Multi-parameter data cleaning method and system for remote battery nuclear capacity

Abstract

This invention discloses a multi-parameter data cleaning method and system for remote battery capacity assessment, relating to the field of battery capacity assessment cleaning technology. The method includes: calculating the time-window integrated capacity based on the group current in the voltage clean frame; calculating a first ampere-hour residual based on the time-window integrated capacity and a set capacity; calculating a second ampere-hour residual based on the time-window integrated capacity and the transmitted capacity; identifying the cumulative increase pattern of the first and second ampere-hour residuals; when the first and second ampere-hour residuals continuously deviate in the same direction, inferring the group current drift value; and correcting the group current in reverse based on the group current drift value to form a current clean frame. The assessment capacity is then calculated based on the corrected group current in the current clean frame, and the cleaning category is marked. Finally, a cleaning capacity assessment message is output. This invention eliminates the continuous cumulative impact of slow group current shift during the discharge phase on the time-window integrated capacity and the assessment capacity, improving the stability of remote capacity assessment calculation and the reliability of the cleaning capacity assessment message.

Description

Technical Field

[0001] This invention relates to the field of battery capacity verification and cleaning technology, and in particular to a multi-parameter data cleaning method and system for remote battery capacity verification. Background Technology

[0002] With the increasing application of remote battery capacity verification in communication power supplies, substation DC panels, and backup power supply maintenance scenarios, related technologies typically collect capacity verification frame content such as group voltage, plate voltage, throttling voltage, group current, transmitted capacity, and capacity verification status values. Combined with a remote communication link, the capacity verification process is transmitted to the management platform, and then the capacity verification is calculated and judged according to the collection cycle. This enables maintenance personnel to complete battery discharge capacity verification and capacity status verification under off-site conditions.

[0003] In scenarios where network jitter, repeated frame transmission, frame timestamp offset, local contamination of voltage measurement points, and slow drift of group current coexist in remote capacity links, related technologies often process capacity frames by eliminating single anomalies or correcting single parameter thresholds. This can easily lead to a lack of joint constraints between the representativeness of capacity frames within the same time window, the consistency of battery pack layer voltage, and the continuity of ampere-hour integration. This is because remote capacity frames not only contain communication timing attributes, but also the hierarchical relationship of electrical measurement points composed of group voltage, plate voltage, and section voltage, as well as the capacity accumulation relationship formed by group current integration. It is difficult to locate the contamination channel and determine the source of capacity deviation based on a single parameter threshold. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multi-parameter data cleaning method for remote capacity assessment of batteries to solve the problem of capacity calculation distortion under multi-parameter contamination in remote capacity assessment.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a multi-parameter data cleaning method for remote capacity verification of batteries, comprising: acquiring capacity verification frames during the remote capacity verification process; performing format verification and range verification on the capacity verification frames to form standard capacity verification frames; mapping the standard capacity verification frames to a unified time axis based on the capacity verification start time; determining the communication reliability value based on the continuous state of the frame sequence number of the standard capacity verification frames, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the receiving time and the acquisition timestamp; sorting the standard capacity verification frames within the same time window according to the communication reliability value; retaining the standard capacity verification frame with the highest ranking; and performing capacity verification stage calibration on the standard capacity verification frames according to the capacity verification status value to form aligned capacity verification frames; and calculating the group based on the battery pack measurement point topology corresponding to the aligned capacity verification frames. The residual voltage is calculated by first calculating the plate voltage residual when the group voltage residual exceeds the limit, and then calculating the section voltage residual when the plate voltage residual exceeds the limit. The contamination channel is located sequentially according to the residual level, and the abnormal voltage is reconstructed according to the section voltage sorting sequence to form a voltage clean frame. The time window integration capacity is calculated based on the group current in the voltage clean frame. The first ampere-hour residual is calculated based on the time window integration capacity and the set capacity. The second ampere-hour residual is calculated based on the time window integration capacity and the upload capacity. The cumulative increase pattern of the first ampere-hour residual and the second ampere-hour residual is identified. When the first ampere-hour residual and the second ampere-hour residual deviate continuously in the same direction, the group current drift value is reversed, and the group current is corrected in reverse according to the group current drift value to form a current clean frame. The core capacity is calculated based on the corrected group current in the current clean frame, the cleaning category is marked, and the cleaning core capacity message is output.

[0007] As a preferred embodiment of the multi-parameter data cleaning method for remote battery capacity verification described in this invention, the standard capacity verification frame is formed by matching the position of the frame structure fields of the capacity verification frame to determine that the capacity verification frame meets the remote capacity verification transmission format, then performing consistency verification between the verification field and the payload to confirm that the capacity verification frame has not been damaged during remote transmission, and then comparing the capacity verification parameters with the corresponding allowable capacity verification range, marking and removing capacity verification frames with capacity verification frame cleaning abnormalities.

[0008] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity assessment of batteries described in this invention, the specific steps for retaining the highest-ranked standard capacity assessment frame are as follows: Identify the first frame of the core capacity execution from the standard core capacity frame, determine the acquisition timestamp of the first frame of the core capacity execution as the core capacity start time, and form the core capacity time axis reference; Set the time axis reference of the nuclear capacity to a unified time axis zero point, convert the collection timestamp of the standard nuclear capacity frame into a relative nuclear capacity time, and classify it into the corresponding unified time window according to the remote nuclear capacity collection cycle to form a time windowed standard nuclear capacity frame. The frame sequence trust item is determined based on the continuous state of the frame sequence number of the time-windowed standard core frame, the timestamp trust item is determined based on the offset between the relative core time and the unified time window center time, the link trust item is determined based on the difference between the receiving time and the acquisition timestamp, and the communication trust value is formed by combining them according to a fixed priority. The time-windowed standard capacity frames are sorted according to the communication trust value. The time-windowed standard capacity frame with the highest ranking is retained as the representative frame of the time window, and the corresponding unified time window is determined as the effective unified time window.

[0009] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity verification of batteries described in this invention, the specific steps for performing capacity verification stage calibration on standard capacity verification frames based on capacity verification status values ​​to form aligned capacity verification frames are as follows: The parsing window represents the core capacity status value in the frame, and the core capacity preparation state corresponds to the core capacity preparation stage, the constant current discharge state corresponds to the core capacity discharge stage, the discharge stop state corresponds to the core capacity cutoff stage, the capacity calculation state corresponds to the capacity calculation stage, and the core capacity end state corresponds to the core capacity end stage, thus forming the candidate core capacity stage. Compare the candidate capacity phase with the confirmed capacity phase of the previous effective unified time window. If the candidate capacity phase has not changed, a capacity phase identifier is generated based on the candidate capacity phase. If the candidate capacity phase has changed, the continuity is confirmed through the candidate capacity phase of the next effective unified time window, and a capacity phase identifier is generated based on the continuity confirmation result. The capacity stage identifier, the valid unified time window number where the time window representative frame is located, and the relative capacity time are bound to the time window representative frame, and the capacity parameters in the time window representative frame are retained to form an aligned capacity frame.

[0010] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity verification of batteries described in this invention, the steps of calculating the group voltage residual based on the battery pack measurement point topology corresponding to the aligned capacity verification frame, calculating the plate voltage residual when the group voltage residual exceeds the limit, and calculating the section voltage residual when the plate voltage residual exceeds the limit are as follows: Using the valid unified time window number in the aligned capacity frame as an index, extract the group pressure value, plate pressure value and section pressure value from the aligned capacity frame corresponding to the same valid unified time window, and retrieve the battery pack measurement point topology according to the battery pack identifier corresponding to the aligned capacity frame; Based on the battery pack test point topology, determine the plate voltage test points belonging to the same group of test points. Within the same effective unified time window, calculate the difference between the group voltage value and the sum of the plate voltage values ​​belonging to the same group of test points to form the group voltage residual. When the group voltage residual exceeds the limit, determine the plate voltage test point corresponding to the over-limit group voltage test point according to the voltage test point attribution table. The difference between the plate voltage value and the sum of the section voltage values ​​belonging to the same plate voltage measurement point is calculated to form the plate voltage residual. When the plate voltage residual exceeds the limit, the section voltage measurement point corresponding to the plate voltage measurement point that exceeds the limit is determined according to the voltage measurement point assignment table. A throttle pressure sorting sequence is formed by connecting the throttle pressure measurement points in series. Candidate abnormal throttle pressure measurement points are determined by Hampel-type robust window verification. The difference between the current throttle pressure value and the throttle pressure reference value corresponding to the candidate abnormal throttle pressure measurement point is calculated to form the throttle pressure residual.

