Bms battery state of health determination method fusing capacity and internal resistance characteristics

Abstract

The application discloses a BMS battery health state determination method fusing capacity and internal resistance characteristics, relates to the technical field of battery management and state evaluation, and comprises the following steps: acquiring a battery capacity increment slope value in a charging end interval, an internal resistance peak value interval duration and an internal resistance peak value amplitude difference proportion, dividing a capacity change convergence segment along a time sequence, and marking an internal resistance multi-peak dense section inside the capacity change convergence segment. The application reconstructs the capacity increment slope value and the internal resistance characteristics in the capacity change convergence segment, forms a consistent change relationship between the capacity and the internal resistance in the time dimension, improves the differential expression capability after the feature fusion, eliminates the inconsistent problems caused by the multi-path derivation by constructing a single determination track, and thus improves the continuity and stability of the battery health state determination result.

Description

Technical Field

[0001] This invention relates to the field of battery management and state assessment technology, specifically to a BMS battery health status determination method that integrates capacity and internal resistance characteristics. Background Technology

[0002] Battery health status assessment using a BMS that integrates capacity and internal resistance characteristics refers to simultaneously extracting effective capacity change information and internal resistance change information during charge-discharge cycles in the battery management process. It correlates the declining energy storage capacity reflected by capacity decay with the internal conduction obstruction reflected by rising internal resistance. By comprehensively analyzing the synergistic changes of these two types of characteristics over time, a multi-dimensional characterization of battery aging, performance degradation paths, and operational stability is constructed. Based on this, a corresponding health status level or range is output to guide subsequent operational adjustments and maintenance decisions, thereby avoiding biases caused by single-feature assessments and improving the accuracy and consistency of health evaluation.

[0003] The existing technology has the following shortcomings:

[0004] In the process of determining battery health status by integrating capacity and internal resistance characteristics, when the battery is in the final stage of charging, the battery capacity gradually approaches its saturation value, and the amplitude of capacity change converges significantly, thus weakening the ability of capacity characteristics to distinguish state differences. Simultaneously, due to enhanced polarization effects and internal reaction instability, the internal resistance exhibits multiple fluctuations within a short period, forming a multi-peaked distribution. When capacity and internal resistance characteristics at this stage are used together for health status determination, the relatively flat capacity information coupled with the multiple fluctuations in internal resistance information will lead to various feature combinations at different time points. This results in multiple feasible state derivation paths during the determination process, causing the same battery to correspond to different determination results at similar times, ultimately leading to non-unique output results and affecting the consistency of the determination.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a BMS battery health status determination method that integrates capacity and internal resistance characteristics, so as to solve the problems in the background art mentioned above.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a BMS battery health status determination method integrating capacity and internal resistance characteristics, comprising the following steps:

[0008] Obtain the battery capacity increment slope, internal resistance peak interval duration, and internal resistance peak amplitude difference ratio within the charging end interval. Divide the capacity change convergence segment along the time series and mark the internal resistance multi-peak dense segment within the capacity change convergence segment.

[0009] Within the capacity change convergence segment, for the densely populated multi-peak section of internal resistance, non-uniform stretching processing is performed on the interval duration of the internal resistance peaks, and the peak positions are rearranged according to the ratio of the amplitude difference of the internal resistance peaks. At the same time, the capacity increment slope value within the capacity change convergence segment is reversed and pulled, and the capacity response range with difference recognition capability is output.

[0010] Based on the capacity response interval, the ratio of the peak amplitude difference of internal resistance within the capacity change convergence segment is continuously raised and identified. The time segment with continuously increasing elevation amplitude is extracted as the dominant change interval. The non-dominant peak positions are progressively migrated to adjacent time segments along the capacity change convergence segment to construct a single dominant advance trajectory.

[0011] The capacity response interval is retrieved along the single dominant driving trajectory. The slope values ​​of the capacity increment within the capacity change convergence segment are reordered in direction. The smooth change segment is segmented and offset according to the dominant change interval. The interval duration of the internal resistance peak is compressed and converged into a synchronously advancing coordinated change trajectory.

[0012] Based on the coordinated change trajectory, the slope value of capacity increment and the ratio of peak internal resistance amplitude difference within the capacity change convergence segment are mapped segment by segment, and the change relationship of each time segment is merged into a single judgment trajectory to output the battery health status judgment result.

[0013] Preferably, the steps for calibrating the multi-peak dense region of internal resistance within the capacity change convergence segment are as follows:

[0014] Record current, voltage and cumulative charge amount, calculate the capacity change rate to obtain the capacity increment slope value, extract the internal resistance response and identify the peak position, calculate the time difference between adjacent peaks to obtain the internal resistance peak interval duration and the amplitude difference to obtain the internal resistance peak amplitude difference ratio.

[0015] Compare the slope values ​​of capacity increments in chronological order, extract continuous intervals with consistent change direction and amplitude within a set range, filter out the convergent segments of capacity change and record the time range;

[0016] Around the set of internal resistance peaks within the convergence segment of capacity change, identify the set of continuous peaks with time intervals within a set range and amplitude differences that change progressively, and label them as the dense segment of internal resistance multi-peaks;

[0017] Record the time range of the densely populated multi-peak internal resistance section, align the corresponding capacity increment slope values ​​point by point, and number each section in chronological order and associate the interval between internal resistance peaks with the ratio of the amplitude difference between internal resistance peaks.

[0018] Preferably, the capacity response range output steps are as follows:

[0019] Extract the time position and corresponding amplitude of each peak in the densely populated section of internal resistance within the capacity change convergence segment, and read the time interval and amplitude difference ratio of adjacent peaks. Arrange them in time order and correspond them point by point with the capacity increment slope value. At the same time, divide the time segment and assign it a label.

