A method for monitoring the operation of an energy storage high current connector

Abstract

The present application relates to the technical field of operation determination, in particular to a kind of energy storage large current connector operation monitoring method, in the present application, isolated forest is introduced to execute multiple rounds of random split determination, and the number of times that window is divided into independent interval in multiple split is counted and sorted, so that abnormal identification is not dependent on single threshold trigger, but is based on the consistency of multiple random split to form stable determination basis, so as to reduce the influence of instantaneous disturbance or single-point fluctuation on abnormal identification result, after abnormal window screening, further introduce bayesian change point detection, perform continuous difference calculation on the sequence corresponding to current-carrying loop in window dimension and combine prior distribution to recursively update change probability, so that continuous overrun position can be systematically labeled in time series, and the section boundary is determined by first and last overrun window, so that abnormal identification result is expanded from discrete window level to continuous operation section level.

Description

Technical Field

[0001] This invention relates to the field of operational determination technology, and in particular to a method for monitoring the operation of a high-current energy storage connector. Background Technology

[0002] The field of operational judgment technology aims to perform conditional judgments on the operational status of business objects and data objects during the operation of a computer system. By comparing, performing logical operations, and verifying thresholds on the input data, it generates deterministic judgment results and associates these results with preset management actions and control processes. This enables automatic triggering of object status identification, operational classification, and management strategies, ensuring that the system's operational behavior is stably executed according to predetermined rules.

[0003] The purpose of a big data-based social data analysis and management method is to quantitatively determine the behavioral status and interaction characteristics of social objects by performing statistical calculations on social data and combining them with preset operation judgment rules, thereby forming clear management judgment results to support the classification management and status control of social data objects, achieve consistency in management decision execution, and improve the efficiency of social data management.

[0004] Existing operational judgment technologies primarily rely on comparative operations, logical operations, and threshold checks on input data to determine operational status. Operational status is often judged based on single sampling and short-term statistical results, lacking a characterization of the evolution of status during continuous operation. This leads to jumps and frequent switching in judgment results when there are phased fluctuations and gradual changes. In social data analysis and management scenarios, existing technologies typically directly map statistical results to preset rules. The data processing emphasizes result aggregation while weakening the analysis of the change process, causing management judgments to reflect the state of the result rather than the formation process. When object behavior and operational status exhibit phased anomalies and interval changes, existing judgment methods struggle to accurately identify the boundaries of abnormal segments, resorting only to discrete processing through repeated rule triggering, increasing the frequency of management actions. Furthermore, existing technologies often employ single indicators and simple weighting in risk and status level judgments, with low correlation between indicator selection and time span, failing to reflect the impact of operational duration and status stability on overall risk assessment. This results in short-term anomalies being overemphasized and long-term anomalies being underestimated. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for monitoring the operation of high-current energy storage connectors.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for monitoring the operation of a high-current energy storage connector, comprising the following steps: S1: Based on the current, voltage, contact resistance, temperature rise and power loss values ​​generated by the energy storage high current connector in a continuous operating cycle, align the parameters of the conductive terminal and contact interface according to the cycle, calculate the difference between adjacent cycles and combine them to establish the operating window state sequence and obtain the connector operating window state sequence. S2: Based on the connector operating window state sequence, perform multiple rounds of numerical segmentation judgment on the current change difference, temperature rise change difference and contact resistance change difference corresponding to the contact interface, count the number of separations and sort them, compare them with the abnormal threshold and then filter the window state to obtain the abnormal window filtering result. S3: Based on the abnormal window screening results, select adjacent windows within the connector working section, segment the corresponding sequence of the current-carrying circuit according to the window, calculate the segment difference and mark the continuous over-limit positions to form the section boundary and obtain the abnormal operation section identifier. S4: Based on the abnormal operation section identifier, perform a weighted summation operation on the current load ratio, cumulative running time and operation stability values ​​within the section around the connector load range to form the corresponding value of the section and obtain the section operation risk score; S5: Based on the section operation risk score, compare the score value with the connector operation level range item by item, select the corresponding risk level identifier and match the operation control number to form the section status classification result and obtain the connector operation judgment result.

[0007] As a further aspect of the present invention, the connector operating window state sequence includes an operating window number, a current state identifier, a voltage state identifier, a contact resistance state identifier, and a temperature rise state identifier; the abnormal window screening result includes an abnormality level identifier, an abnormal window number, and a corresponding contact interface identifier; the abnormal operating segment identifier includes a segment start window number, a segment end window number, and a segment corresponding current-carrying circuit identifier; the segment operating risk score includes a segment number, a risk score value, and a score weight identifier; and the connector operating judgment result includes an operating state category identifier, a corresponding abnormal segment number, and an operating control number.

