magnification, traverse all segment numbers, and determine whether the magnification value of the current segment is the same as that of the previous segment. If the same, it is included in the existing magnification sampling group. If different, a new magnification sampling group is added and the current segment is included. During the classification process, a corresponding voltage sampling value list is established for each magnification sampling group, and all voltage data under the magnification are spliced ​​and merged in chronological order. The voltage data must maintain the original sampling accuracy and time interval unchanged. For example, the magnification of segment A is 1C, and the voltage sequence is 3.70V, 3.68V, and 3.65V. Segment C is also 1C, and its voltage sequence is 3.60V, 3.58V, and 3.55V. The two segments of voltage data are merged and included. 1C rate sampling group, and generate combined data with a voltage sequence length of 6. The judgment standard of the rate needs to maintain a unified format to avoid the problem of misgrouping due to numerical precision errors. Therefore, the rate should be formatted before comparing the rate value, such as converting 1C and 1.00C into floating point numbers 1.0, and then performing logical comparison operations. After completing the grouping of each rate value, it is necessary to extract the time interval of each group of voltage sampling points for subsequent analysis of voltage response characteristics by rate. All rate sampling groups must meet the minimum sample number requirement. If there are less than 10 sample points in a single group, they need to be merged into adjacent rate groups. The merging strategy is based on time continuity, and the rate groups with adjacent time and the closest rate values ​​are preferentially merged. After completing the classification operation, the segment number, voltage value sequence and time index range contained in each group are output to obtain the rate voltage sampling classification set.

[0069] The specific steps of S2 are:

[0070] S201: extracting voltage sampling values ​​and time information based on a data sequence belonging to the same rate segment in the rate voltage sampling classification set, screening the locations of change points where the voltage direction is reversed and exists continuously between adjacent sampling points, and obtaining a direction reversal change point set;

[0071] First, extract the voltage sampling value and the corresponding timestamp information from each segment of data, perform a point-by-point traversal operation on each continuous voltage data, compare the voltage value relationship between the current sampling point and the previous sampling point, and determine whether the change direction changes from rising to falling or from falling to rising. For example, in a certain segment of data, the voltage is 3.72V at the 10th second, 3.74V at the 11th second, and 3.69V at the 12th second. Since the voltage rises from the 10th second to the 11th second and falls from the 11th second to the 12th second, the 11th second is a reversal point in the voltage direction. This process continues on the entire segment of data. After each reversal point is found, continue to track whether the change trend of several subsequent points continues to maintain the opposite direction. If there are 3 or more points in a row that maintain the reversal direction, for example, the voltage continues to fall from the 11th second to the 14th second, it means that the reversal is not a short-term jitter but a continuous wave When judging continuity, a minimum continuous point threshold needs to be set. For example, if it is set to 3 points, it means that if the directions of 3 consecutive points are consistent after reversal, it can be judged as continuous reversal. If it does not meet the requirements, the point will be removed to avoid misjudgment. At the same time, short-term high-frequency noise points need to be removed. In actual operation scenarios, the battery may be disturbed by signals during high-speed sampling, resulting in single-point abnormalities. Therefore, the data should be smoothed first during processing, such as 3-point sliding average processing. The average of each sampling point and the points before and after it is used for trend judgment, further enhancing the stability of direction judgment. After the judgment is completed, all point indexes that meet the reversal and continuous change conditions are recorded together with their corresponding timestamps and voltage values, and the marks are listed separately in the data set for use in subsequent cycle splicing operations. Finally, all direction change points that meet the conditions are extracted from the voltage data under all rate segments to obtain a set of direction reversal change points.

[0072] S202: extracting voltage sampling data of fixed periods before and after the direction reversal change point set based on the position of each change point, calculating the root mean square value of all sampling values ​​in the periodic segment, and obtaining a periodic voltage root mean square sequence;

[0073] The calculation formula for the average value of the angle between the voltage change direction of each segment and the change direction of the main path is as follows:

[0074]

[0075] in, Indicates the root mean square value of all sampling values ​​within the period of the i-th rate segment, u ij Represents the voltage value of the jth voltage sampling point in the i-th rate segment, Represents the arithmetic mean of the voltage values ​​of all sampling points in the i-th rate segment, N i Represents the total number of sampling points in the i-th rate segment, λ i Represents the dynamic adjustment coefficient of voltage offset in the i-th rate cycle, α iRepresents the voltage change trend response adjustment factor of the i-th rate cycle, Δt ij represents the sampling time interval of the jth voltage point in the i-th segment, Represents the instantaneous rate of change obtained by dividing the voltage difference before and after the jth sampling point in the i-th segment.