[0011] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity assessment of batteries described in this invention, the steps of sequentially locating contamination channels according to residual levels and reconstructing abnormal voltages according to voltage sorting sequences to form voltage clean frames are as follows: When the group pressure residual exceeds the limit but the plate pressure residual does not exceed the limit, locate the group pressure measurement point channel; when the plate pressure residual exceeds the limit but the section pressure residual does not exceed the limit, locate the plate pressure measurement point channel; when the section pressure residual exceeds the limit, locate the section pressure measurement point channel, thus forming a pollution channel positioning mark. The abnormal voltage is reconstructed based on the pollution channel location marker. The abnormal group voltage value is replaced by the sum of the corresponding plate voltage values, the abnormal plate voltage value is replaced by the sum of the corresponding section voltage values, and the abnormal section voltage value is replaced by the average of the adjacent non-limited section voltage values ​​in the section voltage sorting sequence, thus forming a voltage clean frame.

[0012] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity assessment of batteries described in this invention, the steps of calculating the time window integral capacity based on the group current in the voltage clean frame, calculating the first ampere-hour residual based on the time window integral capacity and the set capacity, and calculating the second ampere-hour residual based on the time window integral capacity and the uploaded capacity are as follows: Select effective unified time windows that are identified as nuclear capacity discharge stages from the voltage clean frame, arrange the group currents in the effective unified time windows according to the unified time window number, and simultaneously retain the time window interval between adjacent effective unified time windows to form a group current integration sequence; Along the current integration sequence, two adjacent effective unified time windows are selected sequentially. The current endpoints corresponding to the two adjacent effective unified time windows form a trapezoidal current boundary, and the time window interval between adjacent effective unified time windows forms a trapezoidal time boundary, thus forming a single window capacity segment. The single-window capacity segments are continuously aggregated along the time window sequence of the nuclear discharge stage to form the time window integrated capacity; The starting point of the nuclear capacity discharge stage is taken as the starting point of the capacity progress. The discharge process is determined according to the relative nuclear capacity time of the current effective unified time window. The set capacity is mapped to the set capacity progress value under the same effective unified time window. The deviation of the time window integral capacity from the set capacity progress value is used to form the first ampere-hour residual. Using the unified time window number as an index, the transmission capacity corresponding to the same valid unified time window is extracted from the voltage clean frame, and the time window integral capacity is aligned with the transmission capacity. Then, the deviation of the time window integral capacity from the transmission capacity is used to form the second ampere-hour residual.

[0013] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity assessment of batteries described in this invention, the specific steps for identifying the cumulative increase pattern of the first ampere-hour residual and the second ampere-hour residual, and for back-calculating the group current drift value when the first ampere-hour residual and the second ampere-hour residual continuously deviate in the same direction, are as follows: The first and second ampere-hour residuals are arranged according to the relative core capacity time. The cumulative slope of the first and second ampere-hour residuals is extracted by Theil-Sen median slope estimation to form a residual cumulative slope pattern identifier. When the cumulative residual pattern indicates that the first ampere-hour residual and the second ampere-hour residual deviate continuously in the same direction, the cumulative slope of the first ampere-hour residual and the cumulative slope of the second ampere-hour residual are fused together, and the group current drift value is formed according to the conversion relationship between the ampere-hour change rate and the current drift.

[0014] As a preferred embodiment of the multi-parameter data cleaning method for remote capacity assessment of batteries described in this invention, the steps of reversely correcting the group current based on the group current drift value to form a current clean frame, calculating the capacity based on the corrected group current in the current clean frame, marking the cleaning category, and outputting a cleaning capacity assessment message are as follows: Using the group flow drift value as the group flow compensation benchmark, the group flow in the group flow integral sequence is reversed to form the corrected group flow; The corrected group flow is backfilled into the voltage clean frame to form the current clean frame, and the core capacity is regenerated based on the current clean frame. The cleaning category is marked according to the contamination channel location identifier and the group flow drift value, and the cleaning core capacity message is output.

[0015] Secondly, the present invention provides a multi-parameter data cleaning system for remote capacity verification of batteries, including: a capacity verification frame verification module, used to collect capacity verification frames in the remote capacity verification process, perform format verification and range verification on the capacity verification frames, and form a standard capacity verification frame; The time axis calibration module is used to map standard nuclear capacity frames to a unified time axis based on the nuclear capacity start time. It determines the communication reliability value based on the continuous status of the frame sequence number of the standard nuclear capacity frame, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the reception time and the acquisition timestamp. The standard nuclear capacity frames in the same time window are sorted according to the communication reliability value, and the standard nuclear capacity frame with the highest ranking is retained. The module also performs nuclear capacity stage calibration on the standard nuclear capacity frames according to the nuclear capacity status value to form aligned nuclear capacity frames. The voltage cleaning module is used to calculate the group voltage residual based on the battery pack measurement point topology corresponding to the aligned core capacity frame, calculate the plate voltage residual when the group voltage residual exceeds the limit, calculate the section voltage residual when the plate voltage residual exceeds the limit, locate the contamination channel according to the residual level in sequence, and reconstruct the abnormal voltage according to the section voltage sorting sequence to form a voltage clean frame. The capacity cleaning module is used to calculate the time window integrated capacity based on the group current in the voltage clean frame, calculate the first ampere-hour residual based on the time window integrated capacity and the set capacity, calculate the second ampere-hour residual based on the time window integrated capacity and the transmitted capacity, identify the cumulative increase pattern of the first and second ampere-hour residuals, back-calculate the group current drift value when the first and second ampere-hour residuals deviate continuously in the same direction, and correct the group current in reverse based on the group current drift value to form a current clean frame. The core capacity is calculated based on the corrected group current in the current clean frame, the cleaning category is marked, and a cleaning core capacity message is output.

[0016] The beneficial effects of this invention are as follows: by back-calculating the group current drift value through the cumulative pattern of the first and second ampere-hour residuals, and then using the group current drift value to reverse-correct the group current integral sequence, the source of capacity deviation is realized from result deviation to the cause of current drift. This is used to eliminate the continuous cumulative effect of the slow shift of the group current during the discharge stage on the time window integral capacity and core capacity, and improves the stability of remote core capacity calculation and the reliability of cleaning core capacity messages. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a multi-parameter data cleaning method for remote capacity assessment of batteries.

[0019] Figure 2 This is a schematic diagram of a multi-parameter data cleaning system for remote capacity assessment of batteries.

[0020] Figure 3 This is a flowchart of the time axis calibration and the calibration of the core capacity frame.

[0021] Figure 4 This is a flowchart for capacity residual accumulation identification and group flow drift value compensation.

[0022] Figure 5 This is a comparison chart showing the cumulative number of misselected frames and a magnified view of a specific area within the time window.

[0023] Figure 6 This is a comparison chart showing the cumulative capacity deviation as a function of relative core capacity over time. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 This is one embodiment of the present invention, which provides a multi-parameter data cleaning method for remote capacity assessment of batteries, including the following steps: S1. Collect the verification frames during the remote verification process, perform format verification and range verification on the verification frames, and form a standard verification frame.

[0028] It should be noted that the data collection target is the battery pack undergoing remote capacity verification. The data collection includes the group voltage value from the group voltage test point, the board voltage value from the board voltage test point, the galvanic voltage value from the junction voltage test point, and the group current from the group current acquisition circuit. The data collection also includes the set capacity, the transmitted capacity, the verification status value, and the verification execution identifier from the remote capacity verification execution terminal. At the same time, the data collection includes the frame sequence number, the data collection timestamp, the reception time, the frame source identifier, and the verification field from the remote capacity verification communication link. The above data collection is arranged and encapsulated according to the remote capacity verification transmission format to form the capacity verification frame in the remote capacity verification process.

[0029] Group current refers to the discharge current flowing through the main circuit of the entire battery pack during remote core capacity discharge, and is used to calculate the core capacity by time window integration.

[0030] S1.1. Perform position matching on the frame structure fields of the core capacity frame to determine that the core capacity frame meets the remote core capacity transmission format. Then, perform consistency verification on the check field and the payload to confirm that the core capacity frame has not been damaged during remote transmission. Subsequently, compare the core capacity parameters with the corresponding core capacity allowable range, mark and remove core capacity frames with core capacity frame cleaning abnormalities, and form a standard core capacity frame.