[0008] As a further aspect of the present invention, the specific steps for generating the connector operating window state sequence are as follows: Based on the current, voltage, contact resistance, temperature rise, and power loss values ​​generated by the high-current energy storage connector during continuous operation cycles, the corresponding parameters of the conductive terminals and contact interfaces are arranged in the order of the operation cycles and labeled with cycle numbers to obtain the cycle parameter alignment set. Based on the cycle parameter alignment set, numerical difference calculation is performed on the corresponding parameters of adjacent operating cycles, and the difference results are combined and arranged with the original cycle parameters in a fixed field order to establish a time continuity relationship and obtain the connector operating window state sequence.

[0009] As a further aspect of the present invention, the arrangement according to a fixed field order specifically refers to arranging the parameter values ​​within the same operating cycle and the corresponding differences between adjacent cycles according to a pre-set field position order. The field order includes the operating cycle number field, the current value field, the voltage value field, the contact resistance calculation value field, the temperature rise calculation value field, the power loss calculation value field, and the cycle difference field corresponding to each of the above fields. The original parameter field and the difference field maintain a corresponding relationship in the arrangement, and the field order remains consistent in each operating cycle.

[0010] As a further aspect of the present invention, the specific steps for generating the abnormal window filtering results are as follows: Based on the connector operating window state sequence, the current change difference, temperature rise change difference, and contact resistance change difference corresponding to the contact interface are arranged in order of window number, and the corresponding window positions are marked to establish the change amount alignment relationship and generate the window change amount alignment set. Based on the window change alignment set, an isolated forest is used to set a split boundary value for each window change and perform multiple split judgments. The number of times each window is split into an independent interval is recorded and sorted according to the number of times to generate a window separation sorting set. Based on the window separation and sorting set, the sorting results are compared with the preset abnormal threshold item by item, the window numbers that exceed the threshold are filtered and the time order is retained to form a window state set and obtain the abnormal window filtering results.

[0011] As a further aspect of the present invention, the isolated forest, for each window change in the window change alignment set, randomly selects the numerical dimension corresponding to the change, and generates a splitting boundary value within the range of the selected dimension. The window change is divided into different numerical intervals by the splitting boundary. For window changes falling into the numerical interval, the random dimension selection and splitting boundary generation operation is performed again, and the division process is repeated until the window change is divided into an independent numerical interval that does not contain other window changes. The number of splits experienced in completing the division process is recorded to form a separation count value that corresponds one-to-one with the window change.

[0012] As a further aspect of the present invention, the specific steps for generating the abnormal operating segment identifier are as follows: Based on the abnormal window screening results, adjacent windows before and after the abnormal window are selected in chronological order within the connector working section, and the current sequence, temperature rise sequence and contact resistance sequence of the corresponding current-carrying circuit are extracted to form a section window sequence set. Based on the segment window sequence set, Bayesian variable point detection is used to perform difference calculation on the corresponding values ​​of adjacent windows in the same sequence, and the difference is compared with the over-limit threshold item by item to mark the window positions of continuous over-limit and generate a continuous over-limit position set. Based on the set of continuous out-of-limit locations, the first and last out-of-limit windows are determined as the segment boundary, and the range of segment coverage window numbers is recorded to form a segment identifier, thereby obtaining the abnormal operation segment identifier.

[0013] As a further aspect of the present invention, the Bayesian change point detection, within the segment window sequence set, establishes adjacent window numerical difference sequences in chronological order for the current sequence, temperature rise sequence, and contact resistance sequence, and constructs corresponding probability distribution representations based on the difference sequences. A change point state variable is introduced for each time position, and combined with a preset prior probability, the posterior probability that the current time position belongs to the change point state is recursively calculated. The posterior probability values ​​obtained from consecutive time positions are compared with the over-limit threshold item by item, and time positions that meet the threshold conditions are marked. When multiple adjacent time positions are marked at the same time, the corresponding window number is recorded to form a continuous over-limit position set.

[0014] As a further aspect of the present invention, the specific steps for generating the section operation risk score are as follows: Based on the abnormal operation segment identifier, the current load ratio, cumulative running time and operation stability value corresponding to each segment are extracted and collected according to the segment number to form a segment parameter set. Based on the set of parameters for the section, the current load ratio, cumulative running time, and operating stability values ​​are multiplied by their respective weighting coefficients and then summed to form the corresponding values ​​for the section, thereby obtaining the section's operating risk score.