[0076] Assume that within the rate segment i, the number of sampling points N i =6, time interval Δt ij = 0.1s, the voltage sequence of the sampling points is as follows (unit: V):

[0077] (3.21,3.26,3.33,3.30,3.24,3.18);

[0078] Calculate the cycle average voltage

[0079]

[0080] Calculate the difference change rate of each point before and after

[0081]

[0082] ...and so on

[0083] Set parameter values:

[0084] λ i =0.6:

[0085] α i =0.4:

[0086] Substitute the third point (3.33V) into the calculation:

[0087] |u i3 |=3.33;

[0088]

[0089] but

[0090] Combined calculation:

[0091] (3.33+0.6×0.0767+0.4×0.07) 2 ;

[0092] =(3.33+0.04602+0.028) 2 ;

[0093] =(3.40402) 2 ;

[0094] ≈11.591;

[0095] Expand all sampling points and calculate the square root of the average sum of squares after substituting them into the formula to obtain:

[0096]

[0097] The results show that the voltage state in the periodic sampling segment has a response offset, and the value can be used as the basis for identifying the subsequent path residual fluctuation segment.

[0098] S203: Based on the continuous fluctuation segments in the periodic voltage RMS sequence, identifying the voltage path segments of the turning behavior, marking the start and end positions and change characteristics of the associated segments, and obtaining a rate segment inflection point feature combination set;

[0099] First, extract the RMS voltage value and its time index corresponding to each cycle, and then scan the RMS value in continuous segments in the form of a time series to identify the parts with direction change characteristics in several adjacent cycles. For example, if the RMS voltage values ​​of five consecutive cycles are 0.08V, 0.07V, 0.06V, 0.07V, and 0.09V respectively, it can be determined that the voltage fluctuation direction changes in the opposite direction after the third cycle and lasts for two cycles. It is preliminarily identified as a manifestation of an inflection point behavior. When performing the identification operation, it is necessary to set the minimum number of continuous fluctuation cycles, for example, to 3 cycles, that is, the RMS value trend must change in the same direction for more than 3 consecutive cycles and then reverse before it can be judged as a valid inflection point. At the same time, exclude short-term repeated changes caused by signal sampling jitter. After confirming the inflection point segment, propose Take the time segment corresponding to the segment in the original voltage path, mark its starting point and ending point point by point, and classify and annotate the voltage change trend of the segment. For example, the change from rising to falling is marked as "rising and falling type", and the change from falling to rising is marked as "falling and rising type". If the voltage levels of the starting and ending points are similar but there is obvious fluctuation in the middle, it is marked as "fluctuation type". The annotation process needs to combine the direction and amplitude of the voltage change to determine whether it belongs to a certain change feature type. For example, if the starting point voltage in a certain segment is 3.6V and the end point voltage is 3.61V, but the fluctuation range in the middle is ±0.1V, it should not be marked as stable type. The inflection point behavior should have continuity and obvious structural change trend. After all inflection point segments are extracted, the start and end time, voltage change direction identification, change feature classification and other information of each segment are unified and sorted out to obtain the rate segment inflection point feature combination set.

[0100] The specific steps of S3 are:

[0101] S301: Based on the path voltage sampling sequence corresponding to each rate segment in the rate segment inflection point feature combination set, extract the voltage change trend of the main path and the auxiliary path under the same rate segment, compare the change direction and continuity in the path according to the sampling time sequence, identify the data segments with direction deviation and lack of trend continuation, and obtain the path deviation identification sequence;