[0031] Furthermore, the frame structure fields in the nuclear capacity frame are read according to the field arrangement order specified in the remote nuclear capacity transmission format. The field position, field length, and field boundary of each frame structure field are matched with the content specified in the remote nuclear capacity transmission format item by item. Nuclear capacity frames that do not match the field position, field length, or field boundary are marked as nuclear capacity frame cleaning abnormalities. After the frame structure field completes the position matching, the payload is read, the payload check value is calculated according to the existing check method corresponding to the check field, and the payload check value is checked against the check field for consistency. The core capacity frame that fails the consistency check is marked as a core capacity frame cleaning abnormality. Read the kernel capacity parameter according to the kernel capacity parameter name, retrieve the kernel capacity allowable range corresponding to the kernel capacity parameter, compare the kernel capacity parameter with the corresponding kernel capacity allowable range item by item, and mark the kernel capacity frame that exceeds the corresponding kernel capacity allowable range as kernel capacity frame cleaning abnormality; remove the kernel capacity frames marked as kernel capacity frame cleaning abnormality, and retain the kernel capacity frames that are not marked as kernel capacity frame cleaning abnormality to form standard kernel capacity frames.

[0032] It should be noted that the remote capacity transfer format is derived from the device communication protocol used by the remote capacity execution end and the remote capacity receiving end, and is obtained by reading the provisions on field arrangement order, field length and field boundary in the device communication protocol.

[0033] The payload originates from the nuclear capacity frame. After removing the frame structure control field and check field according to the remote nuclear capacity transmission format, the content of the field carrying the nuclear capacity parameters is read.

[0034] The corresponding allowable capacity range is derived from the rated parameters of the battery pack, the parameters of the remote capacity assessment task, and the rated range of the data acquisition equipment. It is obtained by matching the corresponding allowable range according to the capacity parameter name.

[0035] S2. Based on the core capacity start time, the standard core capacity frame is mapped to a unified time axis. The communication reliability value is determined according to the continuous status of the frame sequence number of the standard core capacity frame, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the reception time and the acquisition timestamp. The standard core capacity frames in the same time window are sorted according to the communication reliability value. The standard core capacity frame with the highest ranking is retained. The core capacity stage is calibrated according to the core capacity status value to form an aligned core capacity frame.

[0036] S2.1 Identify the first frame of the core capacity execution from the standard core capacity frame, determine the acquisition timestamp of the first frame of the core capacity execution as the core capacity start time, and form the core capacity time axis reference.

[0037] Furthermore, standard nuclear capacity frames are arranged from earliest to latest according to the collection timestamp. The nuclear capacity execution identifier in the standard nuclear capacity frame is read frame by frame, and standard nuclear capacity frames whose nuclear capacity execution identifier does not indicate the start of remote nuclear capacity are filtered out. Within the retained standard nuclear capacity frames, the nuclear capacity status value is read again. The standard nuclear capacity frame whose nuclear capacity status value first enters the nuclear capacity preparation state or constant current discharge state is selected. The consecutive status of the frame sequence number is combined to check that there is no disorder in the adjacent standard nuclear capacity frames. The standard nuclear capacity frame that first meets the conditions of valid nuclear capacity execution identifier, valid nuclear capacity status value and consecutive frame sequence number is determined as the first frame of nuclear capacity execution. The collection timestamp of the first frame of nuclear capacity execution is read and determined as the nuclear capacity start time. The nuclear capacity start time is used as the nuclear capacity time axis reference.

[0038] S2.2 Set the core capacity time axis reference to a unified time axis zero point, convert the collection timestamp of the standard core capacity frame into a relative core capacity time, and classify it into the corresponding unified time window according to the remote core capacity collection cycle to form a time windowed standard core capacity frame.

[0039] Furthermore, using the core capacity time axis reference as the unified time axis zero point, the acquisition timestamps of standard core capacity frames are read frame by frame, and the time interval between the acquisition timestamp and the core capacity time axis reference is calculated to generate the relative core capacity time. The acquisition cycle sequence to which the relative core capacity time belongs is determined according to the remote core capacity acquisition cycle, and the unified time window number is determined according to the acquisition cycle sequence number. The standard core capacity frames are assigned to the unified time window corresponding to the unified time window number. Standard core capacity frames whose acquisition timestamps are earlier than the core capacity time axis reference or cannot be assigned to the unified time window are removed. The standard core capacity frames that have completed the relative core capacity time marking and unified time window assignment are used as time windowed standard core capacity frames.

[0040] It should be noted that the remote capacity collection cycle is determined by reading the collection interval configuration in the remote capacity task parameters, checking the collection timestamp interval of adjacent standard capacity frames, eliminating abnormal intervals caused by obvious transmission delays, and selecting the time interval that is consistent with the collection interval configuration and appears stably in consecutive standard capacity frames.

[0041] S2.3. Determine the frame sequence trust item based on the continuous state of the frame sequence number of the time-windowed standard core frame, determine the timestamp trust item based on the offset between the relative core time and the unified time window center time, determine the link trust item based on the difference between the receiving time and the acquisition timestamp, and combine them according to a fixed priority to form a communication trust value.

[0042] Furthermore, using the unified time window number as the processing order, each windowed standard core frame within the same unified time window is continuously checked against the windowed standard core frame with the closest frame number in the previous unified time window. When the frame number conforms to the increasing relationship and does not repeat with other frame numbers in the same unified time window, the frame sequence confidence item is marked as normal and continuous. When the frame number has a skip number but does not fall back, the frame sequence confidence item is marked as boundary skip. When the frame number is repeated or falls back, the frame sequence confidence item is marked as repeated and fallen back. The unified time window center time is determined according to the remote capacity acquisition cycle. The absolute time offset between the relative capacity time and the unified time window center time is calculated. When the absolute time offset is within the range of the unified time window center, the timestamp trust item is marked as center proximity. When the absolute time offset is close to the boundary of the unified time window, the timestamp trust item is marked as boundary offset. The transmission interval between the reception time and the acquisition timestamp is calculated. The transmission interval is compared with the normal arrival interval of the remote capacity communication link. When the transmission interval is within the normal arrival interval, the link trust item is marked as normal arrival. When the transmission interval exceeds the normal arrival interval, the link trust item is marked as delayed arrival. The communication trust value is generated by arranging the frame order trust item, timestamp trust item and link trust item in order of priority.

[0043] The absolute time offset between the relative core time and the center time of the unified time window is calculated using the following expression: ; in: It is the first The absolute time offset corresponding to the standard core capacity frame of the frame-time windowing, in seconds; It is the first The relative capacity time corresponding to the standard capacity frame of the frame-time windowing, in seconds; It is the first The unified window number to which the frame windowing standard core frame belongs, and the unified window number is counted starting from 0; This is the remote capacity data acquisition cycle, in seconds; It is the frame sequence index of the time-windowed standard core frame.

[0044] It should be noted that the normal arrival interval is determined by the transmission interval between the reception time and the acquisition timestamp of consecutive standard nuclear capacity frames in the same remote nuclear capacity communication link. After removing the transmission intervals corresponding to obvious out-of-order and excessively long delays, the range of stable transmission intervals is taken as the normal arrival interval.

[0045] The range near the center of the unified time window is determined by the remote core capacity acquisition cycle, the start and end times of the unified time window are determined, the midpoint of the start and end times is taken as the center time of the unified time window, and the stable offset amplitude of continuously arriving core capacity frames relative to the center time is statistically calculated. The stable offset coverage interval on both sides of the center time is then defined. The boundary of the unified time window is determined by the remote core capacity acquisition cycle to determine the start and end points of the unified time window. After deducting the range near the center of the unified time window, the side time intervals near the start and end points are retained. The state of the core capacity frame falling into the side time interval relative to the core capacity time is obtained after the state is determined.

[0046] S2.4 Sort the time-windowed standard core frame according to the communication trust value, retain the time-windowed standard core frame with the highest ranking as the time window representative frame, and determine the corresponding unified time window as the effective unified time window.

[0047] Furthermore, windowed standard capacity frames belonging to the same unified time window are aggregated according to the unified time window number. Within the same unified time window, these frames are arranged from highest to lowest based on their communication trust value. When comparing communication trust values, the frame sequence trust item is used to prioritize distinguishing the continuity of frame sequence numbers. For windowed standard capacity frames with the same frame sequence trust item, the timestamp trust item is used to further distinguish the degree to which their relative capacity time is close to the center time of the unified time window. For windowed standard capacity frames with the same timestamp trust item, the link trust item is used to further distinguish the degree to which their transmission interval is close to the normal arrival interval. After sorting, the windowed standard capacity frame with the highest ranking is retained as the representative frame of the time window, and the unified time window containing the representative frame is determined as the valid unified time window.

[0048] S2.5. The core capacity status value in the time window representative frame is analyzed, the core capacity preparation state is mapped to the core capacity preparation stage, the constant current discharge state is mapped to the core capacity discharge stage, the discharge stop state is mapped to the core capacity cutoff stage, the capacity calculation state is mapped to the capacity calculation stage, and the core capacity end state is mapped to the core capacity end stage, thus forming the candidate core capacity stage.