[0015] As a further aspect of the present invention, the specific steps for generating the connector operation determination result are as follows: Based on the section operation risk score, the score values ​​of each section are compared with the boundary of the connector operation level interval in order of size, and the matching interval number is recorded to form a section level correspondence table. Based on the segment level correspondence table, the segment level identifier is matched one by one with the preset operation control number, and the results are summarized and classified according to the segment number to form the segment status classification result and obtain the connector operation judgment result.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, an isolated forest is introduced to perform multiple rounds of random segmentation judgment, and the number of times the window is divided into independent intervals in multiple segmentations is statistically sorted. This makes anomaly identification not dependent on a single threshold trigger, but based on the consistency of multiple random segmentations to form a stable judgment basis, thereby reducing the impact of instantaneous disturbances or single-point fluctuations on the anomaly identification results. In this invention, after the abnormal window screening, Bayesian change point detection is further introduced. The corresponding sequence of the current-carrying loop is subjected to continuous difference calculation in the window dimension and the change probability is recursively updated in combination with the prior distribution, so that the continuous over-limit positions can be systematically marked in the time series. The segment boundary is determined by the first and last over-limit windows, so that the abnormal identification results are extended from the discrete window level to the continuous running segment level. In this invention, during the risk assessment stage, the proportion of current load within the section, cumulative operating time, and operational stability are incorporated into a unified quantitative framework. A comparable score is formed through weighted summation, so that the risk expression takes into account load intensity, duration, and fluctuation characteristics. By mapping the score to the operating level interval and associating it with the operating control number, the operation judgment result forms a complete link from multi-dimensional operating data and abnormal structure identification to control identifier output. The processing path forms a superimposed gain in terms of the reliability of abnormal identification, the continuity of section boundary identification, and the consistency of risk judgment, making the operation status judgment closer to the actual operation evolution process. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the main steps of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Example 1 Please see Figure 1 This invention provides a technical solution: a method for monitoring the operation of a high-current energy storage connector, comprising the following steps: S1: Based on the current, voltage, contact resistance, temperature rise and power loss values ​​generated by the energy storage high current connector in a continuous operating cycle, align the parameters of the conductive terminal and contact interface according to the cycle, calculate the difference between adjacent cycles and combine them to establish the operating window state sequence and obtain the connector operating window state sequence. S2: Based on the connector operating window state sequence, perform multiple rounds of numerical segmentation judgment on the current change difference, temperature rise change difference and contact resistance change difference corresponding to the contact interface, count the number of separations and sort them, compare with the abnormal threshold and filter the window state to obtain the abnormal window filtering results. S3: Based on the abnormal window screening results, select adjacent windows within the connector working section, segment the corresponding sequence of the current-carrying circuit according to the window, calculate the segment difference and mark the continuous over-limit positions to form the section boundary and obtain the abnormal operation section identifier. S4: Based on the abnormal operation section identifier, perform a weighted summation operation on the current load ratio, cumulative running time and operation stability values ​​within the section around the connector load range to form the corresponding value of the section and obtain the section operation risk score; S5: Based on the section operation risk score, the score value is compared with the connector operation level range item by item, the corresponding risk level identifier is selected and matched with the operation control number to form the section status classification result and obtain the connector operation judgment result.

[0020] The connector operation window status sequence includes the operation window number, current status identifier, voltage status identifier, contact resistance status identifier, and temperature rise status identifier. The abnormal window screening results include the abnormal level identifier, abnormal window number, and corresponding contact interface identifier. The abnormal operation section identifier includes the section start window number, section end window number, and section corresponding current-carrying circuit identifier. The section operation risk score includes the section number, risk score value, and score weight identifier. The connector operation judgment result includes the operation status category identifier, corresponding abnormal section number, and operation control number.