[0102] First, the voltage sampling values ​​and time index information of the main path and the auxiliary path in the same rate segment are extracted to construct two time series data sequences of equal length, where the main path is the reference sequence under normal operating conditions, and the auxiliary path is the observation sequence for comparative analysis. During the comparison process, the time axis must be aligned, that is, the change direction and continuous trend of the voltage values ​​between the two paths are compared at the same time point. The first step in the judgment is to extract the voltage difference sign between two adjacent sampling points. If the voltage difference direction of more than three consecutive points is consistent, it is considered to be the same trend segment. For example, the voltages from 10 to 14 seconds in the main path are 3.7V, 3.72V, 3.75V, 3.77V, and 3.80V respectively, then the trend of this segment is rising. If the voltage of the auxiliary path is 3.7V, 3.68V, 3.66V, 3.64V, and 3.63V in the same time period, then the trend is falling, and the direction is opposite. To determine whether a segment is a trend deviation segment, in addition to the direction, the continuity must also be judged, that is, whether there is an interruption in the trend segment. If the direction changes suddenly or fluctuates between two points, it is considered to have no trend continuity. At the same time, local abnormal segments caused by occasional jumps are excluded. By setting the minimum number of trend continuity points to 3, it is determined whether the trend constitutes a continuous trend segment. After completing the above judgment, all path segments in each rate segment are compared and analyzed. The segments that meet the direction deviation and trend discontinuity are marked and their start and end time range, corresponding rate segment, main path voltage segment number, and auxiliary path segment number are recorded. For example, in the 1.5C rate range, the main path is a stable rising segment from 200 to 210 seconds, while the auxiliary path is a falling segment. In this way, this time interval is marked as a deviation segment and written into the sequence as the result. Finally, the deviation segment position information under all rate segments is sorted to obtain the path deviation identification sequence.

[0103] S302: Based on the marked data segments in the path deviation identification sequence, extract the voltage sampling sequence within the corresponding rate segment, calculate the average angle between the voltage change direction of each segment and the main path change direction, analyze the degree of deviation based on the difference in the angle change amplitude, filter out segments with inconsistent directions, and obtain a direction deviation angle distribution sequence;

[0104] The calculation formula for the average value of the angle between the voltage change direction of each segment and the change direction of the main path is:

[0105]

[0106] in, Represents the average value of the angle within the magnification segment e, Represents the voltage change at sampling point y in the auxiliary path at the rate segment e. Represents the time interval corresponding to the voltage change of the auxiliary path sampling point y, Represents the voltage change at sampling point y in the main path, represents the time interval corresponding to the voltage change of sampling point y on the main path, and n represents the number of sampling points in the rate segment e;

[0107] Assumptions:

[0108] In the rate range e = 2, the voltage changes of the auxiliary path collected at four points are: 0.15V, 0.18V, 0.17V, 0.16V, and the corresponding time interval is uniformly 0.5s. The voltage changes of the main path at the corresponding positions are: 0.13V, 0.16V, 0.15V, 0.14V, and the time interval is also 0.5s.

[0109] Take the first sampling point as an example:

[0110] The voltage direction difference term is calculated as

[0111] 0.15·0.5-0.13·0.5=0.075-0.065=0.01;

[0112] The modulus lengths of the two path vectors are

[0113]

[0114] The cosine difference of the angle is

[0115]

[0116] Similarly, the following three points also have an absolute angle difference of 0.037 according to the above calculation method, and the average angle is:

[0117]

[0118] The results show that the average deviation degree of the voltage change direction between the main path and the auxiliary path in the rate segment e is that the closer the average angle value is to 0, the higher the consistency of the voltage change direction between the paths. If the value exceeds the set angle difference threshold γ, it means that there is a significant deviation between the paths in this rate segment.

[0119] S303: Based on the position of the data segments with directional differences in the direction deviation angle distribution sequence, remove the data interval covered by the corresponding deviation segments, and define the current path range according to the continuous distribution relationship of the remaining path segments in the magnification stage to obtain the main path detection target set;

[0120] First, all path segments marked as inconsistent in direction are extracted from the distribution sequence. Each segment must include its starting time point, ending time point, and the identification information of the rate segment in which it is located. For example, under the rate segment 1.5C, it is identified that there is an obvious direction difference from 150 seconds to 160 seconds. This time period is recorded as a segment to be removed. Then, the data segment consistent with the time of the segment is located in the main path data, and the voltage sampling value, voltage change direction and other information in the corresponding time interval are removed as a whole, and no longer included in the subsequent path range judgment. The removal operation needs to be performed segment by segment to avoid the disconnection of adjacent segments caused by batch removal. For the case where adjacent segments are connected, for example, a deviation segment is from 160 to 170 seconds, and the remaining data segment of the main path is from 140 to 160 seconds. The time interval is from 100 to 180 seconds. After removing the deviated segments, it is necessary to determine whether a new interruption zone has appeared. If the removal operation causes the path to be divided into multiple discontinuous segments, the remaining data segments need to be grouped and sorted according to time continuity. The next step is to group the remaining paths under each rate segment and construct a continuous path structure in chronological order. If the time interval between two segments does not exceed the set interval threshold (such as 5 seconds), it can be determined as a continuous path. Otherwise, it is considered an interrupted path and needs to be separately identified. Combined with the time start and end points, data integrity, voltage change continuity and other parameters of the remaining path segments, the valid paths in each rate stage are re-summarized and the updated start and end ranges are recorded. Finally, a record result containing continuous main path segments is formed in units of rate segments to obtain the main path detection target set.