[0049] Furthermore, the window representative frames corresponding to the effective unified window are processed one by one in order of effective unified window number; the core capacity status value is located in the core capacity status frame, and the core capacity status value is compared with the core capacity preparation status. When the core capacity status value is consistent with the core capacity preparation status, the stage corresponding to the effective unified window number is recorded as the core capacity preparation stage. If the nuclear capacity status value is inconsistent with the nuclear capacity preparation status, the nuclear capacity status value is further compared with the constant current discharge status. If the nuclear capacity status value is consistent with the constant current discharge status, the stage corresponding to the effective unified time window number is recorded as the nuclear capacity discharge stage. If the nuclear capacity status value is inconsistent with the constant current discharge status, the nuclear capacity status value is further compared with the discharge stop status. If the nuclear capacity status value is consistent with the discharge stop status, the stage corresponding to the effective unified time window number is recorded as the nuclear capacity cutoff stage. When the capacity status value is inconsistent with the discharge stop status, the capacity status value is compared with the capacity calculation status. When the capacity status value is consistent with the capacity calculation status, the stage corresponding to the effective unified time window number is recorded as the capacity calculation stage. When the capacity status value is inconsistent with the capacity calculation status, the capacity status value is compared with the capacity end status. When the capacity status value is consistent with the capacity end status, the stage corresponding to the effective unified time window number is recorded as the capacity end stage. After completing all valid unified time windows and corresponding stage records, arrange the stage records according to the valid unified time window number to form candidate capacity approval stages.

[0050] S2.6. Compare the candidate capacity phase with the confirmed capacity phase of the previous effective unified time window. If the candidate capacity phase has not changed, generate a capacity phase identifier based on the candidate capacity phase. If the candidate capacity phase has changed, perform continuity confirmation through the candidate capacity phase of the next effective unified time window, and generate a capacity phase identifier based on the continuity confirmation result.

[0051] Furthermore, the phase confirmation is carried out in ascending order of the effective unified time window number. If the first effective unified time window does not have a confirmation and approval phase of the previous effective unified time window, the candidate approval phase of the first effective unified time window is taken as the confirmation approval phase, and approval phase identifiers are generated according to the confirmation approval phases. Starting from the second effective unified time window, the candidate capacity approval stage of the current effective unified time window is compared with the confirmed capacity approval stage of the previous effective unified time window. If the candidate capacity approval stage is consistent with the confirmed capacity approval stage of the previous effective unified time window, the candidate capacity approval stage of the current effective unified time window is taken as the confirmed capacity approval stage, and a capacity approval stage identifier is generated according to the confirmed capacity approval stage. When the candidate capacity approval stage is inconsistent with the confirmation capacity approval stage of the previous valid unified time window, the candidate capacity approval stage of the current valid unified time window and the candidate capacity approval stage of the next valid unified time window are continuously confirmed. When the candidate capacity approval stage of the current valid unified time window and the candidate capacity approval stage of the next valid unified time window are consistent, the candidate capacity approval stage of the current valid unified time window is used as the confirmation capacity approval stage, and a capacity approval stage identifier is generated according to the confirmation capacity approval stage. When the candidate capacity approval stage of the current effective unified time window is inconsistent with the candidate capacity approval stage of the next effective unified time window, the confirmation capacity approval stage of the previous effective unified time window will be extended to the confirmation capacity approval stage of the current effective unified time window, and a capacity approval stage identifier will be generated according to the extended confirmation capacity approval stage.

[0052] S2.7. Bind the capacity stage identifier, the valid unified time window number where the time window representative frame is located, and the relative capacity time to the time window representative frame, and retain the capacity parameters in the time window representative frame to form an aligned capacity frame.

[0053] Furthermore, taking the valid unified time window corresponding to the time window representative frame as the processing object, the capacity stage identifier generated after continuity confirmation is appended to the time window representative frame, the unified time window number to which the time window representative frame belongs is appended to the time window representative frame, the relative capacity time corresponding to the time window representative frame is appended to the time window representative frame, and the communication trust value used when the time window representative frame is sorted within the same unified time window is appended to the time window representative frame. The original capacity parameters and the bound time window alignment content in the time window representative frame are retained to form an aligned capacity frame.

[0054] like Figure 5 As shown, the upper overview diagram uses the unified time window number as the horizontal axis and the cumulative number of misselected time windows as the vertical axis, comparing the earliest received frame retention method, the latest received frame retention method, the timestamp center proximity method, the frame sequence number consecutive priority method, and the processing method of this invention under remote communication disturbance conditions to show the cumulative misselection of representative frames for time windows. The red dashed rectangle shows the local magnified area where the stage switching interference is concentrated, and is connected to the lower local magnified diagram by a red leader. The lower local magnified diagram magnifies the unified time window interval corresponding to the red dashed rectangle and marks the jump points and the points of greatest difference for each processing method during the stage switching interference period. It can be seen that the latest received frame retention method and the earliest received frame retention method show obvious misselection jumps near the stage switching, the timestamp center proximity method and the frame sequence number consecutive priority method still have misselection accumulation, while the processing method of this invention has the smallest increase in the number of misselections in the same interval. This indicates that this invention, by forming a communication credibility value through frame sequence credibility item, timestamp credibility item and link credibility item and then sorting it, can more accurately retain the representative frame of the time window and provide a reliable basis for the subsequent generation of capacity stage identifiers.

[0055] The earliest received frame retention method refers to directly selecting the earliest received standard capacity frame as the representative frame of the time window when multiple standard capacity frames exist within the same unified time window. When out-of-order transmission, retransmission, or phase switching jitter occurs in the remote communication link, the earliest received standard capacity frame may not be closest to the center time of the unified time window, and may also carry the capacity status value of the previous capacity phase. Therefore, it is easy to cause the effective unified time window to be missed during the discharge phase, which ultimately manifests as an increase in the cumulative number of incorrectly selected time windows and a negative cumulative capacity error.

[0056] The latest received frame retention method refers to directly selecting the latest received standard capacity frame as the representative frame of the time window when multiple standard capacity frames exist within the same unified time window. When there are delayed received frames or duplicate uploaded frames in the remote capacity link, the latest received standard capacity frame is prone to becoming a delayed frame or a duplicate frame. This can lead to the capacity cutoff stage being misjudged as the capacity discharge stage, thus incorporating the effective unified time window of the non-discharge stage into the group flow integration sequence. Ultimately, this results in the highest cumulative number of misselected time windows and a positive cumulative capacity error.

[0057] The timestamp center proximity method selects the standard capacity frame that is closest to the center time of the unified capacity window within the same unified time window as the representative frame of the time window. The timestamp center proximity method can reduce the impact of some acquisition timestamp offsets, but it only focuses on the time position and does not consider the frame sequence number continuity and link arrival status. When duplicate frames carry acquisition timestamps that are close to the center time of the unified time window, they may still be mistakenly selected. Therefore, the cumulative number of misselected time windows and the cumulative capacity error continue to increase.

[0058] The frame sequence number continuation priority method prioritizes standard capacity frames whose frame sequence numbers are continuous with adjacent unified time windows within the same unified time window. While this method reduces misselection caused by duplicate or skipped frames, it lacks sufficient constraints on reception delay and acquisition timestamp offset. Furthermore, when frame sequence numbers are continuous but the reception time is significantly delayed or deviates from the center of the unified time window relative to the capacity time, it still affects the generation of capacity stage identifiers and subsequent capacity integration.

[0059] S3. Calculate the group voltage residual based on the battery pack measurement point topology corresponding to the aligned core capacity frame. Calculate the plate voltage residual when the group voltage residual exceeds the limit. Calculate the section voltage residual when the plate voltage residual exceeds the limit. Locate the contamination channel according to the residual level in sequence, and reconstruct the abnormal voltage according to the section voltage sorting sequence to form a voltage clean frame.

[0060] S3.1 Using the valid unified time window number in the aligned capacity frame as an index, extract the group voltage value, plate voltage value and section voltage value from the aligned capacity frame corresponding to the same valid unified time window, and retrieve the battery pack measurement point topology according to the battery pack identifier corresponding to the aligned capacity frame.

[0061] Furthermore, taking the aligned capacity frames as the processing object, the aligned capacity frames are collected according to the valid unified time window number. Under the same valid unified time window number, the battery pack identifier, pack voltage test point number and pack voltage value, plate voltage test point number and plate voltage value, and junction voltage test point number and junction voltage value are extracted. The corresponding battery pack test point topology is retrieved according to the battery pack identifier. The battery pack test point topology includes the attribution relationship between the pack voltage test point and the plate voltage test point, the attribution relationship between the plate voltage test point and the junction voltage test point, and the series sequence of the junction voltage test point. According to the battery pack test point topology, the pack voltage value, plate voltage value, and junction voltage value under the same valid unified time window are mapped to the corresponding test point positions to form topological voltage data.

