An electronic ignition module production system

Abstract

The application discloses an electronic ignition module production system, and particularly relates to the technical field of electronic ignition module production. The electronic ignition module production system is characterized in that: batch quality data of the same batch of electronic ignition modules is collected, the batch quality data is merged, breakpoints are marked, and short-time impact and stable deviation are distinguished; then, local deviation segments of different processes are extracted and matched with historical fault rework records to identify high-risk segments; the direction of transmission of the high-risk segments, the span of continuous stations, and the consistency of terminal detection deviation are analyzed to form a batch coupling risk result; finally, the batch coupling risk result is used to determine rework processes and correction processes, a closed-loop management and control of quality detection, risk identification and process correction is formed, and the batch consistency and quality stability of the electronic ignition module are improved.

Description

Technical Field

[0001] This invention relates to the field of electronic ignition module manufacturing technology, and more specifically, to an electronic ignition module manufacturing system. Background Technology

[0002] Existing electronic ignition module production processes often suffer from incomplete data coverage, lack of risk correlation in handling abnormal data, insufficient prediction accuracy, and non-closed-loop control processes. For example, patent application CN121576868A mentions that in the existing production process of ignition-free electronic detonators, quality data between production stages are usually collected and analyzed independently. Outlier handling is often limited to simple rejection or uniform treatment, lacking effective utilization of historical failure rework data and failing to differentiate the severity of the impact of different abnormal data on quality. Furthermore, existing technologies lack precision control and risk warning mechanisms for sensitivity index prediction results. Risk identification is often delayed, and timely process adjustments and secondary verification are difficult to implement, resulting in low rework efficiency for non-conforming batches, poor batch quality stability, and significant resource waste. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an electronic ignition module production system to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: An electronic ignition module production system, comprising: Quality acquisition module: Collects charge and discharge curve data, connection and continuity data, material status data, assembly process data and ignition response detection data of the same production batch of electronic ignition modules to form batch quality data; Breakpoint merging module: Merges batch quality data according to process nodes, marks the collected breakpoint data, retains adjacent data before and after exceeding the limit, and distinguishes between short-term impact segment and stable offset segment; Deviation extraction module: Extracts the inflection point of the charge-discharge curve based on the short-term impact segment and the stable offset segment, connects the conduction fluctuation segment, the material state gradual change segment and the assembly process reversal segment to form local deviation segments; Risk matching module: Matches local deviation segments with the rework locations, number of reworks, and re-inspection results in historical fault rework records to form deviation risk segments; Coupling determination module: Based on the sequential transmission direction of deviation risk segments among electronic control module, material state and assembly process, continuous work station span and terminal detection offset consistency, batch coupling risk results are generated; Re-mining and calibration module: Based on the batch coupling risk results, determine the re-mining process and calibration process, and generate batch quality data again after executing the calibration process.

[0005] In a preferred embodiment, charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response detection data of the same production batch of electronic ignition modules are collected to form batch quality data, specifically: When the same production batch of electronic ignition modules enters the process node, the following data are collected respectively: charge and discharge curve data formed by the electronic control module, connection and conduction data formed by the conductive connection position, material state data formed by the input production materials, assembly process data formed by the assembly operation position, and ignition response detection data formed by the ignition response verification position. Establish a correspondence between production batch identifiers, process nodes, and data collection times to form batch quality data.

[0006] In a preferred embodiment, batch quality data is merged according to process nodes, data collection breakpoints are marked, adjacent data before and after exceeding limits are retained, and short-term impact segments and stable offset segments are distinguished. Specifically: Batch quality data are merged according to production batch identifier and process node, and the merged batch quality data are arranged according to the time of collection. Data that is missing at the acquisition time or whose acquisition sequence is interrupted is marked as acquisition breakpoint data; Data that deviates from the preset range of the corresponding process node is considered out-of-limit data, and adjacent data before and after the out-of-limit data are retained. Based on the continuous state of the out-of-limit data and the adjacent data before and after the out-of-limit data, the short-term impact segment and the stable offset segment are distinguished.

[0007] In a preferred embodiment, the short-term impact segment and the stable offset segment are distinguished based on the continuous state of the out-of-limit data and the adjacent data before and after the out-of-limit data, specifically as follows: Connect the out-of-limit data with the adjacent data before and after the out-of-limit data according to the collection time to form the out-of-limit neighborhood data; In the out-of-limit neighborhood data, when the data deviates from the preset range of the corresponding process node only appears within the adjacent collection time and the adjacent data before and after the out-of-limit data returns to the preset range of the corresponding process node, the out-of-limit neighborhood data is classified into the short-term impact period. When data deviating from the preset range of the corresponding process node maintains the same offset direction continuously along the acquisition time, the data in the out-of-limit neighborhood will be classified into the stable offset segment.

[0008] In a preferred embodiment, the inflection point of the charge-discharge curve is extracted based on the short-term impact segment and the stable offset segment, and the conduction fluctuation segment, the material state gradual change segment, and the assembly process reversal segment are connected to form a local deviation segment, specifically: Using the short-term impact segment and the stable offset segment as positioning objects, the location where the change direction changes is extracted from the charge-discharge curve data to form the inflection point of the charge-discharge curve; Extract the continuous reciprocating range of the corresponding short-term impact segment and stable offset segment from the connection and conduction data to form the connection and conduction fluctuation segment; Extract the continuous unidirectional change interval of the corresponding stable offset segment from the material state data to form the material state gradual change segment; Extract the intervals where the direction of process change changes from the assembly process data to form the assembly process reversal segment; The inflection points of the charge / discharge curves, the connection and conduction fluctuation segments, the material state gradual change segments, and the assembly process reversal segments are collected according to the process nodes and the time of data collection to form local deviation segments.

[0009] In a preferred embodiment, the local deviation segment is matched with the rework location, number of reworks, and re-inspection results in the historical fault rework record to form a deviation risk segment, specifically: Read local deviation segments according to process nodes and acquisition times; Establish a correspondence between the inflection point of the charge / discharge curve, the connection and conduction fluctuation segment, the material state gradual change segment, and the assembly process reversal segment in the local deviation segment and the rework location in the historical fault rework record. Based on the corresponding relationship, read the number of reworks and the results of re-inspection, and mark the local deviation segments with consistent rework parts, duplicate rework records, and unqualified re-inspection results as deviation risk segments.

[0010] In a preferred embodiment, the number of rework attempts and the re-inspection results are read according to the correspondence. Local deviation segments where the rework locations are consistent, the number of rework attempts is duplicated, and the re-inspection results are unqualified are marked as deviation risk segments. Specifically: Establish a correspondence between the location of the local deviation segment and the rework part in the historical fault rework record according to the process node and the time of data collection; When the location of the local deviation segment is at the process node corresponding to the rework part or the process node adjacent to the rework part, read the number of reworks and the re-inspection results; Local deviation segments with duplicate rework records and unqualified re-inspection results are classified as deviation risk segments.