[0121] The specific steps of S4 are:

[0122] S401: Based on the residual sequence of the magnification segment in the main path detection target set, the starting and ending positions of the residual data segment are extracted, the changing trend of the residual value of the sampling point on the time axis is analyzed, and the edge segments with sudden changes and amplitude jumps at both ends are identified to obtain the fluctuation edge positioning interval;

[0123] First, perform data extraction on each rate segment to clarify the time index points of the starting and ending positions in each sequence. For example, when the rate segment is 2.0C, extract the 300th to 400th seconds in the residual sequence as the analysis target. Then read the residual values ​​corresponding to all sampling points in this period and arrange them in chronological order to construct a time series residual data set. Then observe the changes in the residual data in the time dimension. The judgment of the change trend is based on the continuous rising and falling sequence of residual values. For example, if the residual values ​​of 5 consecutive sampling points are 0.01, 0.02, 0.05, 0.07, and 0.11, then the segment is an upward trend. If 0.01, 0.02, 0.50, 0.03, and 0.02 appear in the middle, then the third point is a mutation point. When judging the trend, the position index of the starting point and end point of each trend segment must be recorded, and the edge segment must be identified in combination with the jump of the residual value amplitude. The judgment logic is that if the difference between the residual value of a point and its previous position exceeds 3 times the average residual of the entire segment, the point is marked as a mutation. For example, if the average residual is 0.02 and the residual of the current point is 0.08, it is an edge point if the jump condition is met. This type of judgment operation covers the entire residual sequence to ensure that each mutation segment is identified. After the identification is completed, a fluctuation segment is formed according to each pair of continuous edge points and its start and end time is recorded. For example, if there is an abnormal jump and trend break in the edge segment between 315 seconds and 327 seconds, this segment is classified as a fluctuation edge. Finally, the fluctuation segments identified in all magnification segments are sorted out to obtain the fluctuation edge positioning interval.

[0124] S402: Based on the edge segments included in the fluctuation edge positioning interval, sequentially segment the preceding and following time points of each segment, and divide them into a leading edge area and a trailing edge area according to their positional relationship. Mark the start and end states according to the direction of residual trend change in the segment to obtain a residual trend identification sequence;

[0125] The start and end time information of each segment is read in sequence. For example, a certain edge segment starts at 240 seconds and ends at 255 seconds. A fixed sampling step of 5 seconds is extended forward and backward within this time range to define the leading edge and trailing edge regions at the boundary. The leading edge region is set to 235 seconds to 240 seconds, and the trailing edge region is set to 255 seconds to 260 seconds. The residual value sequence within each extended segment is then called, and the direction of residual change corresponding to each position is counted according to the time sequence. For example, if the current residual changes continuously from negative to positive in the leading edge region and gradually decreases in the trailing edge region, the leading edge region status is marked as "rising" and the trailing edge region status is marked as "falling", respectively. If the residual values ​​in the leading edge region of a segment are -0.02, -0.01, and 0, respectively. 0.01, 0.03, 0.05, and the trailing edge area is 0.06, 0.04, 0.03, 0.01, -0.01. It can be confirmed that the residual shows an increasing and shrinking trend in the directions of the front and rear segments, respectively. State identifiers are established for different segments. Two trend types, "start rising" and "end falling", are configured on both sides of each edge segment to distinguish the directions of different edges. Then, the marking results in all edge segments are arranged in time index order to construct a complete trend sequence. In the example, this sequence is manifested as a change type sequence such as "rise-fall-rise-rise-fall", and each edge segment has two state markers. Finally, all edge segments involved in the entire segment are summarized and trend information is sorted out to obtain a residual trend identifier sequence.