[0020] The peak amplitude difference ratio of the internal resistance corresponding to the time segment is sorted and divided into continuous intervals. Each interval is matched with the time adjustment rule, and the peak interval of the internal resistance is adjusted accordingly and the peak time position is updated to maintain the consistency of the time order.

[0021] The peak positions are rearranged according to the updated peak time positions, and the segments are divided according to the direction of change of amplitude ratio. The order of peak positions within the segments is adjusted and spliced ​​together according to the original time order, while maintaining the correspondence with the capacity increment slope value.

[0022] Using the rearranged peak position as the adjustment point, the capacity increment slope value is boosted in the opposite direction, and the expansion range is determined according to the position of the amplitude difference ratio and adjusted point by point.

[0023] The adjusted capacity increment slope value is scanned, and time segments with the same direction of change and corresponding to the order of change of amplitude ratio are extracted, sorted and merged to obtain the capacity response interval.

[0024] Preferably, during the non-uniform stretching process of the internal resistance peak interval, the time adjustment rules are matched according to the sorting results of the internal resistance peak amplitude difference ratio. The length of each time segment is adjusted accordingly and the peak time position is updated synchronously to keep the peak time order from overlapping. At the same time, the peak position is divided into segments and rearranged according to the direction of amplitude difference ratio change. The capacity increment slope value corresponding to the peak position is used as the adjustment point to perform reverse lifting traction and determine the expansion range, thereby obtaining the capacity response interval.

[0025] Preferably, the steps for constructing a single lead actor's trajectory are as follows:

[0026] Extract the proportion of the peak internal resistance amplitude difference for each time segment within the capacity response interval and arrange them in chronological order. Record the difference between adjacent amplitude differences and divide the continuous intervals, then mark the candidate segments.

[0027] For changes in the amplitude ratio within candidate segments, identify time segments that are continuously increasing and whose differences fall within the same interval, mark the segments with elevation changes, and record the range of their numbers.

[0028] The amplitude difference ratio of the uplift change segments is statistically analyzed and sorted. The combination of segments with continuous time and progressive difference is extracted to determine the dominant change interval and record the time number.

[0029] Around the dominant change range, non-dominant peaks are migrated in chronological order, and the number of time segments is determined by sorting the positions according to the amplitude difference ratio and updating the numbers.

[0030] All peak positions after migration are sorted out, rearranged by time number, and the amplitude difference ratio is checked to obtain the single main trajectory.

[0031] Preferably, the amplitude difference ratio of the internal resistance peak corresponding to the dominant change interval is in a continuous progressive relationship on the time axis. The non-dominant peak positions are migrated according to the original time sequence, with the migration direction pointing towards the dominant change interval. The migration process is adjusted according to the amplitude difference ratio to sort the positions and adjust the number of time segments. The peak positions after migration are updated with time numbers, while keeping the time sequence of each peak from overlapping.

[0032] Preferably, the steps for generating the cooperative change trajectory are as follows:

[0033] Read the time segment number of the capacity response interval and remap it according to the peak time order in the single master track. Merge adjacent time segments to the corresponding peak nodes and insert unmerged segments. At the same time, arrange the capacity increment slope values ​​according to the new time order to establish a correspondence.

[0034] Based on the direction of change of the capacity increment slope value, the inconsistent segments are rearranged and divided into smooth change segments. The smooth change segments are then segmented and offset according to the position of the dominant change interval, and the time segment positions are adjusted while maintaining the consistency of the internal order of the segments.

[0035] Read the duration of the interval between adjacent peaks and determine the compression ratio according to the sorting position to perform segment-by-segment shortening. At the same time, update the peak time position and match it with the capacity increment slope value point by point, so that the change direction of the two is consistent and the time number is matched, thus obtaining the coordinated change trajectory.

[0036] Preferably, the time segment merging adopts the peak node neighborhood expansion method. The expansion range is determined by the sorting position of the internal resistance peak amplitude difference ratio. The capacity increment slope value is boosted in reverse order from node to outward. At the same time, the compressed peak time position is matched with the corresponding capacity increment slope value point by point to keep the direction of change consistent and the time numbering continuous.

[0037] Preferably, based on the joint mapping relationship between the capacity increment slope value and the ratio of the peak amplitude difference of internal resistance, a single judgment trajectory is obtained and the health status result is output through segment division and judgment label merging processing. The steps are as follows:

[0038] Read the slope value of capacity increment and the ratio of the peak amplitude difference of internal resistance for each time segment within the coverage area of ​​the coordinated change trajectory, number and bind them according to the time progression order, and record the change direction of adjacent time segments;

[0039] By combining the direction of change of the slope value of capacity increment with the direction of change of the ratio of the peak value difference of internal resistance, the continuous and consistent time segments are divided, the segments that meet the number requirements of time segments are extracted and the number range is recorded.

[0040] The numerical range of the ratio of the slope of the capacity increment to the peak amplitude difference of the internal resistance is extracted for each segment and divided into equal intervals. The corresponding time segments are mapped to the interval numbers and spliced ​​in a fixed order to obtain the judgment identifier.

[0041] All time segment judgment labels are organized and arranged in chronological order. Consecutive and consistent judgment labels are merged and the proportion of time segments is calculated. The judgment label with the largest proportion is selected to output the health status result.

[0042] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0043] This invention performs non-uniform stretching on the densely populated multi-peak section of internal resistance within the capacity change convergence segment and reverses the slope value of the capacity increment, so that the capacity feature and the internal resistance feature form a consistent change structure in the time dimension. This allows the capacity feature to maintain its ability to express differences even at the end of the charging stage. At the same time, the internal resistance fluctuation information is processed in an orderly manner to avoid the information superposition effect caused by multi-peak distribution. This ensures that the two types of features have a stable correspondence during the fusion process, thereby improving the ability to identify subtle changes in the health status determination process.