[0011] In a preferred embodiment, batch coupling risk results are generated based on the sequential transmission direction, continuous workstation span, and terminal detection offset consistency of the deviation risk segments among the electronic control module, material state, and assembly process. Specifically: The deviation risk segments are arranged according to the process node and the time of data collection, and the order of occurrence of the deviation risk segments among the electronic control module, material status and assembly process is identified to form the direction of sequential transmission. The span of continuous workstations is determined based on the consecutive process nodes crossed by the deviation risk segment; The offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node is compared with the offset direction of the deviation risk segment in the corresponding process node to form the terminal detection offset consistency. Based on the sequential transmission direction, continuous workstation span, and consistency of terminal detection offset, batch coupling risk results are generated.

[0012] In a preferred embodiment, the offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node is compared with the offset direction of the deviation risk segment in the corresponding process node to form terminal detection offset consistency, specifically: Read the offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node, and read the offset direction of the deviation risk segment in the corresponding process node; When the offset direction of the ignition response detection data and the offset direction of the deviation risk segment in the corresponding process node are on the same side of the preset range of the corresponding process node, a consistent mark is formed. When the offset direction of the ignition response detection data is opposite to the offset direction of the deviation risk segment in the corresponding process node, an inconsistency mark is formed; Terminal detection offset consistency is formed based on consistent and inconsistent markers.

[0013] In a preferred embodiment, the re-mining process and the calibration process are determined based on the batch coupling risk results. After performing the calibration process, batch quality data is generated again. Specifically: The starting node of the remining process is determined based on the direction of sequential transfer, and the ending node of the remining process is determined based on the span of the continuous work station. Based on the consistency of terminal detection offset, a correction process is determined within the re-sampling process; After the calibration process is performed, charge-discharge curve data, connection continuity data, material status data, assembly process data, and ignition response test data are re-acquired in the re-collection process. Correspondence is established according to the production batch identifier, process node, and collection time to form the regenerated batch quality data.

[0014] The technical effects and advantages of the electronic ignition module production system of the present invention are as follows: By collecting charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response detection data from the same production batch of electronic ignition modules, batch quality data covering electronic performance, connection status, material state, assembly process, and terminal response is generated. Batch quality data is merged according to process nodes, retaining adjacent data before and after exceeding limits, thus avoiding the direct weakening or deletion of short-term impacts and stability deviations by conventional anomaly handling. By extracting inflection points of charge / discharge curves, connection continuity fluctuation segments, material state gradual change segments, and assembly process reversal segments, local deviation characteristics can be identified from different processes. By matching local deviation segments with historical fault rework records, deviation risk segments related to rework locations, rework times, and re-inspection results can be screened. Batch coupling risk results are generated by considering sequential transmission direction, continuous workstation span, and terminal detection offset consistency, identifying hidden quality risks formed by cross-process coupling. Based on the batch coupling risk results, re-collection and correction processes are determined, and batch quality data is regenerated after correction, forming a closed loop of verification and correction in the production process, thereby improving the batch consistency and quality stability of electronic ignition modules. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of an electronic ignition module production system according to the present invention. Detailed Implementation

[0016] The present invention will be further described below through specific embodiments and in conjunction with the accompanying drawings.

[0017] Example 1 Figure 1 The present invention discloses an electronic ignition module production system, comprising: Quality acquisition module: Collects charge and discharge curve data, connection and continuity data, material status data, assembly process data and ignition response detection data of the same production batch of electronic ignition modules to form batch quality data; Breakpoint merging module: Merges batch quality data according to process nodes, marks the collected breakpoint data, retains adjacent data before and after exceeding the limit, and distinguishes between short-term impact segment and stable offset segment; Deviation extraction module: Based on the short-term impact segment and the stable offset segment, extract the inflection point of the charge and discharge curve, connect the conduction fluctuation segment, the material state gradual change segment and the assembly process reversal segment to form a local deviation segment; Risk matching module: Matches local deviation segments with the rework locations, number of reworks, and re-inspection results in historical fault rework records to form deviation risk segments; Coupling determination module: Based on the sequential transmission direction of deviation risk segments among electronic control module, material state and assembly process, continuous work station span and terminal detection offset consistency, batch coupling risk results are generated; Re-mining and calibration module: Based on the batch coupling risk results, determine the re-mining process and calibration process, and generate batch quality data again after executing the calibration process.

[0018] Data on charge / discharge curves, connectivity data, material status data, assembly process data, and ignition response detection data of the same production batch of electronic ignition modules are collected to form batch quality data, including: The term "same production batch" for electronic ignition modules refers to a set of electronic ignition modules under the same production task, tracked using the same production batch identifier, and sequentially passing through the electronic control module detection position, conductive connection position, material input position, assembly operation position, and ignition response verification position. The production batch identifier is formed by combining production date information, production line identification information, batch number information of input materials, and batch sequence information. The production batch identifier is written into the production record before the electronic ignition module enters its first process node. A process node refers to a fixed location during the production process of the electronic ignition module where data acquisition occurs. The acquisition time refers to the point in time when the corresponding data is read and written into the production record. The acquisition time is recorded using a unified clock source, preferably a continuously running clock recording device within the same production site.

[0019] When the electronic ignition module from the same production batch enters the electronic control module's detection position, the charging and discharging process generated by the electronic control module is read, forming charging and discharging curve data. The charging and discharging curve data is stored in a sequential recording method arranged by acquisition time. Each record includes the production batch identifier, process node, acquisition time, detection stage, detection value type, and detection value. The detection stage distinguishes between the charging start stage, charging duration stage, discharging start stage, and discharging duration stage. The detection value type distinguishes between voltage records and current records. To ensure that the charging and discharging curve data accurately reflects the charging and discharging changes generated by the electronic control module, the detection line is read under no-load conditions before reading the data to obtain an no-load reference record. Then, the detection line is connected to the electronic ignition module and continuously read. The detected value after reading is subtracted from the no-load reference record with the closest acquisition time to form the detected value written into the production record. The acquisition time of the no-load reference record and the acquisition time of the charging and discharging curve data are recorded using the same clock source. The matching principle between the no-load reference record and the charging and discharging curve data is the closest acquisition time principle, i.e., the no-load reference record with the smallest time difference is selected as the subtraction object. If the same electronic ignition module is repeatedly read at the detection position of the electronic control module, all repeated reading results are retained, and the corresponding acquisition time is recorded separately. The repeated reading results are not averaged.

[0020] When the electronic ignition module enters the conductive connection position within the same production batch, it reads the conduction status formed at the conductive connection position, generating connection continuity data. This data includes the production batch identifier, process node, acquisition time, conduction resistance record, conduction voltage drop record, and conduction status record. The conduction resistance record is calculated by converting the voltage-current correspondence under stable input current conditions. The conduction voltage drop record is formed by reading the instantaneous potential difference between the two ends of the conductive connection position. The conduction status record is formed based on the positional relationship between the conduction resistance record and the conduction voltage drop record relative to a preset range of the conductive connection position. The preset range of the conductive connection position is established by selecting a set of qualified conduction records formed during continuous stable operation of the same production line, then arranging them separately according to the conduction resistance record and the conduction voltage drop record. Records significantly higher or lower than the concentrated area are removed, retaining the middle concentrated area as the preset range of the conductive connection position. Records significantly higher or lower than the concentrated area can be eliminated by removing a small number of records located at the two ends of the arrangement; for example, a predetermined proportion of records located at the highest and lowest ends of the arrangement can be eliminated. Once the preset range of conductive connection positions is established, it remains unchanged during the same batch of production tasks until the preset range of conductive connection positions is re-established next time.