[0126] S403: Based on the state information and fluctuation direction marked in the residual trend identification sequence, extract the coverage of the edge segment with continuous state in the rate stage, merge the start and end boundaries, and obtain the rate segment detection range segment;

[0127] First, the edge segments with continuous states and consistent labels are retrieved and summarized, and the adjacent and continuous segments showing "rise" or "fall" in the state sequence are uniformly identified as the same group of data. For example, if there is a continuous distribution of "rise-rise-rise" in the state sequence, the three segments are marked as the same edge area, and the start time of the first segment and the end time of the last segment are extracted as the total coverage range of the group of edge segments. Then, the start and end values ​​of the time points in each group are merged to calculate their position distribution in the overall rate segment. For example, if a group of edge segments starts at the 460th second and ends at the 525th second, then its span in the rate stage can be determined by comparing it with the original rate segment time series range (such as the rate segment is from the 400th second to the 600th second). The position features in the segment are then used to form a time period set with the coverage time index of each group of continuous edge segments at the mapping position of the rate stage. If multiple groups of edge segments have time overlap or are adjacent for less than 10 seconds, they are merged into the same interval, and the boundary values ​​of the front and back segments are merged to form the overall segment boundary. Then, all the merged coverage intervals are summarized within the rate segment range, and finally multiple continuous detection target areas are constructed within each rate segment. In practical applications, this operation is used, for example, to compare whether the voltage fluctuations of different segments are stable. For example, if the voltage residual in the same rate stage shows "first rise and then fall" and the start and end times are 200 seconds and 250 seconds respectively, then this paragraph constitutes a valid detection segment. Finally, all the summarized segment boundary time information is output to obtain the rate segment detection range segment.

[0128] The specific steps of S5 are:

[0129] S501: Based on the marked edge segments in the rate detection range, extract the path data content covered by the edge segments within the rate stage, intercept the voltage sampling sequence and rate state information within the corresponding time range, and obtain a set of edge path data segments;

[0130] First, the start and end time values ​​corresponding to each edge segment are retrieved. For example, the start and end time of a certain edge segment is from the 120th second to the 160th second. Based on the obtained time range, the path data record file within the time interval is called to identify the data row content that matches the time index, and further filter out the field group of the recorded path data. The field group usually contains multiple dimensional contents such as timestamp, voltage sampling value, current value, and rate identification. Among them, the "voltage sampling sequence" field and the "rate status" field are extracted as target data participants, and they are intercepted and processed in chronological order to ensure that the path data corresponding to each edge segment is a complete and continuous time period data set. Then, multiple edge segments are processed. The same processing is performed separately. For example, if there are three edge segments, namely 120-160 seconds, 210-245 seconds, and 270-290 seconds, it is necessary to extract the voltage sampling and rate state comparison sequences of each of these three time intervals, and construct three groups of path data fragments. In the process, it is also necessary to eliminate the sampling interruption segments caused by inconsistent sampling intervals or missing records, and smooth the missing parts according to the difference between the previous and next sampling points. For example, if the voltage record is missing between the 134th and 136th seconds, the voltage point at 135th second is inserted as the average of the two points before and after the completion, and then included in the complete data segment. Finally, the data fragments intercepted in the rate stage of all edge segments are unified into the set item to obtain the edge path data fragment set.

[0131] S502: Calling the voltage sampling sequence in the edge path data segment set, comparing the voltage change state in each segment of data with the response characteristics of the path markers one by one, analyzing the correspondence between the change direction and the response behavior, extracting segments with associated characteristics, and obtaining a voltage response associated segment group;

[0132] First, each set of path data in the set is traversed sequentially, and the fields such as time, voltage, and rate status are extracted. A traversal index is established with each fragment as the minimum processing unit. During the traversal process, the change direction of the voltage sampling value is read item by item, and the rising, falling, or flat behavior between adjacent points is identified. The behavior result is temporarily stored as a change label sequence. The corresponding response feature is provided by the previous path annotation record. For example, a certain segment of data has been marked with three types of behaviors such as "sudden rise", "steady state", and "jump drop". If the voltage of the current data segment shows an upward trend continuously, that is, the change label is "rising, rising, rising, flat, falling", it is necessary to compare the correspondence between the label sequence and the response feature label sequence point by point, and find the sampling segments with consistent behavior or similar change trend direction through position matching. Further determine whether there is a time overlapping segment and record it. For example, the response feature is " "Sudden rise" requires that there are two or more continuous rising labels in the voltage sampling sequence, and the average rising amplitude is not less than a certain baseline value. If this condition is met, the data segment is extracted as a response feature associated fragment. The entire processing process is executed for each segment of edge path data, and the start and end time and data index of all data fragments in which behaviors and responses are associated are recorded, and these associated fragments are unified into a summary table. Then, items with time overlap or repeated content are removed, and multiple fragments within the adjacent index range are merged into a continuous data unit, finally forming a group of valid fragments corresponding to voltage changes and response behaviors. For example, if the voltage in a certain fragment rises continuously from the 105th second to the 110th second, and there is a "fast charging response" mark in this interval, then the fragment is identified as a valid response associated fragment and included in the set, and finally the voltage response associated fragment group is obtained.