[0044] This invention constructs a single master-driven trajectory and further forms a synchronously advancing collaborative change trajectory, unifying multiple feature combinations originally scattered in different time segments into a single judgment trajectory. This ensures that the ratio of capacity increment slope value to internal resistance peak amplitude difference value maintains a consistent change order on the time axis, thereby eliminating the result discrepancies caused by multi-path derivation, ensuring that the output of the same battery remains consistent at adjacent times, enhancing the continuity and stability of health status judgment results, and improving the reliability of the overall evaluation results. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0046] Figure 1 This is a flowchart of the BMS battery health status determination method that integrates capacity and internal resistance characteristics according to the present invention. Detailed Implementation

[0047] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0048] This invention provides, for example Figure 1 The BMS battery health status determination method shown, which integrates capacity and internal resistance characteristics, includes the following steps:

[0049] Obtain the battery capacity increment slope, internal resistance peak interval duration, and internal resistance peak amplitude difference ratio within the charging end interval. Divide the capacity change convergence segment along the time series and mark the internal resistance multi-peak dense segment within the capacity change convergence segment.

[0050] In the BMS battery health status determination process that integrates capacity and internal resistance characteristics, the slope value of the capacity increment and related internal resistance parameters are extracted simultaneously at the end of the charging interval. Furthermore, the multi-dimensional features are uniformly time-aligned and segmented to achieve executable partitioning of capacity change convergence segments and internal resistance multi-peak dense segments, thus providing a stable data foundation for subsequent feature reconstruction. The specific steps are as follows:

[0051] The battery operation process during the final stage of charging is continuously sampled and recorded. The sampling content includes current, voltage, and cumulative charge amount, and the original data set is formed in time sequence according to fixed time intervals. In this data set, the difference between the cumulative charge amount at each sampling time and the cumulative charge amount at the previous sampling time is calculated, and combined with the time interval between the two sampling times, it is converted into the capacity change rate per unit time. This rate is recorded as the capacity increment slope value at the corresponding time position.

[0052] Simultaneously, voltage and current changes are extracted at the same time points, and the internal resistance response value is calculated through the correspondence between voltage and current changes. Then, all internal resistance response values ​​are arranged in chronological order to form a continuous change curve. The internal resistance change curve is scanned point by point along the time direction. When the internal resistance value of a certain sampling point is greater than that of the previous sampling point and the next sampling point, and the difference between the internal resistance value and its adjacent sampling points exceeds the preset minimum amplitude threshold, the sampling point is marked as a valid peak position, and the time position and corresponding amplitude of the peak are recorded. All valid peaks are arranged in chronological order, and the time difference between two adjacent peaks is calculated as the internal resistance peak interval duration. At the same time, the amplitude difference between adjacent peaks is proportionally converted, and the amplitude difference is calculated as a ratio to the amplitude of the previous peak to obtain the internal resistance peak amplitude difference ratio. This establishes a correspondence between the capacity increment slope value, the internal resistance peak interval duration, and the internal resistance peak amplitude difference ratio on a unified time axis.

[0053] For the continuous distribution of capacity increment slope values ​​along the time axis, the capacity change trend is divided into segments. Starting from the time start point, the capacity increment slope values ​​of adjacent sampling points are compared sequentially. When the change direction of capacity increment slope values ​​between multiple consecutive sampling points is consistent, and the change amplitude between adjacent sampling points is less than the preset change threshold, and the number of consecutive sampling points meets the preset minimum requirement, the continuous time interval is marked as a candidate segment.

[0054] During the scanning process, if the direction of change of the capacity increment slope value reverses or the change magnitude exceeds the preset change threshold, the segment boundary is defined at the current sampling point, and a new segment identification is restarted. All candidate segments are screened. When the capacity increment slope value in a candidate segment does not exceed the preset abrupt change threshold in the entire interval, the candidate segment is determined as the capacity change convergence segment, and its start time and end time are recorded. At the same time, the capacity increment slope value in the segment is uniformly marked so that the capacity change convergence segment has clear boundaries and continuity on the time axis.

[0055] Within the defined capacity change convergence segment, local density identification is performed on the interval duration of internal resistance peaks and the ratio of internal resistance peak amplitude differences. All internal resistance peaks belonging to the time range of this segment are extracted and arranged in chronological order. For the time interval between adjacent peaks, when the interval duration between multiple consecutive peaks is less than the preset time interval threshold and the number of consecutive peaks reaches the preset density judgment number, the set of consecutive peaks is marked as a candidate clustering area.

[0056] Within the candidate cluster region, the amplitude difference ratio of adjacent peaks is compared item by item. When the amplitude difference ratio maintains a progressive change relationship among multiple consecutive peaks, and the change amplitude between any adjacent amplitude difference ratios does not exceed the preset ratio change threshold, the candidate cluster region is determined as the internal resistance peak cluster region. Subsequently, the internal resistance peak cluster region is matched with the capacity change convergence segment to which it belongs in terms of time range. Only the set of peaks that completely fall within the time interval of the capacity change convergence segment is retained, and this set is marked as the internal resistance multi-peak dense segment, thereby completing the segment division of the internal resistance characteristics in the context of capacity change convergence.

[0057] The distribution of the calibrated multi-peak dense sections of internal resistance in the capacity change convergence segment is uniformly organized. By recording the start and end times of each multi-peak dense section of internal resistance, and aligning each sampling point in the time interval with the corresponding capacity increment slope value, a one-to-one correspondence is established between the position of the internal resistance peak and the capacity change state. For multiple multi-peak dense sections of internal resistance existing in the same capacity change convergence segment, they are numbered in chronological order so that each section forms a continuous arrangement structure on the time axis.