[0062] S3.2. Determine the plate voltage test points belonging to the same group of voltage test points according to the battery pack test point topology. Within the same effective unified time window, perform difference calculation between the group voltage value and the sum of plate voltage values ​​belonging to the same group of voltage test points to form the group voltage residual. When the group voltage residual exceeds the limit, determine the plate voltage test point corresponding to the over-limit group voltage test point according to the voltage test point attribution table.

[0063] Furthermore, the group voltage test point numbers in the topology voltage data are processed one by one. After selecting a group voltage test point within the same valid unified time window, all plate voltage test points belonging to the group voltage test point are searched according to the battery pack test point topology. The group voltage value of the group voltage test point and the plate voltage values ​​of all corresponding plate voltage test points are extracted from the topology voltage data. When all corresponding plate pressure test points have valid plate pressure values, the corresponding plate pressure values ​​are summed to form a plate pressure value. Then, the group pressure value is calculated by subtracting the sum of plate pressure values ​​from the group pressure value to form the group pressure residual under the current valid unified time window. When the group pressure residual exceeds the group pressure residual limit condition, the group pressure test point is marked as the limit-exceeding group pressure test point. According to the battery pack test point topology, the set of plate pressure test points corresponding to the limit-exceeding group pressure test point is output for subsequent plate pressure residual calculation.

[0064] It should be noted that the group voltage residual exceeding the limit condition is determined by confirming the number of plate voltage measurement points under the same group voltage measurement point according to the voltage measurement point assignment table, then determining the allowable deviation of the sum of plate voltage values ​​based on the plate voltage value acquisition error, and finally combining the group voltage value acquisition error.

[0065] S3.3. Perform a difference calculation on the sum of the plate voltage value and the junction voltage value belonging to the same plate voltage measurement point to form the plate voltage residual. When the plate voltage residual exceeds the limit, determine the junction voltage measurement point corresponding to the plate voltage measurement point that exceeds the limit according to the voltage measurement point assignment table.

[0066] Furthermore, taking the plate voltage test points corresponding to the over-limit group voltage test points as the processing range, the affiliation relationship of the section voltage test points corresponding to each plate voltage test point is determined according to the voltage test point affiliation table; within the same effective unified time window, the section voltage values ​​belonging to the same plate voltage test point are collected and the sum of section voltage values ​​is calculated, and the difference between the plate voltage value corresponding to the plate voltage test point and the sum of section voltage values ​​is calculated to form the plate voltage residual; The plate voltage residual is compared with the plate voltage residual exceeding the limit. When the plate voltage residual does not exceed the limit, the voltage state of the plate voltage measurement point is retained. When the plate voltage residual exceeds the limit, the plate voltage measurement point is marked as the plate voltage measurement point exceeding the limit. The corresponding section voltage measurement point is determined according to the voltage measurement point assignment table.

[0067] It should be noted that the plate voltage residual exceeding the limit condition is determined by confirming the number of section voltage measurement points under the same plate voltage measurement point according to the voltage measurement point assignment table, then forming the allowable deviation of the section voltage value summation based on the section voltage value acquisition error, and then combining it with the plate voltage value acquisition error.

[0068] S3.4. Form a pressure sorting sequence according to the series connection order of the pressure measurement points, determine the candidate abnormal pressure measurement points through Hampel type robust window verification, and perform difference calculation on the current pressure value and the pressure reference value corresponding to the candidate abnormal pressure measurement points to form the pressure residual.

[0069] Furthermore, taking the voltage measurement points corresponding to the over-limit plate voltage measurement points as the processing range, the current voltage values ​​corresponding to each voltage measurement point are arranged in the series order of the voltage measurement points in the voltage measurement point attribution table to form a voltage measurement sorting sequence. The voltage measurement points to be verified are selected one by one along the voltage measurement sorting sequence. The current voltage values ​​of the voltage measurement points adjacent to the voltage measurement points on both sides of the voltage measurement point to be verified form a Hampel-type robust window, which is used to select neighboring voltage measurement points around the voltage measurement point to be verified in the voltage measurement sorting sequence. The current voltage values ​​of the neighboring voltage measurement points form a voltage reference value and a robust discrete limit, thereby determining whether the current voltage value of the voltage measurement point to be verified deviates significantly from the neighboring voltage level, and thus determining the candidate abnormal voltage measurement points.

[0070] The median value of the current throttling pressure within the Hampel-type robust window is used as the throttling pressure reference value. The robust dispersion limit is then determined based on the degree of dispersion of the current throttling pressure value relative to the throttling pressure reference value within the Hampel-type robust window. The deviation between the current throttling pressure value of the throttling pressure measurement point to be verified and the throttling pressure reference value is checked. When the deviation exceeds the robust dispersion limit, the throttling pressure measurement point to be verified is identified as a candidate abnormal throttling pressure measurement point. The difference between the current throttling pressure value corresponding to the candidate abnormal throttling pressure measurement point and the throttling pressure reference value is calculated to form the throttling pressure residual.

[0071] S3.5 When the group pressure residual exceeds the limit but the plate pressure residual does not exceed the limit, locate the group pressure measurement point channel; when the plate pressure residual exceeds the limit but the section pressure residual does not exceed the limit, locate the plate pressure measurement point channel; when the section pressure residual exceeds the limit, locate the section pressure measurement point channel, thus forming a pollution channel positioning mark.

[0072] Furthermore, taking the aligned core frame corresponding to the effective unified time window as the order, the group pressure residual is compared with the group pressure residual over-limit condition, the board pressure residual is compared with the board pressure residual over-limit condition, and the section pressure residual is compared with the section pressure residual over-limit condition; when the group pressure residual exceeds the limit but the board pressure residual corresponding to the over-limit group pressure test point does not exceed the limit, the source of the anomaly is limited to the group pressure test point channel of the over-limit group pressure test point, and the group pressure test point channel is marked under the over-limit group pressure test point; When the plate pressure residual exceeds the limit but the section pressure residual corresponding to the plate pressure test point that exceeds the limit does not exceed the limit, the source of the anomaly is limited to the plate pressure test point channel of the plate pressure test point that exceeds the limit, and the plate pressure test point channel is marked under the plate pressure test point that exceeds the limit; when the section pressure residual exceeds the limit, the source of the anomaly is limited to the section pressure test point channel of the candidate abnormal section pressure test point, and the section pressure test point channel is marked under the candidate abnormal section pressure test point; the effective unified time window number, the pressure test point of the exceeding limit group, the plate pressure test point that exceeds the limit, the candidate abnormal section pressure test point and the corresponding channel mark are recorded accordingly to form a pollution channel location identifier.

[0073] S3.6. Reconstruct abnormal voltages based on the pollution channel location markers. Replace abnormal group voltages with the sum of corresponding plate voltages, replace abnormal plate voltages with the sum of corresponding section voltages, and replace abnormal section voltages with the average of adjacent non-limited section voltages in the section voltage sorting sequence to form a voltage clean frame.

[0074] Furthermore, following the order of effectively unifying the time window, for the over-limit group pressure test points marked as group pressure test point channels in the pollution channel positioning identification, the plate pressure values ​​belonging to the over-limit group pressure test points are collected according to the voltage test point attribution table, and the sum of the corresponding plate pressure values ​​is calculated. The abnormal group pressure value corresponding to the over-limit group pressure test point is replaced with the sum of the corresponding plate pressure values. For the plate pressure test points that exceed the limit in the pollution channel location identification, collect the voltage values ​​belonging to the plate pressure test points according to the voltage test point classification table, calculate the sum of the corresponding voltage values, and replace the abnormal plate pressure value corresponding to the plate pressure test point with the sum of the corresponding voltage values. For candidate abnormal pressure measurement points marked as pressure measurement point channels in the pollution channel location identification, search for adjacent non-limited pressure measurement points on both sides of the candidate abnormal pressure measurement point along the pressure sorting sequence, calculate the average value of adjacent non-limited pressure values, and replace the abnormal pressure value corresponding to the candidate abnormal pressure measurement point with the average value of adjacent non-limited pressure values; after completing the replacement of abnormal group pressure value, abnormal plate pressure value and abnormal pressure value, retain the voltage value not marked by the pollution channel location identification to form a voltage clean frame.