[0133] S503: Based on the change trend and response performance of the voltage response associated segment group, the associated segments are grouped and labeled according to sequence distribution, and classified into a path structure of a unified rate stage to obtain a rate segment detection feature set;

[0134] First, each segment of data in the associated segment group is numbered sequentially, and the time start and end positions and magnification status of each segment in the original sampling data are extracted. At the same time, the directional label of the voltage change in the segment is extracted, such as rising, falling, oscillating, etc., and each segment change trend is marked. For example, the continuous rising sequence is marked as "U", the falling sequence is marked as "D", and the oscillating or no obvious direction segment is marked as "N". This type of mark is combined with the response feature content to form a unique identification label. After all segments are marked, a time sequence index table is established. The segment labels are traversed in sequence according to the numbering order and the continuity of the time position between adjacent segments is analyzed. Segments with an interval less than twice the sampling period are regarded as continuous event segments and merged. These segments are uniformly classified into the same grouping label, and the number of segments, time span, total voltage change amplitude, etc. within the grouping range are recorded to construct a complete grouping identification information. Then, according to the magnification corresponding to each group of segments, the segment number, time span, and total voltage change amplitude are recorded. The rate stage number to which the group belongs is extracted from the state, and the path structure hierarchical mapping is completed. The grouped segments are then classified into the rate structure path table according to the rate number. For example, if three segments numbered P001, P002, and P003 are sequentially located in the same rate state, and the time interval between P001 and P002 is less than 10 seconds, and the time interval between P002 and P003 is less than 12 seconds, they can be merged into group R1 and classified into rate stage B3. Finally, a group mapping index table is established under the rate path structure, and the group label is bound to the original path label. The voltage trend information covered by each grouping tag in the rate structure path is bound to the response performance. For example, if the trend of group R1 is "UUU" and its response is "charging peak segment", the mapping pair "R1–B3–UUU–charging peak segment" is recorded in the structure table and included in the final detection feature reference table. Finally, the grouping tags and their changing performance in all rate stages are summarized to obtain the rate segment detection feature set.

[0135] See also Figure 2 , a detection system for a new energy lithium battery management system, comprising:

[0136] The rate recognition module obtains the voltage sampling data and corresponding rate status of the new energy lithium battery in the differentiated charging and discharging stages, identifies the rate status change points and divides them into continuous segments, arranges the data segments according to the time sequence of the rate change, and assigns the voltage sampling data corresponding to adjacent rate segments to the data group under each rate value. The voltage sample sequence and label content are divided by rate to obtain the rate voltage sampling classification set;

[0137] The inflection point extraction module identifies the points where the voltage change direction reverses and has continuity based on the data sequence of the same rate segment in the rate voltage sampling classification set. It connects the data range of the adjacent cycles before and after the change point and marks the start and end positions. It then selects the data segments with turning points and classifies them into a sequence to obtain a combination set of rate segment inflection point features.

[0138] The main path identification module combines the voltage sequences of the paths associated with the rate segments based on the rate segment inflection point feature set. It compares the voltage change directions of the main and auxiliary paths under the same rate conditions, identifies segments with deviations in direction but no continuous trend response, removes trend-independent data segments, and adjusts the corresponding intervals of the main path to obtain the main path detection target set.

[0139] The residual positioning module detects the residual sequence of the rate segment in the target set based on the main path, locates the fluctuation edges at both ends of the data segment, divides the start and end states of the residual trend and extracts the edge trend direction. It extracts the fluctuation performance area based on the rate segment and obtains the rate segment detection range segment.

[0140] The judgment construction module extracts the path data content under the covered rate stage based on the edge segments marked in the rate segment detection range, compares the relationship characteristics between the voltage state change and the response path, and classifies the fragments showing correlation into the same group to obtain the rate segment detection feature set.