[0021] The specific steps for generating the connector runtime window state sequence are as follows: Based on the current, voltage, contact resistance, temperature rise, and power loss values ​​generated by the high-current energy storage connector during continuous operation cycles, the corresponding parameters of the conductive terminals and contact interfaces are arranged in the order of the operation cycles and labeled with cycle numbers to obtain the cycle parameter alignment set. Based on the periodic parameter alignment set, numerical difference calculation is performed on the corresponding parameters of adjacent running cycles, and the difference results are combined and arranged with the original periodic parameters in a fixed field order to establish a time continuity relationship and obtain the connector running window state sequence. Based on the current, voltage, contact resistance, temperature rise, and power loss calculations generated by the high-current energy storage connector during continuous operation cycles, a sequence parameter alignment method is adopted to process multiple parameters in a unified order within the operation cycle. This sequence parameter alignment method is implemented by constructing a fixed-field mapping table. Specifically, the operation cycle number is set as a unique primary key field. The cycle number is generated in an integer incrementing manner and starts recording from the initial value 1. Under each cycle number, the current, voltage, contact resistance, temperature rise, and power loss calculations are written sequentially. During the writing process, a fixed storage bit width is set for each parameter field, and decimal values ​​are used for representation. After the fields are written, a field order verification operation is performed, including checks on the consistency of the number of fields and the consistency of the field arrangement. After verification, a cycle number label is added to each cycle parameter record, and all cycle parameter records are arranged in chronological order of the operation cycles to generate a cycle parameter alignment set. Based on the period parameter alignment set, an adjacent period difference calculation method is adopted to calculate the numerical difference of corresponding parameters in adjacent operating cycles. This adjacent period difference calculation method is performed through a fixed period interval traversal. Specifically, it uses the cycle number order as the traversal basis, selecting two adjacent cycle records one by one, and performing subtraction operations on the current value field, voltage value field, contact resistance calculation value field, temperature rise calculation value field, and power loss calculation value field. After each difference calculation is completed, the difference result is temporarily stored in the corresponding difference field position. After all field difference calculations are completed, the difference field and the original cycle parameter field are combined and arranged in a preset fixed field order. The combined arrangement includes cycle number, current value, voltage value, contact resistance calculation value, temperature rise calculation value, power loss calculation value, current difference, voltage difference, contact resistance difference, temperature rise difference, and power loss difference. During the combination process, the data is continuously written into the storage structure in the order of cycle number, and a continuous time index is added to each combined record to establish a continuous time relationship and generate a connector operation window state sequence.

[0022] Arranged in a fixed field order, specifically, the parameter values ​​within the same operating cycle and their corresponding differences between adjacent cycles are arranged according to a pre-defined field position order. The field order includes the operating cycle number field, current value field, voltage value field, contact resistance calculation value field, temperature rise calculation value field, power loss calculation value field, and the corresponding cycle difference field for each of the above fields. The original parameter field and the difference field maintain a corresponding relationship in the arrangement, and the field order remains consistent in each operating cycle.

[0023] The specific steps for generating the abnormal window filtering results are as follows: Based on the connector operating window state sequence, the current change difference, temperature rise change difference, and contact resistance change difference corresponding to the contact interface are arranged in order of window number, and the corresponding window positions are marked to establish the change amount alignment relationship and generate the window change amount alignment set. Based on the window change alignment set, an isolated forest is used. A split boundary value is set for each window change and multiple split judgments are performed. The number of times each window is split into an independent interval is recorded and sorted according to the number of times to generate a window separation sort set. Based on the window separation sorting set, the sorting results are compared with the preset abnormal threshold item by item, the window numbers that exceed the threshold are filtered and the time order is retained to form a window state set and obtain the abnormal window filtering results. Based on the connector operating window state sequence, a window change alignment method is adopted to uniformly process the current change difference, temperature rise change difference, and contact resistance change difference corresponding to the contact interface. The window change alignment method is executed by constructing a fixed window index mapping table. Specifically, it reads the window number corresponding to each window in the connector operating window state sequence, uses the window number as a unique index field, traverses the sequence data in ascending order of window number, and writes the three values ​​of current change difference, temperature rise change difference, and contact resistance change difference under each window index in sequence. During the writing process, the three change differences are uniformly stored in double-precision numerical format, and missing fields are filled with preset placeholder values ​​of zero. After each window record is written, a field quantity consistency check operation is performed. After the check is completed, window number annotation information is added to each window change record, and all window change records are arranged in order of window number to establish a change alignment relationship and generate a window change alignment set. Based on the window change alignment set, the isolated forest algorithm is used to perform multiple random splitting judgments on each window change. The isolated forest algorithm is executed by constructing multiple isolated trees. Specifically, the number of isolated trees is set to one hundred, and the maximum splitting depth of each isolated tree is set to the value corresponding to the number of windows in the alignment set. During the construction of each isolated tree, a subset is randomly selected from the window change alignment set and splitting nodes are generated layer by layer. In each splitting node, the difference in current change, the difference in temperature rise, and the difference in contact resistance are used as candidate splitting dimensions in turn. A splitting boundary value is randomly generated on the current splitting dimension, and the window change is divided into two parts according to the splitting boundary. The splitting level corresponding to the independent interval of each window change in the current isolated tree is recorded. The splitting level obtained by the same window in all isolated trees is accumulated and the accumulated result is recorded as the number of times the window is split into independent intervals. After the splitting judgment of all isolated trees is completed, the corresponding number values ​​of all windows are summarized and sorted from largest to smallest to generate a window separation sorting set. Based on a window-separated sorted set, a threshold comparison filtering method is used to compare the sorting results with a preset abnormal threshold item by item. The threshold comparison filtering method is executed by sequential traversal. Specifically, the window number and its corresponding count value are read item by item from the window-separated sorted set in sorting order, and the count value is compared with the preset abnormal threshold. During the comparison, the greater than relationship is used as the judgment condition, and the window numbers that meet the condition are recorded. During the recording process, the original time order of the window numbers is kept unchanged. All window numbers that meet the condition are written into a set structure in sequence. After the traversal is completed, an integrity check operation is performed on the set structure to form a window state set and obtain the abnormal window filtering results.