[0021] When the electronic ignition module receives materials from the same production batch, it reads the material state of the materials and generates material state data. This data includes the production batch identifier, process node, acquisition time, batch number record of the materials, storage environment temperature record, storage environment humidity record, particle state record, and appearance record. The storage environment temperature and humidity records are read before the materials enter the process node and written to the production record at the acquisition time closest to that node. The particle state record is generated based on sieving results, particle size observation results, or particle uniformity inspection results. The appearance record is generated based on color, agglomeration, adhesion, and foreign matter inspection results. To ensure the correspondence between the material state data and the same production batch of the electronic ignition module, the batch number record is written to the production record simultaneously when the materials enter the process node and is stored together with the production batch identifier. If materials from different sources are used in the same production batch at different times, separate batch number records are generated, with the acquisition time used to distinguish the order.

[0022] When the electronic ignition module enters the assembly work position within the same production batch, it reads the assembly process formed at the assembly work position, generating assembly process data. This assembly process data includes production batch identifier, process node, acquisition time, assembly pressure record, assembly displacement record, assembly temperature record, assembly speed record, and work status record. The assembly pressure record is generated using a pressure reading device, the assembly displacement record using a displacement reading device, the assembly temperature record using a temperature reading device, and the assembly speed record is obtained by dividing the difference in assembly displacement records at adjacent acquisition times by the time difference between adjacent acquisition times. The work status record is determined based on whether the assembly pressure record, assembly displacement record, assembly temperature record, and assembly speed record form a complete and continuous record. A complete and continuous record means that data records exist continuously arranged according to the acquisition time from the start to the end of assembly.

[0023] When the electronic ignition module enters the ignition response verification position within the same production batch, it reads the response process formed at the ignition response verification position to generate ignition response detection data. The ignition response detection data includes the production batch identifier, process node, acquisition time, verification start record, response occurrence record, response duration record, and response status record. The verification start record is written at the start of the verification action; the response occurrence record is written when a response change is detected; the response duration record is written continuously according to the acquisition time after the response occurs; and the response status record is formed based on the time correspondence between the verification start record, response occurrence record, and response duration record. To ensure the feasibility of the ignition response detection data, the response occurrence record is determined as follows: when the continuously read response value deviates from the pre-verification baseline value beyond a preset range, the acquisition time at which the first deviation from the preset range occurs is written as the response occurrence record. The method for establishing the preset variation range is to read the continuous benchmark records when no review action is performed, then calculate the fluctuation range of the continuous benchmark records, and take the range after reserving a safety margin outside the fluctuation range as the preset variation range. The safety margin can be set according to a fixed proportion of the fluctuation range of the continuous benchmark records, for example, it can be set to a small proportion of the width of the fluctuation range of the continuous benchmark records.

[0024] After collecting charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response test data, all data are written to the same data record set. Each data entry in the data record set retains the production batch identifier, process node, acquisition time, data category, and test content. The data category distinguishes between charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response test data, while the test content stores the specific record value or status record under the corresponding data category. To avoid data confusion between different process nodes, a source location record is also written to each data entry when writing to the data record set, maintaining a one-to-one correspondence between the source location record and the process node. If multiple data entries of different data categories are written at the same acquisition time, multiple records are retained separately. If acquisition fails at the same process node, an acquisition failure marker is written to the data record set, and the production batch identifier, process node, acquisition time, and data category are retained so that a complete correspondence can still be established later based on the production batch identifier, process node, and acquisition time.

[0025] When establishing the correspondence based on production batch identifier, process node, and data acquisition time, the production batch identifier is used as the affiliation condition to group all data belonging to the same electronic ignition module and the same production batch into the same batch data set. Then, the process node is used as the classification condition to divide the same batch data set into electronic control module detection location data set, conductive connection location data set, material input location data set, assembly operation location data set, and ignition response verification location data set. Finally, the acquisition time is used as the sorting condition to arrange the data within each process node in chronological order. For data with the same acquisition time in the electronic control module detection location data set, conductive connection location data set, material input location data set, assembly operation location data set, and ignition response verification location data set, a parallel storage method is used. For data with missing acquisition times, a missing acquisition time marker is retained, and the corresponding record is not deleted. For data with empty detection content, a missing detection content marker is retained, and the production batch identifier, process node, and data category are not overwritten with null values.

[0026] The same batch of data is connected sequentially according to the process nodes to form a batch quality data table. Each row in the batch quality data table corresponds to an independent data record, and each column corresponds to the production batch identifier, process node, acquisition time, data category, source location record, and test content, respectively. The batch quality data formed in this way retains the source differences between the charge and discharge curve data generated by the electronic control module, the connection continuity data generated by the conductive connection location, the material state data generated by the input production materials, the assembly process data generated by the assembly operation location, and the ignition response test data generated by the ignition response verification location, while also retaining the sequential relationship of the same production batch of the same electronic ignition module at each process node.

[0027] Batch quality data is merged according to process nodes, data collection breakpoints are marked, adjacent data before and after exceeding limits are retained, and short-term impact segments and stable offset segments are distinguished, including: The system reads the production batch identifier from the batch quality data and groups all data records with the same production batch identifier into the data set corresponding to the same production batch of the same electronic ignition module. Data records with different production batch identifiers are not included in the same merging process to avoid data overlap between different electronic ignition modules of the same production batch. After merging the production batch identifiers, the system continues to read the process nodes in the data set corresponding to the same production batch of the same electronic ignition module. Based on the process nodes, the system divides the data set corresponding to the same production batch of the same electronic ignition module into the following sets: electronic control module detection position record set, conductive connection position record set, production material input position record set, assembly operation position record set, and ignition response verification position record set.

[0028] After merging batch quality data according to production batch identifiers and process nodes, the record sets corresponding to each process node are arranged according to the time of data collection. The time of data collection uses a unified clock source to generate a specific point in time, and the collection times are arranged chronologically.

[0029] Missing acquisition time refers to the presence of a missing acquisition time marker in the data records under the same production batch identifier and the same process node, or the absence of a corresponding acquisition time in a testing process that should form continuous records. Interrupted acquisition sequence refers to a time gap exceeding the allowable acquisition interval between adjacent acquisition times in the data records arranged according to acquisition time under the same production batch identifier and the same process node. The allowable acquisition interval is determined by the continuous trial acquisition records of the same process node before the start of the same production task. The determination method is to read the time interval between adjacent acquisition times in the continuous trial acquisition records, remove the time intervals marked as acquisition failures, and take the concentrated area of ​​the remaining time intervals as the allowable acquisition interval. For example, the remaining time intervals can be arranged by size, and a predetermined proportion of the records at the highest and lowest ends can be removed. The maximum time interval of the retained records can then be used as the upper limit of the allowable acquisition interval. Data records with missing acquisition times or interrupted acquisition sequences are marked as acquisition breakpoint data. Acquisition breakpoint data is still retained in the batch quality data and is not deleted as out-of-limit data.