[0058] Simultaneously, the peak resistance interval duration and peak resistance amplitude difference ratio within each segment are associated and stored with the corresponding capacity increment slope value, so that each time segment contains a complete joint feature expression of capacity and internal resistance. Through the above processing, the discrete feature points within the capacity change convergence segment are transformed into a structured data form with clear segment division and hierarchical relationship, thereby providing a consistent and executable data foundation for subsequent non-uniform stretching of the peak resistance interval duration and capacity response interval extraction.

[0059] Within the capacity change convergence segment, for the densely populated multi-peak section of internal resistance, non-uniform stretching processing is performed on the interval duration of the internal resistance peaks, and the peak positions are rearranged according to the ratio of the amplitude difference of the internal resistance peaks. At the same time, the capacity increment slope value within the capacity change convergence segment is reversed and pulled, and the capacity response range with difference recognition capability is output.

[0060] Within the capacity change convergence segment, for the densely populated multi-peak section of internal resistance, the duration of the internal resistance peak interval is redistributed, and the slope of the capacity increment is adjusted accordingly. This causes the capacity characteristics and internal resistance characteristics to form a directionally correlated change structure on a unified time axis, thereby outputting a capacity response interval with differential recognition capabilities. The specific steps are as follows:

[0061] Select the defined capacity change convergence segment, and extract the time position and corresponding amplitude of all peaks in the dense multi-peak section of internal resistance point by point within the time range of the capacity change convergence segment. At the same time, read the internal resistance peak interval between adjacent peaks and the internal resistance peak amplitude difference ratio between adjacent peak amplitudes. Establish a multi-attribute record for each peak, including time position, amplitude, time interval before and after, and amplitude difference ratio.

[0062] Subsequently, all peak values ​​are arranged in chronological order, and the arrangement results are matched point by point with the capacity increment slope values ​​at the corresponding time points in the capacity change convergence segment, so that each peak position is associated with the corresponding capacity change state. At the same time, the time interval between adjacent peak values ​​is divided into independent time segments, and each time segment is assigned a unique identifier, so that the peak interval duration has a clear time range definition, thereby forming a unified data structure that can be used for subsequent processing.

[0063] For the internal resistance peak amplitude difference ratio corresponding to each time segment, all amplitude difference ratio values ​​are sorted according to their distribution range in the current capacity change convergence segment. Based on the sorting results, the amplitude difference ratio is divided into a preset number of continuous intervals. Each interval corresponds to a fixed time adjustment rule. The number of intervals is determined in the initialization stage according to the number of peak values ​​in the capacity change convergence segment and remains unchanged in the same capacity change convergence segment.

[0064] For each time segment, the interval duration of the internal resistance peak is adjusted according to the interval to which the amplitude difference ratio belongs. When the amplitude difference ratio is in the upper-ranked interval, the corresponding time segment is shortened according to the preset compression ratio. When the amplitude difference ratio is in the lower-ranked interval, the corresponding time segment is extended according to the preset expansion ratio. After each adjustment, the time positions of all subsequent peaks are updated sequentially to ensure that the order of all peaks on the time axis remains consistent and does not overlap, thereby completing the non-uniform stretching of the internal resistance peak interval duration.

[0065] After the non-uniform stretching process is completed for the peak interval of internal resistance, all peaks are rearranged according to the updated time position and sorted in order according to the corresponding internal resistance peak amplitude difference ratio, so that the amplitude difference ratio shows a monotonically changing relationship in the same continuous interval. At the same time, the position where the change direction of the amplitude difference ratio changes is used as the segmentation node to divide the peak set into segments.

[0066] For each segment, the peak time positions within the segment are repositioned and arranged sequentially according to the direction of amplitude difference ratio change. All segments are then spliced ​​together according to the original time order, so that the entire internal resistance multi-peak dense segment forms a continuous structure composed of multiple ordered sub-segments on the time axis. At the same time, it is ensured that each rearranged peak position still falls within the time range of the capacity change convergence segment and maintains a corresponding relationship with the capacity increment slope value at the corresponding time point, thereby completing the peak position rearrangement process.

[0067] Using the rearranged peak positions as adjustment points, a reverse boosting traction process is performed on the capacity increment slope value within the capacity change convergence segment. The capacity increment slope value corresponding to each peak position is used as the central reference point. Based on the sorting position of the internal resistance peak amplitude difference ratio corresponding to the peak in the current capacity change convergence segment, the time range for forward and backward expansion of the central point is determined. The earlier the amplitude difference ratio sorting position is, the larger the corresponding expansion time range is, and the later the sorting position is, the smaller the corresponding expansion time range is.

[0068] After determining the expansion range, the slope value of the capacity increment within the range is adjusted point by point so that the slope value of the capacity increment at each sampling point increases in a way that gradually decreases from the center point outwards. This forms a directional capacity change structure near each peak position, and the structure is kept consistent with the order of change of the ratio of the peak amplitude difference of the internal resistance. At the same time, within the same capacity change convergence segment, the adjustment rules corresponding to each peak maintain a unified mapping relationship.

[0069] The slope value of the capacity increment after the reverse lifting traction process is scanned as a whole. Time segments on the time axis where the slope value of the capacity increment changes continuously and the change amplitude remains consistent are extracted one by one. The change trend of the slope value of the capacity increment in each time segment is matched with the change trend of the ratio of the peak amplitude difference of the internal resistance in the corresponding time range. When the change direction of the two is consistent and the change order maintains the corresponding relationship, the time segment is marked as an effective response segment.

[0070] All valid response segments are organized in chronological order, and adjacent segments with consistent trends are merged to form a continuous time interval structure. Finally, these continuous time intervals are defined as capacity response intervals, so that segments with difference recognition capabilities in the capacity change convergence segment can be fully expressed.