[0024] Isolation Forest, for each window change in the window change alignment set, randomly selects the corresponding numerical dimension of the change, and generates a split boundary value within the range of the selected dimension. The window change is divided into different numerical intervals by the split boundary. For window changes that fall into the numerical interval, the random dimension selection and split boundary generation operations are repeated, and the division process is repeated until the window change is divided into an independent numerical interval that does not contain other window changes. The number of splits in the division process is recorded to form a separation count value that corresponds one-to-one with the window change.

[0025] The specific steps for generating abnormal operation segment identifiers are as follows: Based on the abnormal window screening results, adjacent windows before and after the abnormal window are selected in chronological order within the connector working section, and the current sequence, temperature rise sequence and contact resistance sequence of the corresponding current-carrying circuit are extracted to form a section window sequence set. Based on the segment window sequence set, Bayesian variable point detection is used to calculate the difference between the corresponding values ​​of adjacent windows in the same sequence, and the difference is compared with the out-of-limit threshold item by item to mark the window positions of continuous out-of-limit and generate a set of continuous out-of-limit positions. Based on the continuous set of over-limit locations, the first and last over-limit windows are determined as the segment boundary, and the range of segment coverage window numbers is recorded to form a segment identifier, thereby obtaining the abnormal operation segment identifier. Based on the abnormal window screening results, a segment window selection method is adopted. Within the connector working section, adjacent windows before and after the abnormal window are selected in chronological order. This segment window selection method is executed using a fixed window expansion rule. Specifically, it reads all abnormal window numbers from the abnormal window screening results, arranges them in ascending order of window number, uses each abnormal window number as a central index, and selects two window numbers forward and two window numbers backward. During the selection process, it determines whether the selected window numbers exceed the range of the connector working section's start and end numbers, and removes any exceeding the range. After completing the window number selection, it extracts the corresponding current-carrying circuit current sequence, temperature rise sequence, and contact resistance sequence from the original operating data based on the selected window numbers. These three types of sequences are written into the sequence storage structure according to the window number order. During the writing process, each sequence is uniformly formatted as a floating-point number and retains four decimal places. After each window data is written, a sequence length consistency check is performed. After the check is completed, the sequences corresponding to multiple windows are combined and stored according to the window number order to form a segment window sequence set. Based on a segmented window sequence set, a Bayesian change point detection method is employed. This method calculates the difference between corresponding values ​​in adjacent windows within the same sequence and performs out-of-limit comparisons. The Bayesian change point detection method is implemented through a sequence-by-sequence recursive calculation. Specifically, independent detection processes are established for the current sequence, temperature rise sequence, and contact resistance sequence. In each detection process, the prior distribution type is set to normal distribution, the prior mean is set to the mean of the first five window values ​​in the corresponding sequence, and the prior variance is set to the variance of the first five window values ​​in the corresponding sequence. During sequence traversal, windows are read sequentially according to their numerical numbers. The corresponding values ​​of two adjacent windows are calculated, and the difference is written to the difference buffer. After the difference is generated, it is compared with the preset over-limit threshold item by item. The greater than relationship is used as the judgment condition during the comparison. When the judgment is true, the over-limit labeling operation is performed on the current window number. During the sequence traversal, the window numbers of consecutive over-limit labeled windows are merged and recorded. During the merging process, the original time order of the window numbers is kept unchanged. After completing the traversal of all three types of sequences, all consecutive over-limit window number records are summarized to generate a set of consecutive over-limit positions. Based on a set of continuous out-of-limit locations, a segment boundary determination method is used to identify the segment boundaries of continuous out-of-limit windows and generate segment identifiers. The segment boundary determination method is executed in a sequential scanning manner. Specifically, it reads all window numbers in the set of continuous out-of-limit locations and arranges them in chronological order. The first window number in the sequence is determined as the segment start boundary, and the last window number in the sequence is determined as the segment end boundary. During the boundary determination process, the start and end values ​​of the segment coverage window numbers are recorded. The start and end values ​​are combined and written into the segment identifier field. During the writing process, a segment index number is appended and numbered in an integer incrementing manner. After the segment identifier is generated, a field integrity verification operation is performed on the identifier content to form the segment identifier and obtain the abnormal operation segment identifier.