[0030] The preset range for corresponding process nodes is used to determine whether the detection content within the same process node deviates from the data range corresponding to the production record. The preset range for corresponding process nodes is established separately according to process node and data category. For example, the preset range for charge / discharge curve data corresponds to the detection position of the electronic control module; the preset range for connection continuity data corresponds to the detection position of the conductive connection; the preset range for material status data corresponds to the input material entry position; the preset range for assembly process data corresponds to the assembly operation position; and the preset range for ignition response verification corresponds to the ignition response detection data. The preset range for corresponding process nodes is established based on qualified batch quality data generated during the continuous and stable operation of the same production line. Qualified batch quality data is filtered according to production batch identifier, process node, and data category. Data records corresponding to data acquisition breakpoints, acquisition failure markers, and missing detection content markers are then removed. Subsequently, the preset range for corresponding process nodes is determined according to the type of detection content. For numerical detection content, the data is arranged according to the magnitude of the detection content value. Records deviating from the concentrated area at both ends of the arrangement are removed, and the minimum to maximum value of the retained records is used as the preset range for the corresponding process node. For status-based detection content, the set of status records allowed to appear in the qualified batch quality data is determined as the preset range for the corresponding process node. For curve-type detection content, the curve sampling points are read according to the acquisition time, and then the corresponding process node preset ranges are established for the voltage records and current records in the same acquisition stage.

[0031] The determination of out-of-limit data is based on the merged and arranged batch quality data, the preset range of the corresponding process node, and the detection content. For numerical detection content, if the detection content is lower than the lower limit or higher than the upper limit of the preset range of the corresponding process node, the data record containing the detection content is considered out-of-limit data. For status-based detection content, if the detection content does not belong to the status record set within the preset range of the corresponding process node, the data record containing the detection content is considered out-of-limit data. For curve-based detection content, if the voltage or current record deviates from the preset range of the corresponding process node within the same acquisition stage, the data record containing the voltage or current record is considered out-of-limit data. After the out-of-limit data is determined, the adjacent data before and after the out-of-limit data are read according to the production batch identifier, process node, data category, and acquisition time. Adjacent data before and after the out-of-limit data refers to the preceding and following data records that are immediately adjacent to the out-of-limit data at the acquisition time under the same production batch identifier, the same process node, and the same data category. If the out-of-limit data is located at the beginning of the record set corresponding to the same process node, only the adjacent data after the out-of-limit data is retained. If the out-of-limit data is located at the end of the record set corresponding to the same process node, only the adjacent data before the out-of-limit data will be retained. If the preceding data record or the following data record of the out-of-limit data is a data acquisition breakpoint, the data acquisition breakpoint marker will be retained.

[0032] Out-of-limit neighborhood data is formed by connecting out-of-limit data with its immediate neighbors according to the acquisition time. The connection process is performed within the same production batch identifier, the same process node, and the same data category. The connection order is: adjacent data before the out-of-limit data, the out-of-limit data itself, and adjacent data after the out-of-limit data. If multiple out-of-limit data appear consecutively at the acquisition time, these consecutive out-of-limit data, along with their nearest neighbors before and after, are connected to form a single out-of-limit neighborhood data. If there is an acquisition breakpoint between two out-of-limit data sets, separate out-of-limit neighborhood data sets are formed for each set. The acquisition breakpoint data is retained at the corresponding position to indicate that the connection was interrupted. After the out-of-limit neighborhood data is formed, it stores the production batch identifier, process node, data category, acquisition time, detection content, out-of-limit data marker, adjacent data markers before and after the out-of-limit data, and acquisition breakpoint data marker.

[0033] The distinction between short-term impact periods is based on the data in the out-of-limit neighborhood. Within the out-of-limit neighborhood data, if data deviating from the preset range of the corresponding process node only appears within adjacent acquisition times, and both adjacent data before and after the out-of-limit data return to the preset range of the corresponding process node, then the out-of-limit neighborhood data is classified into the short-term impact period. "Appearing within adjacent acquisition times" means that the data deviating from the preset range of the corresponding process node does not cross the acquisition breakpoint data within the out-of-limit neighborhood data, and the acquisition time interval between the data deviating from the preset range of the corresponding process node does not exceed the allowed acquisition interval range. "Returning to the preset range of the corresponding process node" means that both adjacent data before and after the out-of-limit data fall within the preset range of the corresponding process node, or the state-type detection content re-belongs to the state record set within the preset range of the corresponding process node. If the out-of-limit data is located at the beginning or end of the process node record set, causing adjacent data before and after the out-of-limit data to be unable to be obtained simultaneously, then the out-of-limit neighborhood data is not classified into the short-term impact period, and the pending judgment mark for the out-of-limit neighborhood data is retained.

[0034] The distinction between stable offset segments is also based on the data in the out-of-limit neighborhood. The offset direction is determined according to the positional relationship of the detected content relative to the preset range of the corresponding process node. For numerical detected content, the offset direction above the upper limit of the preset range of the corresponding process node is recorded as an upper offset, and the offset direction below the lower limit of the preset range of the corresponding process node is recorded as a lower offset. For state-type detected content, the offset direction is determined based on the pre-mapping relationship between the abnormal state category outside the state record set and the preset state category of the process node. The pre-mapping relationship is jointly established by the historical qualified records and historical rework records of the same production line. For example, the open circuit state in the conduction state record can be mapped to the upper offset, and the short circuit state in the conduction state record can be mapped to the lower offset. For curve-type detected content, the offset direction is determined based on the positional relationship of the voltage record or current record relative to the preset range of the corresponding process node within the same acquisition phase. If the data deviating from the preset range of the corresponding process node continuously maintains the same offset direction along the acquisition time, and the process of continuously maintaining the same offset direction is not truncated by the acquisition breakpoint data, then the out-of-limit neighborhood data is classified into the stable offset segment. If different offset directions alternate in the out-of-limit neighborhood data, the out-of-limit neighborhood data will not be included in the stable offset segment, and the mixed offset label of the out-of-limit neighborhood data will be retained.

[0035] Based on the short-term impact segment and the stable offset segment, the inflection point of the charge-discharge curve is extracted, and the conduction fluctuation segment, the material state gradual change segment, and the assembly process reversal segment are connected to form a local deviation segment, including: When extracting inflection points of the charge-discharge curve, the charge-discharge curve data is read within the production batch identifier, process node, and acquisition time range corresponding to the short-term impact segment and the stable offset segment. The charge-discharge curve data includes voltage and current records arranged by acquisition time, each corresponding to a detection stage. The voltage and current records are arranged by acquisition time, and the direction of change between adjacent acquisition times is determined. The direction of change refers to the increase or decrease relationship of the detected content between adjacent acquisition times: when the detected content at a later acquisition time is higher than that at a previous acquisition time, it is recorded as an upward direction; when the detected content at a later acquisition time is lower than that at a previous acquisition time, it is recorded as a downward direction; and when the detected content at a later acquisition time is equal to that at a previous acquisition time, it is recorded as a holding direction. The determination of "equal to the previous acquisition time" uses the minimum distinguishable change corresponding to the resolution of the detection instrument as the criterion. For example, if the absolute value of the difference between the detected content at a later acquisition time and the detected content at a previous acquisition time is less than the minimum distinguishable change corresponding to the resolution of the detection instrument, it is recorded as a holding direction.