[0071] Based on the capacity response interval, the ratio of the peak amplitude difference of internal resistance within the capacity change convergence segment is continuously raised and identified. The time segment with continuously increasing elevation amplitude is extracted as the dominant change interval. The non-dominant peak positions are progressively migrated to adjacent time segments along the capacity change convergence segment to construct a single dominant advance trajectory.

[0072] Within the capacity response range, by refining the analysis of the relationship between the peak amplitude difference ratio of internal resistance over time and continuously reordering the peak distribution positions, a time segment with a continuously rising characteristic can be extracted from the capacity change convergence segment as the dominant change range. Simultaneously, other peak values ​​are gradually guided to the vicinity of this range, thereby constructing a single dominant trajectory. The specific steps are as follows:

[0073] Select the determined capacity response interval, and extract the peak internal resistance amplitude difference ratio of each time segment within the capacity change convergence segment corresponding to the capacity response interval. At the same time, record the time sequence position of each time segment in the capacity change convergence segment, arrange all amplitude difference ratios in chronological order, and assign consecutive numbers to each time segment.

[0074] During the arrangement process, the amplitude difference ratios of two adjacent time segments are compared item by item, and the difference between adjacent amplitude difference ratios is recorded. Based on the maximum and minimum values ​​of all amplitude difference ratios within the current capacity response interval, the numerical range is divided into a fixed number of continuous intervals, so that any adjacent difference can be classified into a certain interval. When the amplitude difference ratios of multiple consecutive time segments change in the same direction and their adjacent differences fall into the same difference interval, the consecutive time segments are divided into initial candidate segments, and the start number and end number are recorded for each candidate segment, so that the amplitude difference ratio changes within the capacity response interval form a segment structure with a unified division standard.

[0075] Further screening is performed on the amplitude difference ratio changes within each initial candidate segment. Within each candidate segment, the amplitude difference ratio between adjacent time segments is checked point by point. When the amplitude difference ratios of multiple consecutive time segments increase sequentially and the corresponding adjacent differences always fall within the same difference range, and the number of such consecutive time segments reaches the preset minimum number of segments, the consecutive time segments are marked as an upward change segment, and the number range of all time segments within this segment is recorded. Candidate segments that do not meet the condition of continuous increase or whose differences span different intervals are not included in the upward change segment processing. This ensures that the upward change segment retains only a set of time segments with consistent change direction and uniform change magnitude, thereby ensuring that the upward change segment has consistent change characteristics.

[0076] Around the multiple uplift change segments that have been obtained, the amplitude difference ratio changes in each segment are statistically analyzed one by one. By calculating the amplitude difference ratio difference between the start time segment and the end time segment in each uplift change segment, all segments are sorted according to the size of this difference. At the same time, the distribution relationship of the sorted segments on the time axis is checked. When multiple segments are arranged continuously in time and their amplitude difference ratio differences show a progressive relationship, the continuous segments are extracted.

[0077] When there are multiple combinations of segments that meet the conditions, the combination of segments with the longest time span and the continuous progressive relationship of the internal amplitude difference ratio is selected as the unique dominant change interval. The start and end numbers of the dominant change interval are recorded so that it has a unique identifier in the capacity change convergence segment.

[0078] Based on the established dominant change interval, a progressive migration process is performed on the peak positions that are not included in the dominant change interval within the capacity change convergence segment. Each non-dominant peak is processed one by one according to its original time sequence. For peaks located before the dominant change interval, their time positions are gradually moved backward by time segments as the smallest unit. For peaks located after the dominant change interval, their time positions are gradually moved forward by time segments as the smallest unit.

[0079] During the movement, the number of time segments to be moved is determined according to the position of the internal resistance peak amplitude difference ratio corresponding to the peak in the sort of all amplitude difference ratios. This ensures that peaks with earlier sort positions correspond to smaller movement step sizes, while peaks with later sort positions correspond to larger movement step sizes. At the same time, the movement range of each peak is limited to the time boundary of the capacity change convergence segment. The time number of the peak is updated after each movement, and it is ensured that the time order between all peaks does not overlap, so that non-dominant peaks gradually concentrate in the vicinity of the dominant change interval.

[0080] After the migration process is completed, all peak positions are uniformly organized. The peaks in the dominant change interval and the non-dominant peaks after migration are arranged in a new time number order. The amplitude difference ratio change relationship between adjacent peaks is checked point by point, so that all peaks show a continuous change trend in a single direction along the time axis. At the same time, the amplitude difference ratio change path corresponding to each time segment is continuously marked, so that the amplitude difference ratio change of the internal resistance peak in the entire capacity change convergence segment forms a unique evolution direction, thereby constructing a single dominant evolution trajectory, so that the original multi-path change relationship converges into a single path expression in the time dimension.

[0081] The capacity response interval is retrieved along the single dominant driving trajectory. The slope values ​​of the capacity increment within the capacity change convergence segment are reordered in direction. The smooth change segment is segmented and offset according to the dominant change interval. The interval duration of the internal resistance peak is compressed and converged into a synchronously advancing coordinated change trajectory.

[0082] Given that a single dominant driving trajectory has been established, the capacity response interval is reconstructed by backtracking, and the slope of the capacity increment and the interval duration of the internal resistance peak are adjusted in a coordinated manner within the capacity change convergence segment. This ensures that the capacity characteristics and internal resistance characteristics have a consistent driving relationship in the time dimension, thus converging into a synchronously advancing coordinated change trajectory. The specific steps are as follows:

[0083] Around a single dominant trajectory, the time distribution of the capacity response interval within the capacity change convergence segment is back-referenced. Within the time range of the capacity change convergence segment, the time segment number corresponding to the capacity response interval is read point by point, and each time segment is remapped according to the time order of the peak positions in the single dominant trajectory, ensuring that the time distribution of the capacity response interval is consistent with the change order of the single dominant trajectory. During the mapping process, each peak position in the single dominant trajectory is used as an alignment node. The time segments within the capacity response interval closest to this node are merged into that node, and a fixed number of time segments are extended forward and backward from this node, forming a continuous coverage interval in the neighborhood of the node. The number of extended time segments is determined by the ranking position of the peak amplitude difference ratio of the corresponding internal resistance peak within the capacity change convergence segment, so that peaks with earlier ranking positions correspond to larger extension ranges, and peaks with later ranking positions correspond to smaller extension ranges.