[0075] S4. Calculate the time window integration capacity based on the group current in the voltage clean frame. Calculate the first ampere-hour residual based on the time window integration capacity and the set capacity. Calculate the second ampere-hour residual based on the time window integration capacity and the transmitted capacity. Perform cumulative pattern recognition on the first and second ampere-hour residuals. When the first and second ampere-hour residuals continuously deviate in the same direction, back-calculate the group current drift value. Correct the group current in reverse based on the group current drift value to form a current clean frame. Calculate the core capacity based on the corrected group current in the current clean frame, mark the cleaning category, and output the cleaning core capacity message.

[0076] S4.1 Select the effective unified time window marked as the nuclear capacity discharge stage from the voltage clean frame, arrange the group currents in the effective unified time window according to the unified time window number, and simultaneously retain the time window interval between adjacent effective unified time windows to form the group current integration sequence.

[0077] Going further, S4.2. Select two adjacent effective unified time windows sequentially along the group current integration sequence. Use the group current endpoints corresponding to the two adjacent effective unified time windows to form a trapezoidal current boundary. Use the time window interval between adjacent effective unified time windows to form a trapezoidal time boundary, thus forming a single window capacity segment.

[0078] Furthermore, the capacity stage identifiers are checked in ascending order of the valid unified time window numbers in the voltage clean frame. Valid unified time windows with the capacity stage identifier being the capacity discharge stage are retained, while valid unified time windows corresponding to the capacity preparation stage, capacity cutoff stage, capacity calculation stage, and capacity end stage are removed. The retained valid unified time windows are arranged consecutively according to their unified time window numbers. Within each valid unified time window, the current group is determined, and the current group is recorded in pairs with the corresponding relative capacity time. The time window interval between two adjacent effective unified time windows is calculated based on the relative core capacity time of the latter effective unified time window and the relative core capacity time of the former effective unified time window. The time window interval is then retained synchronously with the group flow corresponding to the two adjacent effective unified time windows to form the group flow integral sequence.

[0079] S4.3. The single-window capacity segments are continuously aggregated along the time window sequence of the nuclear discharge stage to form the time window integrated capacity.

[0080] Furthermore, taking the first effective unified time window of the nuclear capacity discharge stage as the starting point for capacity aggregation, the single-window capacity segments are arranged in ascending order according to the unified time window number, and the capacity is continuously accumulated along the time window sequence; each time the accumulation of a single-window capacity segment is completed, it is bound to the corresponding unified time window number to form the time window integral capacity.

[0081] S4.4. Taking the starting point of the nuclear capacity discharge stage as the starting point of the capacity progress, the discharge process is determined according to the relative nuclear capacity time of the current effective unified time window, and the set capacity is mapped to the set capacity progress value under the same effective unified time window. Then, the deviation of the time window integral capacity from the set capacity progress value is used to form the first ampere-hour residual.

[0082] Furthermore, the effective unified time window for the first time that the nuclear capacity stage identifier is the nuclear capacity discharge stage is taken as the starting point of the nuclear capacity discharge stage, and the effective unified time window for the first time that the nuclear capacity stage identifier changes from the nuclear capacity discharge stage to the nuclear capacity cutoff stage is taken as the ending point of the nuclear capacity discharge stage; the relative nuclear capacity time corresponding to the starting point of the nuclear capacity discharge stage is determined as the starting point of the capacity progress, and the relative nuclear capacity time corresponding to the ending point of the nuclear capacity discharge stage is determined as the ending point of the capacity progress. For any current effective unified time window within a capacity discharge phase, calculate the duration of discharge relative to the capacity progress start point for the current effective unified time window relative to the capacity time window's relative capacity time window. Determine the discharge progress based on the total duration of the capacity discharge phase relative to the capacity progress start point. Proportionally scale the set capacity according to the discharge progress to obtain the set capacity progress value for the current effective unified time window, expressed as: ; in: It is the first in the nuclear discharge stage The set capacity progress value corresponding to each effective unified time window, in ampere-hours; This is the set capacity in the remote capacity control task, in ampere-hours; It is the first in the nuclear discharge stage The relative core capacity time corresponding to each effective unified time window, in seconds; It is the relative core capacity time corresponding to the starting point of capacity progress, in seconds; It is the relative core capacity time corresponding to the end point of the capacity progress, in seconds; It is the time window index of the effective unified time window within the nuclear discharge phase; In the above expression, This represents the discharge process; after the time dimensions cancel each other out, it becomes a dimensionless proportional value. After calculation with the dimensionless proportional value, It is still measured in ampere-hour units and can represent the set capacity progress value under the current effective unified time window.

[0083] The difference between the time window integral capacity under the current effective unified time window and the set capacity progress value is calculated to form the first ampere-hour residual.

[0084] S4.6. Using the unified time window number as an index, extract the transmission capacity corresponding to the same valid unified time window from the voltage clean frame, and align the time window integral capacity with the transmission capacity. Then, use the deviation of the time window integral capacity from the transmission capacity to form the second ampere-hour residual.

[0085] Furthermore, using the unified time window number within the nuclear capacity discharge stage as the search key, the time window integration capacity is processed sequentially from smallest to largest according to the unified time window number; under each unified time window number, the valid unified time window corresponding to the same unified time window number is located from the voltage clean frame, and the upload capacity in the valid unified time window is obtained; When the same unified time window number corresponds to only one uploading capacity, a one-to-one correspondence is established between the uploading capacity and the time window integration capacity under the same unified time window number; when the same unified time window number corresponds to multiple uploading capacities, the uploading capacity corresponding to the voltage clean frame with the highest communication reliability value is retained; when there is no uploading capacity under the same unified time window number, no second ampere-hour residual is generated; after the time window alignment is completed, the difference between the time window integration capacity under the same unified time window number and the uploading capacity is used as the deviation amount to form the second ampere-hour residual.

[0086] S4.7 Arrange the first ampere-hour residual and the second ampere-hour residual according to the relative core capacity time. Use Theil-Sen type median slope estimation to extract the cumulative slope of the first ampere-hour residual and the cumulative slope of the second ampere-hour residual to form a residual cumulative shape identifier.

[0087] Furthermore, taking the effective unified time window with the first ampere-hour residual within the nuclear capacity discharge stage as the object, the first ampere-hour residuals are arranged from early to late according to the relative nuclear capacity time. Any two first ampere-hour residuals with different relative nuclear capacity times are selected in sequence, and the residual difference between the two first ampere-hour residuals and the time interval between the two relative nuclear capacity times are calculated. The two-point slope of the first ampere-hour residual is generated according to the ratio of the residual difference to the time interval. After collecting all the two-point slopes of the first ampere-hour residuals, the median slope of all the two-point slopes of the first ampere-hour residuals is selected by Theil-Sen type median slope estimation to obtain the cumulative slope of the first ampere-hour residual. Taking the effective unified time window with the second ampere-hour residual within the nuclear capacity discharge stage as the object, the second ampere-hour residuals are arranged from early to late according to the relative nuclear capacity time. Any two second ampere-hour residuals with different relative nuclear capacity times are selected in sequence. The residual difference between the two second ampere-hour residuals and the time interval between the two relative nuclear capacity times are calculated. The two-point slope of the second ampere-hour residual is generated according to the ratio of the residual difference to the time interval. After collecting all the two-point slopes of the second ampere-hour residuals, the median slope of all the two-point slopes of the second ampere-hour residuals is selected by Theil-Sen type median slope estimation to obtain the cumulative slope of the second ampere-hour residual. When the cumulative slopes of the first and second ampere-hour residuals are both positive, the residual accumulation pattern is marked as "large capacity, continuous deviation in the same direction". When the cumulative slopes of the first and second ampere-hour residuals are both negative, the residual accumulation pattern is marked as "small capacity, continuous deviation in the same direction". When the cumulative slopes of the first and second ampere-hour residuals are not in the same direction, the residual accumulation pattern is marked as "non-same direction, continuous deviation".

[0088] It should also be noted that the Theil-Sen median slope estimation is an existing robust statistical estimation method used to extract the overall change slope from discrete observations arranged in time. Specifically, it forms observation pairs by taking any two residual observations with different relative core times, calculates the two-point slopes for each pair, sorts all the two-point slopes, and selects the two-point slopes at the median position as the cumulative slope. The Theil-Sen median slope estimation is insensitive to individual outlier residual observations and can reduce the interference caused by occasional jumps on the cumulative slopes of the first and second ampere-hour residuals.

[0089] S4.8 When the residual cumulative increase pattern indicates that the first ampere-hour residual and the second ampere-hour residual deviate continuously in the same direction, the cumulative increase slope of the first ampere-hour residual and the cumulative increase slope of the second ampere-hour residual are fused together, and the group current drift value is formed according to the conversion relationship between the ampere-hour change rate and the current drift amount.