[0036] Locate the points in the charge / discharge curve data where the direction of change reverses. These points include transitions from an upward to a downward direction, from a downward to an upward direction, from an upward to a holding direction and then back to a downward direction, and from a downward to a holding direction and then back to an upward direction. When an upward direction directly reverses to a downward direction, the point of change is the last acquisition time corresponding to the upward direction. Similarly, when a downward direction directly reverses to an upward direction, the point of change is the last acquisition time corresponding to the downward direction. When an upward direction reverses to a holding direction and then back to a downward direction, the point of change is the middle acquisition time within the acquisition time interval corresponding to the holding direction; if the number of acquisition times corresponding to the holding direction is even, the point of change is the acquisition time closest to the downward direction within the acquisition time interval corresponding to the holding direction. Likewise, when a downward direction reverses to a holding direction and then back to an upward direction, the point of change is the middle acquisition time within the acquisition time interval corresponding to the holding direction; if the number of acquisition times corresponding to the holding direction is even, the point of change is the acquisition time closest to the upward direction within the acquisition time interval corresponding to the holding direction. The moment when the direction of change changes is captured is taken as the moment when the charging and discharging curve inflection point is captured.

[0037] If both voltage and current records from the same acquisition moment show a change in direction, inflection points for the charge / discharge curves corresponding to the voltage and current records are established respectively, and these two types of inflection points are distinguished by the type of detection value. If the location of the change in direction crosses the acquisition breakpoint data, no inflection point is formed. If the location of the change in direction does not belong to the acquisition time range corresponding to the short-time impact segment and the stable offset segment, the location of the change in direction is not written into the inflection point of the charge / discharge curve.

[0038] When extracting the continuity fluctuation segment, the continuity data is read within the production batch identifier, process node, and acquisition time range corresponding to the short-term impact segment and stable offset segment. The continuity data includes continuity resistance records, continuity voltage drop records, and continuity status records. The continuity resistance records and continuity voltage drop records are arranged according to the acquisition time, and the direction of change is determined according to the increase or decrease relationship of the detection content at adjacent acquisition times. A continuous reciprocating change interval refers to the data interval where, under the same production batch identifier, the same process node, and the same data category, the continuity resistance record or continuity voltage drop record continuously alternates between rising and falling directions along the acquisition time, or continuously alternates between falling and rising directions, and the alternation process is not truncated by the acquisition breakpoint data. When the continuity status record is a status-type detection content, the continuity status record is converted to an upper offset, a lower offset, or a return to the preset range of the corresponding process node according to the offset direction, and the acquisition time range where the upper and lower offsets alternate is taken as the continuous reciprocating change interval.

[0039] The continuous reciprocating change interval is formed into a connected conduction fluctuation segment. The data record for the connected conduction fluctuation segment includes the production batch identifier, process node, start acquisition time, end acquisition time, data category, connected conduction data participating in forming the connected conduction fluctuation segment, change direction arrangement result, and corresponding short-term impact segment or corresponding stable offset segment. The start acquisition time is the acquisition time before the first change direction alternation position in the continuous reciprocating change interval, and the end acquisition time is the acquisition time after the last change direction alternation position in the continuous reciprocating change interval. If the continuous reciprocating change interval is located at the beginning of the record set, the start acquisition time is the earliest acquisition time in the continuous reciprocating change interval. If the continuous reciprocating change interval is located at the end of the record set, the end acquisition time is the latest acquisition time in the continuous reciprocating change interval. If the continuous reciprocating change interval contains only one unidirectional change, a connected conduction fluctuation segment is not formed. If the continuous reciprocating change interval contains acquisition breakpoint data, the continuous reciprocating change intervals before and after the acquisition breakpoint data are judged separately, and a connected conduction fluctuation segment is formed without crossing the acquisition breakpoint data.

[0040] When extracting the material state gradual change segment, the production batch identifier, process node, and material state data within the acquisition time range corresponding to the stable offset segment are read. Material state data includes records of the storage environment temperature, humidity, particle state, and appearance of the materials used in production. A continuous unidirectional change interval refers to a data range within the stable offset segment where the material state data continuously maintains an upward, downward, upper, or lower offset direction along the acquisition time. When the storage environment temperature and humidity records are numerical detection content, the upward or downward direction is determined according to the increase or decrease relationship of the detection content between adjacent acquisition times. When the particle state and appearance records are state-type detection content, the upper or lower offset is determined according to the pre-mapping relationship between the abnormal state category and the offset direction.

[0041] A continuous unidirectional change interval forms a material state gradual change segment. Data records for this segment include the production batch identifier, process node, start and end acquisition times, data category, material state data contributing to the segment, continuous change direction, and corresponding stable offset segment. The start acquisition time is when the material state data begins to continuously maintain the same change or offset direction; the end acquisition time is when the data stops continuously maintaining the same change or offset direction. If the material state data experiences an interruption in change direction, reversal of offset direction, or acquisition breakpoint within the stable offset segment, the breakpoint is used as a dividing point to determine whether the data intervals before and after the breakpoint can form a material state gradual change segment. If the material state data deviates only within a short-term impact segment, a material state gradual change segment is not formed. If the material state data within the stable offset segment contains a missing detection marker, the continuous unidirectional change interval is interrupted at the missing detection marker.

[0042] When extracting the reversal segment of the assembly process, assembly process data is read within the production batch identifier, process node, and acquisition time range corresponding to the short-term impact segment and the stable offset segment. Assembly process data includes assembly pressure records, assembly displacement records, assembly temperature records, assembly speed records, and work status records. Assembly pressure records, assembly displacement records, assembly temperature records, and assembly speed records are arranged according to their acquisition time, and the direction of change between adjacent acquisition times is determined. The interval where the process change direction changes refers to the acquisition time interval where the change direction of the same data category in the assembly process data changes from an upward direction to a downward direction, from a downward direction to an upward direction, from an upward direction to a holding direction and then back to a downward direction, or from a downward direction to a holding direction and then back to an upward direction. When the work status record is a status-type detection content, the interval where the process change direction changes is determined according to the status change of the work status record relative to the preset range of the corresponding process node.

[0043] The interval where the process direction changes is transformed forms the assembly process reversal segment. Data records for the assembly process reversal segment include production batch identifier, process node, start acquisition time, end acquisition time, data category, assembly process data participating in forming the reversal segment, the direction of change before the transformation, and the direction of change after the transformation. The start acquisition time is the last acquisition time before the process direction change occurs, and the end acquisition time is the first acquisition time after the process direction change occurs. If the process direction changes from an upward direction to a holding direction and then to a downward direction, or from a downward direction to a holding direction and then to an upward direction, the start acquisition time is the last acquisition time before the holding direction begins, and the end acquisition time is the first acquisition time after the holding direction ends. If the assembly speed record in the assembly process data is calculated from the assembly displacement record and the acquisition time, the assembly speed record is read, and the assembly speed record is not recalculated. If the interval where the process direction changes spans the acquisition breakpoint data, an assembly process reversal segment is not formed, but the record position corresponding to the acquisition breakpoint data is retained.