[0026] Bayesian change point detection, within a segment window sequence set, establishes a numerical difference sequence of adjacent windows in chronological order for the current sequence, temperature rise sequence, and contact resistance sequence, and constructs a corresponding probability distribution representation based on the difference sequence. A change point state variable is introduced for each time position, and combined with a preset prior probability, the posterior probability of the current time position belonging to the change point state is recursively calculated. The posterior probability values ​​obtained from consecutive time positions are compared with the over-limit threshold item by item, and time positions that meet the threshold condition are marked. When multiple adjacent time positions are marked at the same time, the corresponding window number is recorded to form a continuous over-limit position set.

[0027] The specific steps for generating a section operation risk score are as follows: Based on the abnormal operation section identifier, the current load ratio, cumulative running time and operation stability values ​​corresponding to each section are extracted and collected according to the section number to form a section parameter set. Based on the section parameter set, the current load ratio, cumulative running time and operation stability values ​​are multiplied by the corresponding weight coefficients and accumulated to form the corresponding values ​​for the section, and the section operation risk score is obtained. Based on the abnormal operation segment identifier, a segment parameter extraction method is adopted to calculate and collect the corresponding parameters for each segment. This method is executed through segment index traversal, specifically by reading all segment numbers from the abnormal operation segment identifier and traversing them sequentially in ascending order. During the traversal, the segment start window number and segment end window number are used as boundary conditions. Within the number range, the current value records of the current-carrying circuits are read window by window from the original operation window data. The ratio of the current value in each window to the rated current value is calculated, and the ratio results are accumulated and divided by the number of windows to obtain the current load ratio. Furthermore, the operating time corresponding to the window within the same segment is statistically analyzed and calculated by window. The duration is accumulated item by item to obtain the cumulative runtime. Within the same segment, three types of time series, namely current value, temperature rise value and contact resistance value, are read and arranged in order of window number. The difference between adjacent window sequence values ​​is calculated item by item and the absolute value of the difference is written into the stability calculation cache. All difference values ​​in the cache are summed and divided by the number of windows to obtain the running stability value. After the parameter calculation of each segment is completed, the segment number, current load ratio, cumulative runtime and running stability value are written into the segment parameter storage structure in a fixed field order. After writing, the field integrity verification operation is performed. After all segments are traversed, the parameters are collected by segment number to form the segment parameter set. Based on the segment parameter set, a weighted cumulative scoring method is adopted to perform weighted calculation and accumulation processing on the current load ratio, cumulative running time, and operational stability values. The weighted cumulative scoring method is executed through a fixed weight mapping rule. Specifically, before the calculation begins, the current load ratio weight coefficient is preset to 0.4, the cumulative running time weight coefficient is 0.3, and the operational stability weight coefficient is 0.3. During the traversal of the segment parameter set, the current load ratio value corresponding to each segment is read one by one and multiplied with the corresponding weight coefficient, and the result is written to the scoring accumulation cache. The cumulative running time value is read one by one and multiplied with the corresponding weight coefficient and accumulated to the scoring accumulation cache. The operational stability value is read one by one and multiplied with the corresponding weight coefficient and accumulated to the scoring accumulation cache. After the weighted calculation of the three parameters is completed, the accumulation result is used as the corresponding value of the segment and written to the segment scoring field. After all segments have been calculated, the scoring results are output in the order of segment number to obtain the segment operation risk score.