[0141] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A detection method for a new energy lithium battery management system, characterized in that: The following steps are involved: S1: Based on the rate state data corresponding to the rate voltage sampling classification set, the charge and discharge segments are divided according to the rate change identification points, and the data segments are sorted and classified according to the rate change, and the corresponding data segments are classified into differentiated rate segment sampling groups to obtain the rate voltage sampling classification set; S2: Based on the data sequence of the same rate segment in the rate voltage sampling classification set, identify the voltage point with reversed direction and continuous change, splice the previous and next cycle data, mark the start and end ranges and change characteristics, and obtain the rate segment inflection point feature combination set; S3: Based on the voltage sequence of the rate segment path in the rate segment inflection point feature combination set, compare the trend changes of the main path and the auxiliary path, remove inconsistent data segments, and obtain the main path detection target set; S4: Based on the residual sequence of the rate segment in the main path detection target set, extract the fluctuation edge segment, divide the front and back trend intervals, assign a state identifier to the boundary position, and obtain the rate segment detection range segment; S5: Based on the edge segments marked in the rate segment detection range, extract the path segments covering the rate stage, compare the voltage change state and the response content in the path, and group the segments with associated features into the same group to obtain the rate segment detection feature set.

2. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The rate voltage sampling classification set includes rate identification, sorting order, segmentation boundary, and voltage data group; the rate segment inflection point feature combination set includes reversal inflection point, periodic splicing range, turning feature label, and sequence structure; the main path detection target set includes auxiliary path trend, main path trend, deviation segment, and update path; the rate segment detection range segment includes residual fluctuation edge, residual trend direction, start and end state, and fluctuation feature interval; the rate segment detection feature set includes voltage change state, path response feature, matching associated segment, and same group sequence.

3. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The specific steps of S1 are: S101: Obtain voltage sampling data corresponding to the rate state in new energy lithium battery management, detect the change point of the rate value in the time series, divide the continuous charge and discharge segments according to the change point, and obtain the rate change segment division result; S102: extracting the magnification value of each section based on the magnification change section division result, sorting the section data in the order of magnification change, and obtaining a magnification sorting classification result by matching the sorting result with the voltage data in the section; S103: Based on the identification information of adjacent magnification segments in the magnification sorting and classification result, each segment of voltage data is classified into a sampling group corresponding to the magnification value to obtain a magnification voltage sampling classification set.

4. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The specific steps of S2 are: S201: extracting voltage sampling values ​​and time information based on a data sequence belonging to the same rate segment in the rate voltage sampling classification set, screening change point locations where voltage direction is reversed and exists continuously between adjacent sampling points, and obtaining a direction reversal change point set; S202: extracting voltage sampling data of fixed periods before and after the direction reversal change point set based on the position of each change point, calculating the root mean square value of all sampling values ​​in the periodic segment, and obtaining a periodic voltage root mean square sequence; S203: Based on the continuously fluctuating segments in the periodic voltage RMS sequence, identifying the voltage path segments of the turning behavior, marking the start and end positions and change characteristics of the associated segments, and obtaining a rate segment inflection point feature combination set.

5. The detection method of the new energy lithium battery management system according to claim 4, characterized in that: The calculation formula for the average value of the angle between the voltage change direction of each segment and the change direction of the main path is specifically: in, Indicates the root mean square value of all sampling values ​​within the period of the i-th rate segment, u ij Represents the voltage value of the jth voltage sampling point in the i-th rate segment, Represents the arithmetic mean of the voltage values ​​of all sampling points in the i-th rate segment, N i Represents the total number of sampling points in the i-th rate segment, λ i Represents the dynamic adjustment coefficient of voltage offset in the i-th rate cycle, α i Represents the voltage change trend response adjustment factor of the i-th rate cycle, Δt ij represents the sampling time interval of the jth voltage point in the i-th segment, Represents the instantaneous rate of change obtained by dividing the voltage difference before and after the jth sampling point in the i-th segment.

6. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The specific steps of S3 are: S301: Based on the path voltage sampling sequence corresponding to each rate segment in the rate segment inflection point feature combination set, extract the voltage change trend of the main path and the auxiliary path in the same rate segment, compare the change direction and continuity in the path according to the sampling time sequence, identify the data segments with direction deviation and lack of trend continuation, and obtain the path deviation identification sequence; S302: Based on the marked data segments in the path deviation identification sequence, extract the voltage sampling sequence within the corresponding rate segment, calculate the average angle between the voltage change direction of each segment and the main path change direction, analyze the degree of deviation based on the difference in the angle change amplitude, filter out segments with inconsistent directions, and obtain a direction deviation angle distribution sequence; S303: Based on the data segment positions of the direction differences in the direction deviation angle distribution sequence, the data interval covered by the corresponding deviation segment is removed, and the current path range is defined according to the continuous distribution relationship of the remaining path segments in the magnification stage to obtain the main path detection target set.