[0084] For time segments that are not merged into the neighborhood of any peak node, they are inserted between adjacent peak nodes according to their original time order, so that all time segments on the time axis of the entire capacity change convergence segment are corresponding to the single master trajectory, and the corresponding capacity increment slope values ​​are rearranged according to the new time order, so that the capacity increment slope values ​​are corresponding to the single master trajectory point by point.

[0085] Under the premise of completing the back-reference processing in the capacity response interval and establishing a time correspondence with the single master pre-trajectory, the capacity increment slope value in the capacity change convergence segment is subjected to direction reordering processing. By comparing the change direction of the capacity increment slope value of adjacent time segments point by point, the time segments whose change direction is inconsistent with the single master pre-trajectory are rearranged so that their change direction is consistent with the master trajectory. In the process, the continuous time segments in which the change amplitude of the capacity increment slope value in the capacity change convergence segment falls into the same difference interval are divided into smooth change segments. The difference interval is divided into multiple equal intervals according to the maximum and minimum values ​​of all capacity increment slope values ​​in the capacity change convergence segment, and one of the intervals is selected as the judgment range.

[0086] Subsequently, taking the dominant change interval in the single master trajectory as a reference, each smooth change segment is segmented and offset according to its relative position in the capacity change convergence segment. For smooth change segments located before the dominant change interval, they are moved backward by a certain number of time segments. For smooth change segments located after the dominant change interval, they are moved forward by a certain number of time segments. The number of time segments moved is determined according to the length ratio of the smooth change segment in the capacity change convergence segment, so that segments with a larger length ratio correspond to more time segments. At the same time, all movement operations are limited to the time boundary of the capacity change convergence segment, and the time order within the segment is kept unchanged during the movement, so that the capacity increment slope value forms a continuous and consistent distribution structure on the time axis.

[0087] After the capacity increment slope value is reordered and segmented for gradual change, the interval duration of the internal resistance peak within the capacity change convergence segment is compressed. This is done by reading the time interval between each adjacent peak and determining the compression ratio based on the order of the time interval within the capacity change convergence segment. Time intervals with earlier order correspond to a larger compression ratio, while those with later order correspond to a smaller compression ratio, thus allowing different time intervals to be shortened differently according to a unified rule. During the compression process, a single time segment is used as the smallest adjustment unit, and each time interval is shortened segment by segment. After each shortening, the time position of the corresponding peak is updated, ensuring that the order of all peaks on the time axis remains consistent and that they do not overlap.

[0088] Simultaneously, the compressed peak time position is matched point by point with the capacity increment slope value that has been directionally adjusted, so that each peak position corresponds to a unique capacity change state. The capacity increment slope value and the internal resistance peak interval duration are aligned in the same direction on the time axis and the change order is maintained. The synchronous advancement relationship is defined as the two changing in the same direction in continuous time segments and the corresponding time segment numbers are matched one by one. This makes the capacity characteristics and internal resistance characteristics in the capacity change convergence segment form a unified advancement structure in the time dimension, and finally converge into a synchronously advancing coordinated change trajectory.

[0089] Based on the coordinated change trajectory, the slope value of capacity increment and the ratio of peak internal resistance amplitude difference within the capacity change convergence segment are mapped segment by segment, and the change relationship of each time segment is merged into a single judgment trajectory to output the battery health status judgment result.

[0090] Given that a coordinated change trajectory has been established, the slope of the capacity increment and the ratio of the peak internal resistance difference within the capacity change convergence segment are mapped segment by segment. The changes in each time segment are then uniformly merged, enabling multi-source features to form a single judgment trajectory in the time dimension, thereby outputting a stable and consistent battery health status judgment result. The specific steps are as follows:

[0091] Select the capacity change convergence segment covered by the coordinated change trajectory, and read the capacity increment slope value and the internal resistance peak amplitude difference ratio corresponding to each time segment one by one within this time range. Using the time progression order in the coordinated change trajectory as a reference, renumber all time segments so that each time segment corresponds to a unique time position identifier. During the numbering process, bind the capacity increment slope value and the internal resistance peak amplitude difference ratio according to the same time number so that each time segment has two characteristic parameters at the same time.

[0092] Simultaneously, the direction of change of the slope value of capacity increment and the direction of change of the ratio of the peak amplitude difference of internal resistance between adjacent time segments are compared point by point, and the direction of change is marked as rising or falling, so that each time segment contains the current feature value and the change relationship information relative to the previous time segment, thereby forming a time segment expression structure containing dual features and change direction.

[0093] In the set of time segments that have completed dual feature binding, the capacity change convergence segments are divided into segments one by one. Starting from the initial time segment, the scan proceeds sequentially. When the direction of change of the capacity increment slope value between adjacent time segments is consistent with the direction of change of the ratio of the peak value difference of internal resistance, the time segment is continued to be included in the current segment. When the change direction of any time segment is inconsistent with the previous time segment, the segment boundary is divided at that position, and a new segment division begins.

[0094] During the partitioning process, the number of time segments within each segment is counted, and only segments with the number of time segments reaching the preset minimum requirement are retained as valid mapping segments. The start and end numbers of each valid mapping segment are recorded, so that the capacity change convergence segment is divided into multiple mapping segments with the same internal change direction, thereby ensuring the consistency of the dual feature changes within the segment.