[0090] Furthermore, when the residual cumulative increase pattern is identified as either a large capacity continuously deviating in the same direction or a small capacity continuously deviating in the same direction, the directional attributes of the cumulative increase slope of the first ampere-hour residual and the cumulative increase slope of the second ampere-hour residual are retained. The cumulative increase slopes of the first ampere-hour residual and the second ampere-hour residual are sorted by numerical value, and the median representative value of the two cumulative increase slopes is taken as the median fusion slope. The median fusion slope characterizes the continuous change in ampere-hour deviation per unit time. Based on the unit conversion relationship from seconds to hours, the median fusion slope is converted from the ampere-hour change rate to the current drift. When the residual accumulation pattern indicates that the capacity is too large and the deviation is continuous in the same direction, the group current drift value is marked as positive drift. When the residual accumulation pattern indicates that the capacity is too small and the deviation is continuous in the same direction, the group current drift value is marked as negative drift, thus forming the group current drift value.

[0091] S4.9 Using the group flow drift value as the group flow compensation benchmark, the group flow in the group flow integral sequence is reversed to form the corrected group flow.

[0092] Furthermore, using the unified time window number in the group current integration sequence as the order, the group current and the direction of the group current drift value corresponding to each valid unified time window are determined one by one. When the group current drift value is positive, negative compensation is performed on the group current of the corresponding valid unified time window to correct the group current towards a lower current direction. When the group current drift value is negative, positive compensation is performed on the group current of the corresponding valid unified time window to correct the group current towards a higher current direction. After completing the reverse compensation of all group currents in the nuclear discharge stage, the compensated group currents are re-bound with the corresponding unified time window number to form the corrected group currents.

[0093] S4.10. Fill the corrected group flow back into the voltage clean frame to form the current clean frame, and regenerate the core capacity based on the current clean frame. Mark the cleaning category according to the contamination channel location identifier and the group flow drift value, and output the cleaning core capacity message.

[0094] Furthermore, based on the unified time window number, the modified group current replaces the group current under the same valid unified time window in the voltage clean frame, and retains the group voltage value, board voltage value, section voltage value, capacity stage identifier, relative capacity time and communication reliability value in the voltage clean frame to form a current clean frame; In the current clean frame, select the effective unified time window marked as the core capacity discharge stage, perform capacity integration calculation on the corrected group current according to the relative core capacity time sequence, and regenerate the core capacity. When the contamination channel location identifier has a group voltage measurement point channel, plate voltage measurement point channel, or section voltage measurement point channel, the cleaning category is marked as voltage channel cleaning. When the group current drift value has positive or negative drift, the cleaning category is marked as current drift cleaning. When both the contamination channel location identifier and the group current drift value have abnormal markings, the cleaning category is marked as voltage and current composite cleaning. The current clean frame, core capacity, and cleaning category are encapsulated, and the cleaning core capacity message is output.

[0095] like Figure 6 As shown, the earliest received frame retention method generates a negative capacity error due to the omission of an effective unified time window for the discharge phase, while the latest received frame retention method generates a positive capacity error due to the erroneous inclusion of an effective unified time window for the non-discharge phase. Although the timestamp center proximity method and the frame sequence number successive priority method can reduce some of the impact of communication disturbances, the cumulative capacity error still gradually increases with the relative core capacity time. The processing method of this invention obtains the representative frame of the time window by sorting the communication reliability value, and combines the core capacity phase identifier to filter the effective unified time window of the core capacity discharge phase, so that the group flow integration sequence participating in the integration is closer to the real core capacity discharge phase. Finally, the cumulative capacity error is always kept within a low error range, which shows that it can improve the stability of the time window integration capacity and core capacity.

[0096] This embodiment also provides a multi-parameter data cleaning system for remote capacity verification of batteries, including: a capacity verification frame verification module, used to collect capacity verification frames during the remote capacity verification process, perform format verification and range verification on the capacity verification frames, and form a standard capacity verification frame; The time axis calibration module is used to map standard nuclear capacity frames to a unified time axis based on the nuclear capacity start time. It determines the communication reliability value based on the continuous status of the frame sequence number of the standard nuclear capacity frame, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the reception time and the acquisition timestamp. The standard nuclear capacity frames in the same time window are sorted according to the communication reliability value, and the standard nuclear capacity frame with the highest ranking is retained. The module also performs nuclear capacity stage calibration on the standard nuclear capacity frames according to the nuclear capacity status value to form aligned nuclear capacity frames. The voltage cleaning module is used to calculate the group voltage residual based on the battery pack measurement point topology corresponding to the aligned core capacity frame, calculate the plate voltage residual when the group voltage residual exceeds the limit, calculate the section voltage residual when the plate voltage residual exceeds the limit, locate the contamination channel according to the residual level in sequence, and reconstruct the abnormal voltage according to the section voltage sorting sequence to form a voltage clean frame. The capacity cleaning module is used to calculate the time window integrated capacity based on the group current in the voltage clean frame, calculate the first ampere-hour residual based on the time window integrated capacity and the set capacity, calculate the second ampere-hour residual based on the time window integrated capacity and the transmitted capacity, identify the cumulative increase pattern of the first and second ampere-hour residuals, back-calculate the group current drift value when the first and second ampere-hour residuals deviate continuously in the same direction, and correct the group current in reverse based on the group current drift value to form a current clean frame. The core capacity is calculated based on the corrected group current in the current clean frame, the cleaning category is marked, and a cleaning core capacity message is output.

[0097] In summary, this invention uses the cumulative increase pattern of the first and second ampere-hour residuals to deduce the group current drift value, and then uses the group current drift value to reverse the group current integral sequence. This enables the source of capacity deviation to be traced back to the cause of current drift, thus eliminating the continuous cumulative impact of the slow shift of the group current during the discharge phase on the time window integral capacity and core capacity. This improves the stability of remote core capacity calculation and the reliability of core capacity cleaning messages.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-parameter data cleaning method for remote capacity assessment of storage batteries, characterized in that, include: Collect verification frames during the remote verification process, perform format and range verification on the verification frames, and form standard verification frames; Based on the start time of the nuclear capacity system, the standard nuclear capacity frames are mapped to a unified time axis. The communication reliability value is determined according to the continuous status of the frame sequence number of the standard nuclear capacity frames, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the reception time and the acquisition timestamp. The standard nuclear capacity frames in the same time window are sorted according to the communication reliability value, and the standard nuclear capacity frame with the highest ranking is retained. The nuclear capacity stage is calibrated according to the nuclear capacity status value to form an aligned nuclear capacity frame. The battery pack voltage residual is calculated based on the topology of the battery pack measurement points corresponding to the aligned core capacity frame. When the battery pack voltage residual exceeds the limit, the plate voltage residual is calculated. When the plate voltage residual exceeds the limit, the section voltage residual is calculated. The contamination channel is located according to the residual level in sequence, and the abnormal voltage is reconstructed according to the section voltage sorting sequence to form a voltage clean frame. The time window integration capacity is calculated based on the group current in the voltage clean frame. The first ampere-hour residual is calculated based on the time window integration capacity and the set capacity. The second ampere-hour residual is calculated based on the time window integration capacity and the transmitted capacity. The cumulative increase pattern of the first ampere-hour residual and the second ampere-hour residual is identified. When the first ampere-hour residual and the second ampere-hour residual deviate continuously in the same direction, the group current drift value is back-calculated. The group current is corrected in reverse based on the group current drift value to form a current clean frame. The core capacity is calculated based on the corrected group current in the current clean frame, the cleaning category is marked, and the cleaning core capacity message is output. The specific steps for performing cumulative pattern recognition on the first and second ampere-hour residuals, and for back-calculating the flow drift value when the first and second ampere-hour residuals continuously deviate in the same direction, are as follows: The first and second ampere-hour residuals are arranged according to the relative core capacity time. The cumulative slope of the first and second ampere-hour residuals is extracted by Theil-Sen median slope estimation to form a residual cumulative slope pattern identifier. When the residual accumulation pattern indicates that the first ampere-hour residual and the second ampere-hour residual deviate continuously in the same direction, the cumulative slope of the first ampere-hour residual and the cumulative slope of the second ampere-hour residual are fused together and the group current drift value is formed according to the conversion relationship between the ampere-hour change rate and the current drift. The process involves reversing the current group based on the current group drift value to form a current clean frame, calculating the core capacity based on the corrected current group in the current clean frame, marking the cleanliness category, and outputting a cleanliness core capacity message. The specific steps are as follows: Using the group flow drift value as the group flow compensation benchmark, the group flow in the group flow integral sequence is reversed to form the corrected group flow; The corrected group flow is backfilled into the voltage clean frame to form the current clean frame, and the core capacity is regenerated based on the current clean frame. The cleaning category is marked according to the contamination channel location identifier and the group flow drift value, and the cleaning core capacity message is output.

2. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 1, characterized in that, The standard core frame is formed by matching the position of the frame structure fields of the core frame to determine that the core frame meets the remote core frame transmission format, and then performing consistency verification between the verification field and the payload to confirm that the core frame has not been damaged during remote transmission. Subsequently, the core frame parameters are compared with the corresponding core frame allowable range, and core frames with core frame cleaning abnormalities are marked and removed.

3. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 2, characterized in that, The specific steps for retaining the highest-ranking standard core capacity frame are as follows: Identify the first frame of the core capacity execution from the standard core capacity frame, determine the acquisition timestamp of the first frame of the core capacity execution as the core capacity start time, and form the core capacity time axis reference; Set the time axis reference of the nuclear capacity to a unified time axis zero point, convert the collection timestamp of the standard nuclear capacity frame into a relative nuclear capacity time, and classify it into the corresponding unified time window according to the remote nuclear capacity collection cycle to form a time windowed standard nuclear capacity frame. The frame sequence trust item is determined based on the continuous state of the frame sequence number of the time-windowed standard core frame, the timestamp trust item is determined based on the offset between the relative core time and the unified time window center time, the link trust item is determined based on the difference between the receiving time and the acquisition timestamp, and the communication trust value is formed by combining them according to a fixed priority. The time-windowed standard capacity frames are sorted according to the communication trust value. The time-windowed standard capacity frame with the highest ranking is retained as the representative frame of the time window, and the corresponding unified time window is determined as the effective unified time window.

4. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 3, characterized in that, The specific steps for performing capacity phase calibration on the standard capacity frame based on the capacity status value to form an aligned capacity frame are as follows: The parsing window represents the core capacity status value in the frame, and the core capacity preparation state corresponds to the core capacity preparation stage, the constant current discharge state corresponds to the core capacity discharge stage, the discharge stop state corresponds to the core capacity cutoff stage, the capacity calculation state corresponds to the capacity calculation stage, and the core capacity end state corresponds to the core capacity end stage, thus forming the candidate core capacity stage. Compare the candidate capacity phase with the confirmed capacity phase of the previous effective unified time window. If the candidate capacity phase has not changed, a capacity phase identifier is generated based on the candidate capacity phase. If the candidate capacity phase has changed, the continuity is confirmed through the candidate capacity phase of the next effective unified time window, and a capacity phase identifier is generated based on the continuity confirmation result. The capacity stage identifier, the valid unified time window number where the time window representative frame is located, and the relative capacity time are bound to the time window representative frame, and the capacity parameters in the time window representative frame are retained to form an aligned capacity frame.

5. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 4, characterized in that, The steps for calculating the battery pack voltage residual based on the battery pack measurement point topology corresponding to the aligned capacity frame, calculating the plate voltage residual when the battery pack voltage residual exceeds the limit, and calculating the section voltage residual when the plate voltage residual exceeds the limit are as follows: Using the valid unified time window number in the aligned capacity frame as an index, extract the group pressure value, plate pressure value and section pressure value from the aligned capacity frame corresponding to the same valid unified time window, and retrieve the battery pack measurement point topology according to the battery pack identifier corresponding to the aligned capacity frame; Based on the battery pack test point topology, determine the plate voltage test points belonging to the same group of test points. Within the same effective unified time window, calculate the difference between the group voltage value and the sum of the plate voltage values ​​belonging to the same group of test points to form the group voltage residual. When the group voltage residual exceeds the limit, determine the plate voltage test point corresponding to the over-limit group voltage test point according to the voltage test point attribution table. The difference between the plate voltage value and the sum of the section voltage values ​​belonging to the same plate voltage measurement point is calculated to form the plate voltage residual. When the plate voltage residual exceeds the limit, the section voltage measurement point corresponding to the plate voltage measurement point that exceeds the limit is determined according to the voltage measurement point assignment table. A throttle pressure sorting sequence is formed by connecting the throttle pressure measurement points in series. Candidate abnormal throttle pressure measurement points are determined by Hampel-type robust window verification. The difference between the current throttle pressure value and the throttle pressure reference value corresponding to the candidate abnormal throttle pressure measurement point is calculated to form the throttle pressure residual.

6. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 5, characterized in that, The steps for sequentially locating contaminated channels according to residual levels and reconstructing abnormal voltages based on voltage sorting sequences to form voltage clean frames are as follows: When the group pressure residual exceeds the limit but the plate pressure residual does not exceed the limit, locate the group pressure measurement point channel; when the plate pressure residual exceeds the limit but the section pressure residual does not exceed the limit, locate the plate pressure measurement point channel; when the section pressure residual exceeds the limit, locate the section pressure measurement point channel, thus forming a pollution channel positioning mark. The abnormal voltage is reconstructed based on the pollution channel location marker. The abnormal group voltage value is replaced by the sum of the corresponding plate voltage values, the abnormal plate voltage value is replaced by the sum of the corresponding section voltage values, and the abnormal section voltage value is replaced by the average of the adjacent non-limited section voltage values ​​in the section voltage sorting sequence, thus forming a voltage clean frame.

7. The multi-parameter data cleaning method for remote capacity assessment of batteries as described in claim 6, characterized in that, The steps for calculating the time window integration capacity based on the group current in the voltage clean frame, calculating the first ampere-hour residual based on the time window integration capacity and the set capacity, and calculating the second ampere-hour residual based on the time window integration capacity and the uploaded capacity are as follows: Select effective unified time windows that are identified as nuclear capacity discharge stages from the voltage clean frame, arrange the group currents in the effective unified time windows according to the unified time window number, and simultaneously retain the time window interval between adjacent effective unified time windows to form a group current integration sequence; Along the current integration sequence, two adjacent effective unified time windows are selected sequentially. The current endpoints corresponding to the two adjacent effective unified time windows form a trapezoidal current boundary, and the time window interval between adjacent effective unified time windows forms a trapezoidal time boundary, thus forming a single window capacity segment. The single-window capacity segments are continuously aggregated along the time window sequence of the nuclear discharge stage to form the time window integrated capacity; The starting point of the nuclear capacity discharge stage is taken as the starting point of the capacity progress. The discharge process is determined according to the relative nuclear capacity time of the current effective unified time window. The set capacity is mapped to the set capacity progress value under the same effective unified time window. The deviation of the time window integral capacity from the set capacity progress value is used to form the first ampere-hour residual. Using the unified time window number as an index, the transmission capacity corresponding to the same valid unified time window is extracted from the voltage clean frame, and the time window integral capacity is aligned with the transmission capacity. Then, the deviation of the time window integral capacity from the transmission capacity is used to form the second ampere-hour residual.

8. A multi-parameter data cleaning system for remote capacity assessment of batteries, based on the multi-parameter data cleaning method for remote capacity assessment of batteries as described in any one of claims 1 to 7, characterized in that, include: The capacity verification module is used to collect capacity verification frames during the remote capacity verification process, perform format verification and range verification on the capacity verification frames, and form standard capacity verification frames. The time axis calibration module is used to map standard nuclear capacity frames to a unified time axis based on the nuclear capacity start time. It determines the communication reliability value based on the continuous status of the frame sequence number of the standard nuclear capacity frame, the offset of the acquisition timestamp relative to the center time of the unified time window, and the difference between the reception time and the acquisition timestamp. The standard nuclear capacity frames in the same time window are sorted according to the communication reliability value, and the standard nuclear capacity frame with the highest ranking is retained. The module also performs nuclear capacity stage calibration on the standard nuclear capacity frames according to the nuclear capacity status value to form aligned nuclear capacity frames. The voltage cleaning module is used to calculate the group voltage residual based on the battery pack measurement point topology corresponding to the aligned core capacity frame, calculate the plate voltage residual when the group voltage residual exceeds the limit, calculate the section voltage residual when the plate voltage residual exceeds the limit, locate the contamination channel according to the residual level in sequence, and reconstruct the abnormal voltage according to the section voltage sorting sequence to form a voltage clean frame. The capacity cleaning module is used to calculate the time window integrated capacity based on the group current in the voltage clean frame, calculate the first ampere-hour residual based on the time window integrated capacity and the set capacity, calculate the second ampere-hour residual based on the time window integrated capacity and the transmitted capacity, identify the cumulative increase pattern of the first and second ampere-hour residuals, back-calculate the group current drift value when the first and second ampere-hour residuals deviate continuously in the same direction, and correct the group current in reverse based on the group current drift value to form a current clean frame. The core capacity is calculated based on the corrected group current in the current clean frame, the cleaning category is marked, and a cleaning core capacity message is output.

Images