7. The detection method of the new energy lithium battery management system according to claim 6, characterized in that: The calculation of the average value of the angle between the voltage change direction of each segment and the change direction of the main path is specifically based on the formula: in, Represents the average value of the angle within the magnification segment e, Represents the voltage change at sampling point y in the auxiliary path at the rate segment e. Represents the time interval corresponding to the voltage change of the auxiliary path sampling point y, Represents the voltage change at sampling point y in the main path, represents the time interval corresponding to the voltage change of sampling point y on the main path, and n represents the number of sampling points in the rate segment e.

8. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The specific steps of S4 are: S401: Based on the residual sequence of the rate segment in the main path detection target set, extract the starting and ending positions of the residual data segment, analyze the change trend of the residual value of the sampling point on the time axis, identify the edge segments with sudden changes and amplitude jumps at both ends, and obtain the fluctuation edge positioning interval; S402: Based on the edge segments included in the fluctuation edge positioning interval, sequentially segment the front and back time points of each segment, and divide them into a leading edge area and a trailing edge area according to the positional relationship. According to the direction of the residual trend change in the segment, the start and end states are marked to obtain a residual trend identification sequence; S403: Based on the state information and fluctuation direction marked in the residual trend identification sequence, extract the coverage of the edge segment with continuous state in the rate stage, merge the start and end boundaries, and obtain the rate segment detection range segment.

9. The detection method of the new energy lithium battery management system according to claim 1, characterized in that: The specific steps of S5 are: S501: Based on the marked edge segments in the rate detection range, extract the path data content covered by the edge segments within the rate stage, intercept the voltage sampling sequence and rate state information within the corresponding time range, and obtain a set of edge path data segments; S502: Calling the voltage sampling sequence in the edge path data segment set, comparing the voltage change state in each segment of data with the response characteristics of the path markers one by one, analyzing the correspondence between the change direction and the response behavior, extracting segments with associated characteristics, and obtaining a voltage response associated segment group; S503: Based on the change trend and response performance in the voltage response associated segment group, the associated segments are grouped and labeled according to sequence distribution, and classified into a path structure of a unified rate stage to obtain a rate segment detection feature set.

10. A detection system for a new energy lithium battery management system, characterized in that: The system is used to implement the detection method of the new energy lithium battery management system according to any one of claims 1 to 9, and the system includes: The rate recognition module obtains the voltage sampling data and corresponding rate status of the new energy lithium battery in the differentiated charging and discharging stages, identifies the rate status change points and divides them into continuous segments, arranges the data segments according to the time sequence of the rate change, and assigns the voltage sampling data corresponding to adjacent rate segments to the data group under each rate value. The voltage sample sequence and label content are divided by rate to obtain the rate voltage sampling classification set; The inflection point extraction module identifies the points where the direction of voltage change reverses and has continuity based on the data sequence of the same rate segment in the rate voltage sampling classification set, connects the data range of adjacent periods before and after the change point and marks the start and end positions, filters the data segments with turning points and classifies them into a sequence form to obtain a rate segment inflection point feature combination set; The main path identification module compares the voltage change directions of the main path and the auxiliary path under the same rate conditions based on the path voltage sequence associated with the rate segment in the rate segment inflection point feature combination set, identifies the segments with direction deviation but no continuous trend response, eliminates data segments unrelated to the trend, and adjusts the corresponding interval of the main path to obtain the main path detection target set; The residual positioning module locates the fluctuation edges at both ends of the data segment based on the residual sequence of the rate segment in the main path detection target set, divides the residual trend start and end states and extracts the edge trend direction, extracts the fluctuation performance area in combination with the rate segment, and obtains the rate segment detection range segment; The judgment construction module extracts the path data content under the covered rate stage based on the edge segments marked in the rate segment detection range segment, compares the relationship characteristics between the voltage state change and the response path, and classifies the fragments showing correlation into the same group to obtain the rate segment detection feature set.

Images