[0095] For each valid mapping segment, the capacity increment slope value and the internal resistance peak amplitude difference ratio are mapped segment by segment. Within each mapping segment, the maximum and minimum values ​​of the capacity increment slope value and the internal resistance peak amplitude difference ratio are extracted, and their respective value ranges are divided into a fixed number of equal intervals, so that each interval corresponds to the same value span. Subsequently, for each time segment within the segment, its capacity increment slope value is mapped to the corresponding interval number, and its internal resistance peak amplitude difference ratio is also mapped to the corresponding interval number, and a one-to-one correspondence is established according to the interval numbers, so that the interval number of the capacity increment slope value and the interval number of the internal resistance peak amplitude difference ratio are synchronized. On this basis, the interval number corresponding to the capacity increment slope value and the interval number corresponding to the internal resistance peak amplitude difference ratio are concatenated in a fixed order to form the judgment identifier of the time segment, so that each time segment has a unique dual-feature combination expression.

[0096] The time segments within all mapped sections are integrated in chronological order. The judgment identifiers corresponding to each time segment are arranged sequentially along the time axis. Intervals with consistent judgment identifiers in consecutive time segments are merged, forming several consecutive judgment intervals on the time axis for the capacity change convergence segment. For each judgment interval, the proportion of its time segment count to the total number of time segments in the entire capacity change convergence segment is calculated. This proportion is used as the weight value of the judgment interval. The weight values ​​of all judgment intervals are compared, and the judgment identifier corresponding to the judgment interval with the largest weight value is selected as the final judgment result. Subsequently, based on the correspondence of the judgment identifier in the preset health status interval division table, the battery health status judgment result is output, ensuring the uniqueness and consistency of the health status output within the capacity change convergence segment.

[0097] This invention performs non-uniform stretching on the densely populated multi-peak section of internal resistance within the capacity change convergence segment and reverses the slope value of the capacity increment, so that the capacity feature and the internal resistance feature form a consistent change structure in the time dimension. This allows the capacity feature to maintain its ability to express differences even at the end of the charging stage. At the same time, the internal resistance fluctuation information is processed in an orderly manner to avoid the information superposition effect caused by multi-peak distribution. This ensures that the two types of features have a stable correspondence during the fusion process, thereby improving the ability to identify subtle changes in the health status determination process.

[0098] This invention constructs a single master-driven trajectory and further forms a synchronously advancing collaborative change trajectory, unifying multiple feature combinations originally scattered in different time segments into a single judgment trajectory. This ensures that the ratio of capacity increment slope value to internal resistance peak amplitude difference value maintains a consistent change order on the time axis, thereby eliminating the result discrepancies caused by multi-path derivation, ensuring that the output of the same battery remains consistent at adjacent times, enhancing the continuity and stability of health status judgment results, and improving the reliability of the overall evaluation results.

[0099] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for determining the health status of a battery via a BMS that integrates capacity and internal resistance characteristics, characterized in that, Includes the following steps: Obtain the battery capacity increment slope, internal resistance peak interval duration, and internal resistance peak amplitude difference ratio within the charging end interval. Divide the capacity change convergence segment along the time series and mark the internal resistance multi-peak dense segment within the capacity change convergence segment. Within the capacity change convergence segment, for the densely populated multi-peak section of internal resistance, non-uniform stretching processing is performed on the interval duration of the internal resistance peaks, and the peak positions are rearranged according to the ratio of the amplitude difference of the internal resistance peaks. At the same time, the capacity increment slope value within the capacity change convergence segment is reversed and pulled, and the capacity response range with difference recognition capability is output. Based on the capacity response interval, the ratio of the peak amplitude difference of internal resistance within the capacity change convergence segment is continuously raised and identified. The time segment with continuously increasing elevation amplitude is extracted as the dominant change interval. The non-dominant peak positions are progressively migrated to adjacent time segments along the capacity change convergence segment to construct a single dominant advance trajectory. The capacity response interval is retrieved along the single dominant driving trajectory. The slope values ​​of the capacity increment within the capacity change convergence segment are reordered in direction. The smooth change segment is segmented and offset according to the dominant change interval. The interval duration of the internal resistance peak is compressed and converged into a synchronously advancing coordinated change trajectory. Based on the coordinated change trajectory, the slope value of capacity increment and the ratio of peak internal resistance amplitude difference within the capacity change convergence segment are mapped segment by segment, and the change relationship of each time segment is merged into a single judgment trajectory to output the battery health status judgment result.

2. The BMS battery health status determination method based on the fusion of capacity and internal resistance characteristics according to claim 1, characterized in that, The steps for calibrating the densely packed multi-peak internal resistance region within the capacity variation convergence segment are as follows: Record current, voltage and cumulative charge amount, calculate the capacity change rate to obtain the capacity increment slope value, extract the internal resistance response and identify the peak position, calculate the time difference between adjacent peaks to obtain the internal resistance peak interval duration and the amplitude difference to obtain the internal resistance peak amplitude difference ratio. Compare the slope values ​​of capacity increments in chronological order, extract continuous intervals with consistent change direction and amplitude within a set range, filter out the convergent segments of capacity change and record the time range; Around the set of internal resistance peaks within the convergence segment of capacity change, identify the set of continuous peaks with time intervals within a set range and amplitude differences that change progressively, and label them as the dense segment of internal resistance multi-peaks; Record the time range of the densely populated multi-peak internal resistance section, align the corresponding capacity increment slope values ​​point by point, and number each section in chronological order and associate the interval between internal resistance peaks with the ratio of the amplitude difference between internal resistance peaks.