[0044] After establishing the inflection points of the charge / discharge curves, connecting conduction fluctuation segments, material state gradual change segments, and assembly process reversal segments, data is aggregated according to production batch identifiers, process nodes, and acquisition times. During the aggregation process, the production batch identifier for each charge / discharge curve inflection point, each connecting conduction fluctuation segment, each material state gradual change segment, and each assembly process reversal segment is read. Data records with the same production batch identifier are included in the same aggregation range. Process nodes are read, and data records belonging to the same or adjacent process nodes within the same aggregation range are arranged within the same local aggregation range. Acquisition times are then read, and data records with the same, adjacent, or overlapping acquisition times within the same local aggregation range are connected. Adjacent acquisition times are determined by the allowable acquisition interval range; if the acquisition time interval between two data records does not exceed the allowable range, the two data records are considered adjacent. If there is a data breakpoint between two data records, the two data records are not connected.

[0045] The collected data records are formed into local deviation segments. A local deviation segment includes a production batch identifier, the involved process nodes, the start and end collection times, the included charge / discharge curve inflection points, the included connection / conduction fluctuation segments, the included material state gradual change segments, the included assembly process reversal segments, and the segment source record. The start collection time is the collection time corresponding to the earliest data record appearing in the local deviation segment, and the end collection time is the collection time corresponding to the latest data record appearing in the local deviation segment. The segment source record is used to store data records of one or more types from the charge / discharge curve inflection points, connection / conduction fluctuation segments, material state gradual change segments, and assembly process reversal segments that form the local deviation segment. If the collection time intervals of different process nodes under the same production batch identifier are not adjacent and do not overlap, they are formed into different local deviation segments. If charge / discharge curve inflection points, connection / conduction fluctuation segments, material state gradual change segments, and assembly process reversal segments exist simultaneously within the same local collection range under the same production batch identifier, these elements are all written into the same local deviation segment.

[0046] Local deviation segments are matched with the rework locations, rework times, and re-inspection results in historical fault rework records to form deviation risk segments, including: The production batch identifier is read from the local deviation segment. Then, the process node, start acquisition time, end acquisition time, and segment source record involved in the local deviation segment are read. The occurrence location of the local deviation segment is determined according to the involved process node and the acquisition time range from the start acquisition time to the end acquisition time. The occurrence location of the local deviation segment refers to the data ownership location of the local deviation segment in the electronic ignition module production process. When the local deviation segment only contains the inflection point of the charge-discharge curve, the occurrence location of the local deviation segment is determined by the process node to which the inflection point of the charge-discharge curve belongs and the acquisition time to which the inflection point of the charge-discharge curve belongs. When the local deviation segment includes the connection conduction fluctuation segment, the material state gradual change segment, or the assembly process reversal segment, the occurrence location of the local deviation segment is determined by the process node to which the connection conduction fluctuation segment, the material state gradual change segment, or the assembly process reversal segment belongs, and the acquisition time range from the start acquisition time to the end acquisition time.

[0047] Historical fault rework records refer to rework processing records generated and stored during the production process of electronic ignition modules. These records include historical production batch identifiers, rework locations, rework occurrence times, rework reason records, number of rework attempts, and re-inspection results. Rework locations refer to the production positions where rework occurred in the historical fault rework records. Rework locations are recorded using production position names consistent with process nodes, including electronic control module detection positions, conductive connection positions, material input entry positions, assembly operation positions, and ignition response verification positions. Rework reason records are records of data anomaly types corresponding to rework processing, with a correspondence established between rework reason records and data categories. Re-inspection results refer to the record of results generated from re-inspection after rework processing, including qualified records, unqualified records, and records that cannot be determined. The number of rework attempts refers to the number of rework records in the historical fault rework records where the same rework location, the same rework reason record, and the same data category are simultaneously present. When writing historical fault rework records, the corresponding process node of the rework part, the time of rework, the data category of the rework reason record, and the re-inspection results are saved simultaneously, so that historical fault rework records can establish a correspondence with local deviation segments.

[0048] Local deviation segments are read according to process nodes and acquisition times, and the inflection points of the charge / discharge curves, the connection conduction fluctuation segments, the material state gradual change segments, and the assembly process reversal segments within these segments are correlated with the rework locations in historical fault rework records. When establishing these correlations, the process node and acquisition time of the charge / discharge curve inflection point are read, and then records in historical fault rework records are searched that match the process node of the rework location. Similarly, the process node, start acquisition time, and end acquisition time of the connection conduction fluctuation segment are read, and records in historical fault rework records are searched that match the process node of the connection conduction fluctuation segment. Finally, the process node, start acquisition time, and end acquisition time of the material state gradual change segment are read, and records in historical fault rework records are searched that match the process node of the rework location. Read the process node to which the assembly process reversal section belongs, the start time of the assembly process reversal section, and the end time of the assembly process reversal section. Then, search for records in the historical fault rework records where the process node corresponding to the rework part is consistent with the process node to which the assembly process reversal section belongs.

[0049] When establishing the correspondence, the adjacent process nodes of the rework location are simultaneously read. Adjacent process nodes of the rework location refer to the process nodes that are immediately before or after the corresponding process node of the rework location within the same production batch of the electronic ignition module. The method for determining adjacent process nodes of the rework location is as follows: a process node sequence record is formed according to the production sequence of the electronic control module detection position, conductive connection position, material input position, assembly operation position, and ignition response verification position. Then, the position of the corresponding process node of the rework location in the process node sequence record is read. The process node one position before and one position after the corresponding process node of the rework location are both considered adjacent process nodes of the rework location. For example, when the corresponding process node of the rework location is the assembly operation position, the material input position and the ignition response verification position can be considered adjacent process nodes of the rework location. When the corresponding process node of the rework location is located at the beginning of the process node sequence record, the adjacent process nodes of the rework location only include the process node one position after the corresponding process node of the rework location. When the process node corresponding to the reworked part is located at the end of the process node sequence record, the adjacent process nodes of the reworked part only include the process node one position before the process node corresponding to the reworked part.

[0050] Based on the process node and data acquisition time, a correspondence is established between the location of local deviation segments and the rework location in the historical fault rework record. When the location of the local deviation segment is at the process node corresponding to the rework location, a direct correspondence is established between the local deviation segment and the corresponding historical fault rework record. When the location of the local deviation segment is at a process node adjacent to the rework location, an adjacent correspondence is established between the local deviation segment and the corresponding historical fault rework record. Both direct and adjacent correspondences retain the production batch identifier, involved process nodes, start acquisition time, end acquisition time, segment source record, historical production batch identifier, rework location, rework occurrence time, rework reason record, number of reworks, and re-inspection results of the local deviation segment. When the location of the local deviation segment is neither at the process node corresponding to the rework location nor at a process node adjacent to the rework location, no correspondence is established between the local deviation segment and the historical fault rework record.