[0028] The specific steps for generating the connector operation judgment result are as follows: Based on the section operation risk score, the score values ​​of each section are compared with the boundary of the connector operation level interval in order of size, and the matching interval number is recorded to form a section level correspondence table. Based on the section level correspondence table, the section level identifier is matched one by one with the preset operation control number, and the results are summarized and classified according to the section number to form the section status classification result and obtain the connector operation judgment result. Based on the segment operation risk score, an interval boundary comparison method is adopted to compare the corresponding score value of each segment with the boundary of the connector operation level interval. The interval boundary comparison method is executed by sequential scanning and interval mapping. Specifically, before the processing starts, a preset connector operation level interval boundary table is loaded and stored as an array of boundaries arranged in ascending order of level number. The array stores the lower boundary value and upper boundary value corresponding to each level in sequence. During the traversal of the segment operation risk score, the segment score value is read one by one in the order of segment number, and an interval judgment operation is performed on the read score value. The interval judgment operation specifically compares the score value with the lower boundary of the first level interval and determines whether it is greater than or equal to the lower boundary and less than the corresponding upper boundary. If it does not meet the condition, it continues to compare with the next level interval boundary. During the judgment process, an interval sequential traversal method is used until a match is successful. When a match is successful, the current level interval number is recorded, and the segment number and level interval number are written into the corresponding mapping record. After each record is written, a number consistency check operation is performed. After all segment score comparisons are completed, the mapping records are sorted in order to form a segment level correspondence table. Based on the segment-level correspondence table, a control number matching and classification method is adopted to match the segment-level identifier with the preset operation control number one by one and perform summary and classification processing. The control number matching and classification method is executed through a level index mapping method. Specifically, before the processing starts, a preset operation control number table is loaded, and the mapping relationship between the level number and the operation control number is stored in the table. During the traversal of the segment-level correspondence table, the segment number and the corresponding level number are read one by one, and the matching operation control number is searched in the operation control number table according to the level number. After the search is completed, the segment number, level number and operation control number are combined and written into the classification record structure. During the writing process, the order of the segment numbers remains unchanged. After all segments are matched, an integrity verification operation is performed on the classification record structure, and the segment numbers are summarized and classified to generate classification output records, forming the segment status classification result and obtaining the connector operation judgment result.

[0029] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for monitoring the operation of a high-current energy storage connector, characterized in that, Includes the following steps: S1: Based on the current, voltage, contact resistance, temperature rise and power loss values ​​generated by the energy storage high current connector in a continuous operating cycle, align the parameters of the conductive terminal and contact interface according to the cycle, calculate the difference between adjacent cycles and combine them to establish the operating window state sequence and obtain the connector operating window state sequence. S2: Based on the connector operating window state sequence, perform multiple rounds of numerical segmentation judgment on the current change difference, temperature rise change difference and contact resistance change difference corresponding to the contact interface, count the number of separations and sort them, compare them with the abnormal threshold and then filter the window state to obtain the abnormal window filtering result. S3: Based on the abnormal window screening results, select adjacent windows within the connector working section, segment the corresponding sequence of the current-carrying circuit according to the window, calculate the segment difference and mark the continuous over-limit positions to form the section boundary and obtain the abnormal operation section identifier. S4: Based on the abnormal operation section identifier, perform a weighted summation operation on the current load ratio, cumulative running time and operation stability values ​​within the section around the connector load range to form the corresponding value of the section and obtain the section operation risk score; S5: Based on the section operation risk score, compare the score value with the connector operation level range item by item, select the corresponding risk level identifier and match the operation control number to form the section status classification result and obtain the connector operation judgment result.

2. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The connector operation window state sequence includes operation window number, current status identifier, voltage status identifier, contact resistance status identifier, and temperature rise status identifier. The abnormal window screening result includes abnormal level identifier, abnormal window number, and corresponding contact interface identifier. The abnormal operation section identifier includes section start window number, section end window number, and section corresponding current-carrying circuit identifier. The section operation risk score includes section number, risk score value, and score weight identifier. The connector operation judgment result includes operation status category identifier, corresponding abnormal section number, and operation control number.

3. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The specific steps for generating the connector's running window state sequence are as follows: Based on the current, voltage, contact resistance, temperature rise, and power loss values ​​generated by the high-current energy storage connector during continuous operation cycles, the corresponding parameters of the conductive terminals and contact interfaces are arranged in the order of the operation cycles and labeled with cycle numbers to obtain the cycle parameter alignment set. Based on the cycle parameter alignment set, numerical difference calculation is performed on the corresponding parameters of adjacent operating cycles, and the difference results are combined and arranged with the original cycle parameters in a fixed field order to establish a time continuity relationship and obtain the connector operating window state sequence.