3. The BMS battery health status determination method based on the fusion of capacity and internal resistance characteristics according to claim 2, characterized in that, The steps for outputting the capacity response range are as follows: Extract the time position and corresponding amplitude of each peak in the densely populated section of internal resistance within the capacity change convergence segment, and read the time interval and amplitude difference ratio of adjacent peaks. Arrange them in time order and correspond them point by point with the capacity increment slope value. At the same time, divide the time segment and assign it a label. The peak amplitude difference ratio of the internal resistance corresponding to the time segment is sorted and divided into continuous intervals. Each interval is matched with the time adjustment rule, and the peak interval of the internal resistance is adjusted accordingly and the peak time position is updated to maintain the consistency of the time order. The peak positions are rearranged according to the updated peak time positions, and the segments are divided according to the direction of change of amplitude ratio. The order of peak positions within the segments is adjusted and spliced ​​together according to the original time order, while maintaining the correspondence with the capacity increment slope value. Using the rearranged peak position as the adjustment point, the capacity increment slope value is boosted in the opposite direction, and the expansion range is determined according to the position of the amplitude difference ratio and adjusted point by point. The adjusted capacity increment slope value is scanned, and time segments with the same direction of change and corresponding to the order of change of amplitude ratio are extracted, sorted and merged to obtain the capacity response interval.

4. The BMS battery health status determination method based on the fusion of capacity and internal resistance characteristics according to claim 3, characterized in that, During the non-uniform stretching process of the internal resistance peak interval, the time adjustment rules are matched according to the sorting results of the internal resistance peak amplitude difference ratio. The length of each time segment is adjusted accordingly and the peak time position is updated synchronously to keep the peak time order from overlapping. At the same time, the peak position is divided into segments and rearranged according to the direction of amplitude difference ratio change. The capacity increment slope value corresponding to the peak position is used as the adjustment point to perform reverse lifting traction and determine the expansion range, thereby obtaining the capacity response interval.

5. The BMS battery health status determination method based on the fusion of capacity and internal resistance characteristics according to claim 3, characterized in that, The steps for constructing a single lead director's trajectory are as follows: Extract the proportion of the peak internal resistance amplitude difference for each time segment within the capacity response interval and arrange them in chronological order. Record the difference between adjacent amplitude differences and divide the continuous intervals, then mark the candidate segments. For changes in the amplitude ratio within candidate segments, identify time segments that are continuously increasing and whose differences fall within the same interval, mark the segments with elevation changes, and record the range of their numbers. The amplitude difference ratio of the uplift change segments is statistically analyzed and sorted. The combination of segments with continuous time and progressive difference is extracted to determine the dominant change interval and record the time number. Around the dominant change range, non-dominant peaks are migrated in chronological order, and the number of time segments is determined by sorting the positions according to the amplitude difference ratio and updating the numbers. All peak positions after migration are sorted out, rearranged by time number, and the amplitude difference ratio is checked to obtain the single main trajectory.

6. The BMS battery health status determination method based on integrated capacity and internal resistance characteristics according to claim 5, characterized in that, The amplitude difference ratio of the internal resistance peak corresponding to the dominant change interval is continuously progressive on the time axis. The non-dominant peak positions are migrated according to the original time sequence, with the migration direction pointing towards the dominant change interval. The migration process adjusts the number of time segments according to the amplitude difference ratio and updates the time number of the peak positions after migration, while ensuring that the time sequence of each peak does not overlap.

7. The BMS battery health status determination method based on integrated capacity and internal resistance characteristics according to claim 5, characterized in that, The steps for generating the coordinated change trajectory are as follows: Read the time segment number of the capacity response interval and remap it according to the peak time order in the single master track. Merge adjacent time segments to the corresponding peak nodes and insert unmerged segments. At the same time, arrange the capacity increment slope values ​​according to the new time order to establish a correspondence. Based on the direction of change of the capacity increment slope value, the inconsistent segments are rearranged and divided into smooth change segments. The smooth change segments are then segmented and offset according to the position of the dominant change interval, and the time segment positions are adjusted while maintaining the consistency of the internal order of the segments. Read the duration of the interval between adjacent peaks and determine the compression ratio according to the sorting position to perform segment-by-segment shortening. At the same time, update the peak time position and match it with the capacity increment slope value point by point, so that the change direction of the two is consistent and the time number is matched, thus obtaining the coordinated change trajectory.

8. The BMS battery health status determination method based on integrated capacity and internal resistance characteristics according to claim 7, characterized in that, The time segment merging adopts the peak node neighborhood expansion method. The expansion range is determined by the sorting position of the peak amplitude difference ratio of the internal resistance. The capacity increment slope value is boosted in reverse order from node to outward. At the same time, the compressed peak time position is matched point by point with the corresponding capacity increment slope value.

9. The BMS battery health status determination method based on the integration of capacity and internal resistance characteristics according to claim 7, characterized in that, To address the joint mapping relationship between the capacity increment slope and the ratio of the peak internal resistance amplitude difference, a single judgment trajectory is obtained and the health status result is output through segmentation and judgment label merging. The steps are as follows: Read the slope value of capacity increment and the ratio of the peak amplitude difference of internal resistance for each time segment within the coverage area of ​​the coordinated change trajectory, number and bind them according to the time progression order, and record the change direction of adjacent time segments; By combining the direction of change of the slope value of capacity increment with the direction of change of the ratio of the peak value difference of internal resistance, the continuous and consistent time segments are divided, the segments that meet the number requirements of time segments are extracted and the number range is recorded. The numerical range of the ratio of the slope of the capacity increment to the peak amplitude difference of the internal resistance is extracted for each segment and divided into equal intervals. The corresponding time segments are mapped to the interval numbers and spliced ​​in a fixed order to obtain the judgment identifier. All time segment judgment labels are organized and arranged in chronological order. Consecutive and consistent judgment labels are merged and the proportion of time segments is calculated. The judgment label with the largest proportion is selected to output the health status result.

Images