[0051] The rework count and re-inspection results are read based on the corresponding relationships. The rework count is read from historical fault rework records that have established a direct or adjacent correspondence with local deviation segments. Historical fault rework records are merged according to rework location, rework reason record, and data category, and the number of rework records under the same rework location, same rework reason record, and same data category is counted to form the rework count. Duplicate rework counts refer to rework counts that are greater than or equal to the preset repetition count. The preset repetition count is determined by the record distribution of historical fault rework records before the production task begins. The determination method is to read historical fault rework records of the same production line within the preset historical statistical period, take the number of rework records corresponding to the same rework reason record and the same data category for each rework location within the preset historical statistical period as the statistical object, arrange all statistical objects in ascending order of value, remove statistical objects with a value of 0, and then select the number of records located on the upper boundary of the concentrated distribution area among the remaining statistical objects as the preset repetition count. The preset historical statistical period is determined by the length of traceable records prior to the start of the production task. For example, the preset historical statistical period can be set to the record period corresponding to several most recently completed production batches. The upper boundary of the concentrated distribution area is determined by removing a small number of the highest-ranking statistical objects from the remaining statistical objects after sorting, and using the maximum value among the retained statistical objects as the upper boundary of the concentrated distribution area. For example, records of the same rework location, the same rework reason, and the same data category appearing more than twice can be used as a way to determine if there are duplicate rework records.

[0052] When reading re-inspection results, historical fault rework records that have established a direct or adjacent correspondence with the local deviation segment are used as the reading objects. The presence of non-conforming records in the re-inspection results means that there is at least one data record with a non-conforming re-inspection result under the corresponding relationship. When the re-inspection result is a conforming record, the re-inspection result is included in the conforming record set. When the re-inspection result is an undeterminable record, the re-inspection result is included in the undeterminable record set, and the undeterminable record is not treated as a non-conforming record. If conforming records, non-conforming records, and undeterminable records exist simultaneously under the same correspondence, the existence of non-conforming records is used as the criterion for determining whether a non-conforming record exists in the re-inspection result, and the original records of conforming records, non-conforming records, and undeterminable records are retained.

[0053] Local deviation segments with consistent rework locations, duplicate rework records, and unqualified re-inspection results are marked as deviation risk segments. Consistent rework locations mean the local deviation segment occurs at the corresponding process node of the rework location, or occurs at an adjacent process node of the rework location, and the data category corresponding to the segment's source record matches the data category corresponding to the rework cause record. Consistent data categories mean the charging / discharging curve inflection point in the segment's source record corresponds to the charging / discharging related cause in the rework cause record; the connection / conduction fluctuation segment in the segment's source record corresponds to the conduction related cause in the rework cause record; the material state gradual change segment in the segment's source record corresponds to the material state related cause in the rework cause record; and the assembly process reversal segment in the segment's source record corresponds to the assembly operation related cause in the rework cause record. The matching rules for consistent data categories are established before the production task begins based on the rework cause records in the historical fault rework records and remain unchanged during the same production task. When establishing matching rules for data categories, the text records or category records in the rework reason records are read, and the text records or category records are categorized according to their position into charging / discharging related reasons, conduction related reasons, material condition related reasons, and assembly operation related reasons. Rework reason records that cannot be categorized into the aforementioned reason categories are not included in the data category consistency judgment.

[0054] Each local deviation segment is individually marked as a deviation risk segment. If a local deviation segment simultaneously meets the criteria of consistent rework location, duplicate rework counts, and non-conforming re-inspection results, it is classified as a deviation risk segment. The deviation risk segment retains the production batch identifier, involved process nodes, start and end acquisition times, included charge / discharge curve inflection points, included connection continuity fluctuation segments, included material state gradual change segments, included assembly process reversal segments, segment source record, correspondence, rework count, and re-inspection results. If a local deviation segment does not simultaneously meet the criteria of consistent rework location, duplicate rework counts, and non-conforming re-inspection results, it is not classified as a deviation risk segment, and the original record of the local deviation segment is retained. If the same local deviation segment corresponds to multiple historical fault rework records, the rework count and re-inspection results are read from each of the multiple historical fault rework records. If one historical fault rework record among multiple historical fault rework records meets the following criteria: consistent rework location, duplicate rework times, and unqualified re-inspection results, the same local deviation segment will be classified into the deviation risk segment, and the historical fault rework record that meets the conditions will be written into the corresponding relationship.

[0055] Based on the sequential transmission direction, continuous workstation span, and consistency of terminal detection offset of deviation risk segments among electronic control modules, material states, and assembly processes, batch coupling risk results are generated, including: When reading deviation risk segments, deviation risk segments belonging to the same electronic ignition module and the same production batch are grouped into the same batch risk set according to the production batch identifier. Within the same batch risk set, deviation risk segments are then arranged according to process nodes and acquisition times. The process node arrangement follows the production sequence: electronic control module detection position, conductive connection position, material input position, assembly operation position, and ignition response verification position. The acquisition times are arranged according to the time sequence recorded by a unified clock source. When a deviation risk segment has a start acquisition time and an end acquisition time, the start acquisition time is used as the starting point and the end acquisition time as the ending point. If multiple deviation risk segments exist within the same process node, they are arranged from earliest to latest according to their start acquisition time. If multiple deviation risk segments have the same start acquisition time, they are arranged from earliest to latest according to their end acquisition time. If both the start and end acquisition times of multiple deviation risk segments are the same, they are saved side-by-side according to the recording order of their source in the charging / discharging curve inflection point, connection conduction fluctuation segment, material state gradual change segment, and assembly process reversal segment.

[0056] When identifying the order in which deviation risk segments appear among the electronic control module, material state, and assembly process, the category of the deviation risk segment is determined according to the segment source record. If the segment source record contains a charge / discharge curve inflection point, the deviation risk segment is classified as belonging to the electronic control module. If the segment source record contains a material state gradual change segment, the deviation risk segment is classified as belonging to the material state. If the segment source record contains an assembly process reversal segment, the deviation risk segment is classified as belonging to the assembly process. If the segment source record only contains a connection conduction fluctuation segment, the connection conduction fluctuation segment is classified as belonging to the assembly process according to the proximity of the corresponding conductive connection position in the production sequence, and the connection conduction fluctuation segment is retained in the deviation risk segment list. If the segment source record contains both a connection conduction fluctuation segment and an assembly process reversal segment, the deviation risk segment is classified as belonging to the assembly process, and both the connection conduction fluctuation segment and the assembly process reversal segment are retained in the deviation risk segment list. When the fragment source record contains at least two types of records from the inflection point of the charge / discharge curve, the gradual change segment of the material state, and the reversal segment of the assembly process, corresponding attribution category records are formed respectively, and the production batch identifier, the process node involved, the start time of collection, and the end time of collection of the same deviation risk fragment are synchronously written into each attribution category record.