4. The method for monitoring the operation of a high-current energy storage connector according to claim 3, characterized in that, The arrangement according to a fixed field order specifically refers to arranging the parameter values ​​within the same operating cycle and the corresponding differences between adjacent cycles according to a pre-set field position order. The field order includes the operating cycle number field, current value field, voltage value field, contact resistance calculation value field, temperature rise calculation value field, power loss calculation value field, and the cycle difference field corresponding to each of the above fields. The original parameter field and the difference field maintain a corresponding relationship in the arrangement, and the field order remains consistent in each operating cycle.

5. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The specific steps for generating the abnormal window filtering results are as follows: Based on the connector operating window state sequence, the current change difference, temperature rise change difference, and contact resistance change difference corresponding to the contact interface are arranged in order of window number, and the corresponding window positions are marked to establish the change amount alignment relationship and generate the window change amount alignment set. Based on the window change alignment set, an isolated forest is used to set a split boundary value for each window change and perform multiple split judgments. The number of times each window is split into an independent interval is recorded and sorted according to the number of times to generate a window separation sorting set. Based on the window separation and sorting set, the sorting results are compared with the preset abnormal threshold item by item, the window numbers that exceed the threshold are filtered and the time order is retained to form a window state set and obtain the abnormal window filtering results.

6. The method for monitoring the operation of a high-current energy storage connector according to claim 5, characterized in that, The isolated forest, for each window change in the window change alignment set, randomly selects the numerical dimension corresponding to the change, and generates a split boundary value within the range of the selected dimension. The window change is divided into different numerical intervals by the split boundary. For window changes falling into the numerical interval, the random dimension selection and split boundary generation operation is performed again, and the division process is repeated until the window change is divided into an independent numerical interval that does not contain other window changes. The number of splits performed to complete the division process is recorded to form a separation count value that corresponds one-to-one with the window change.

7. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The specific steps for generating the abnormal operation segment identifier are as follows: Based on the abnormal window screening results, adjacent windows before and after the abnormal window are selected in chronological order within the connector working section, and the current sequence, temperature rise sequence and contact resistance sequence of the corresponding current-carrying circuit are extracted to form a section window sequence set. Based on the segment window sequence set, Bayesian variable point detection is used to perform difference calculation on the corresponding values ​​of adjacent windows in the same sequence, and the difference is compared with the over-limit threshold item by item to mark the window positions of continuous over-limit and generate a continuous over-limit position set. Based on the set of continuous out-of-limit locations, the first and last out-of-limit windows are determined as the segment boundary, and the range of segment coverage window numbers is recorded to form a segment identifier, thereby obtaining the abnormal operation segment identifier.

8. The method for monitoring the operation of a high-current energy storage connector according to claim 7, characterized in that, The Bayesian change point detection, within the segment window sequence set, establishes adjacent window numerical difference sequences in chronological order for the current sequence, temperature rise sequence, and contact resistance sequence, and constructs corresponding probability distribution representations based on the difference sequences. A change point state variable is introduced for each time position, and the posterior probability of the current time position belonging to the change point state is recursively calculated in combination with a preset prior probability. The posterior probability values ​​obtained from consecutive time positions are compared with the over-limit threshold item by item, and time positions that meet the threshold conditions are marked. When multiple adjacent time positions are marked at the same time, the corresponding window number is recorded to form a continuous over-limit position set.

9. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The specific steps for generating the operational risk score for the aforementioned section are as follows: Based on the abnormal operation segment identifier, the current load ratio, cumulative running time and operation stability value corresponding to each segment are extracted and collected according to the segment number to form a segment parameter set. Based on the set of parameters for the section, the current load ratio, cumulative running time, and operating stability values ​​are multiplied by their respective weighting coefficients and then summed to form the corresponding values ​​for the section, thereby obtaining the section's operating risk score.

10. The method for monitoring the operation of a high-current energy storage connector according to claim 1, characterized in that, The specific steps for generating the connector operation determination result are as follows: Based on the section operation risk score, the score values ​​of each section are compared with the boundary of the connector operation level interval in order of size, and the matching interval number is recorded to form a section level correspondence table. Based on the segment level correspondence table, the segment level identifier is matched one by one with the preset operation control number, and the results are summarized and classified according to the segment number to form the segment status classification result and obtain the connector operation judgment result.

Images