[0057] When establishing the sequential transmission direction, the starting acquisition time of each category record within the same batch of risk sets is read, and the order of appearance of electronic control modules, material states, and assembly processes is arranged from earliest to latest according to the starting acquisition time. The sequential transmission direction refers to the category flow relationship formed between the electronic control module, material state, and assembly process according to the order of acquisition time. When the deviation risk segment corresponding to the electronic control module appears first, followed by the deviation risk segment corresponding to the material state, and lastly the deviation risk segment corresponding to the assembly process, the sequential transmission direction is recorded as electronic control module to material state to assembly process. When the deviation risk segment corresponding to the assembly process appears first, followed by the deviation risk segment corresponding to the electronic control module, the sequential transmission direction is recorded as assembly process to electronic control module. When the starting acquisition time of two category records is the same, the sequential position of the two category records is determined according to the process node order. When the starting acquisition time of two category records is the same and the process node is the same, the two category records are written into the parallel appearance record, without forcibly determining the sequential position between the two category records. When there are records appearing in parallel, the category and collection time corresponding to the records appearing in parallel are retained in the sequential transmission direction.

[0058] The continuous workstation span is determined based on the consecutive process nodes spanned by the deviation risk segment. The continuous workstation span refers to the range of process nodes continuously covered in the process node sequence record for deviation risk segments within the same batch of risk sets that have adjacent acquisition times or overlapping acquisition time intervals. Adjacent acquisition times are determined using the allowable acquisition interval range. If the time interval between two deviation risk segments does not exceed the allowable acquisition interval range, the two deviation risk segments are considered to have adjacent acquisition times. If the time interval between two deviation risk segments exceeds the allowable acquisition interval range, the two deviation risk segments are not considered to have adjacent acquisition times. Overlapping acquisition time intervals refer to the intersection of the interval between the start and end acquisition times of one deviation risk segment and the interval between the start and end acquisition times of another deviation risk segment. The process nodes involved in each deviation risk segment are read from the process node sequence record. The process nodes covered by deviation risk segments with adjacent acquisition times or overlapping acquisition time intervals are connected as consecutive process nodes, and the start and end process nodes of the consecutive process nodes are written into the continuous workstation span.

[0059] When determining the continuous workstation span, if a deviation risk segment covers only one process node, the continuous workstation span is recorded as a single process node span. If there are intermediate process nodes not covered by the deviation risk segment among the process nodes covered by the deviation risk segment, and these intermediate process nodes have data acquisition breakpoints or missing detection content markers, the intermediate process nodes are marked with a span interruption marker, and multiple deviation risk segments spanning these intermediate process nodes are not connected into the same continuous workstation span. If there are intermediate process nodes not covered by the deviation risk segment among the process nodes covered by the deviation risk segment, and these intermediate process nodes do not have data acquisition breakpoints or missing detection content markers, the multiple deviation risk segments are formed into different continuous workstation spans. When multiple deviation risk segments appear repeatedly at the same process node, the continuous workstation span retains records of the number of repetitions, the corresponding acquisition time for the repetition, and the source of the segment corresponding to the repetition, but does not merge multiple deviation risk segments.

[0060] When reading the offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node, the same production batch of the same electronic ignition module is located in the batch quality data according to the production batch identifier. Then, the ignition response detection data is located according to the process node and acquisition time corresponding to the ignition response verification position. When the deviation risk segment has a start acquisition time and an end acquisition time, the ignition response detection data with an adjacent acquisition time relationship or overlapping acquisition time interval with the start acquisition time to end acquisition time of the deviation risk segment is read first. If the ignition response detection data and the deviation risk segment do not have an adjacent acquisition time relationship or overlapping acquisition time interval, the ignition response detection data with the acquisition time closest to the end acquisition time of the deviation risk segment under the same production batch identifier is read, and the ignition response detection data interval marker is written into the reading result. The ignition response detection data includes the verification start record, response occurrence record, response duration record, and response status record. The preset range of the corresponding process node is the preset range of the ignition response detection data.

[0061] When determining the offset direction of ignition response detection data, for numerical detection content in the ignition response detection data, if the detection content is higher than the upper limit of the preset range of the corresponding process node, the offset direction of the ignition response detection data is recorded as an upward offset; if the detection content is lower than the lower limit of the preset range of the corresponding process node, the offset direction of the ignition response detection data is recorded as a downward offset. For status-type detection content in the ignition response detection data, the offset direction of the ignition response detection data is determined according to the pre-mapping relationship between the abnormal status category and the offset direction. When the ignition response detection data falls within the preset range of the corresponding process node, the ignition response detection data is written to the no-offset record. If the same deviation risk segment corresponds to multiple ignition response detection data, and the multiple ignition response detection data form the same offset direction, the same offset direction is taken as the offset direction of the ignition response detection data. If the same deviation risk segment corresponds to multiple ignition response detection data, and the multiple ignition response detection data form different offset directions, the offset direction of the ignition response detection data is determined according to the ignition response detection data whose acquisition time is closest to the end acquisition time of the deviation risk segment, and the original offset direction records of multiple ignition response detection data are retained.

[0062] When reading the offset direction of a deviation risk segment in the corresponding process node, the short-term impact segment marker, stable offset segment marker, limit-crossing data marker, and segment source record retained in the deviation risk segment are read, and the upper or lower offset is read. When the deviation risk segment originates from the inflection point of the charge / discharge curve, the offset direction of the deviation risk segment in the corresponding process node is determined by the positional relationship between the voltage record or current record corresponding to the charge / discharge curve data and the preset range of the corresponding process node. When the deviation risk segment originates from the connection conduction fluctuation segment, the offset direction of the deviation risk segment in the corresponding process node is determined by the positional relationship between the conduction resistance record, conduction voltage drop record, or conduction state record in the connection conduction data and the preset range of the corresponding process node. When the deviation risk segment originates from the material state gradual change segment, the offset direction of the deviation risk segment in the corresponding process node is determined by the positional relationship between the storage environment temperature record, storage environment humidity record, particle state record, or appearance state record of the input material in the material state data and the preset range of the corresponding process node. When a deviation risk segment originates from the reversing section of the assembly process, the offset direction of the deviation risk segment in the corresponding process node is determined by the positional relationship of the assembly pressure record, assembly displacement record, assembly temperature record, assembly speed record, or operation status record in the assembly process data relative to the preset range of the corresponding process node. If a deviation risk segment originates from multiple segment source records and all multiple segment source records correspond to the same offset direction, this same offset direction is used as the offset direction of the deviation risk segment in the corresponding process node. If a deviation risk segment originates from multiple segment source records and all multiple segment source records correspond to different offset directions, the offset direction of the deviation risk segment in the corresponding process node is determined according to the segment source record whose acquisition time is closest to the end acquisition time of the deviation risk segment, and the original offset direction records of the multiple segment source records are retained.

[0063] The offset direction of the ignition response detection data is compared with the offset direction of the deviation risk segment in the corresponding process node to form the terminal detection offset consistency. Terminal detection offset consistency refers to whether the offset direction of the ignition response detection data relative to the preset range of the corresponding process node is on the same side as the offset direction of the deviation risk segment in the corresponding process node. A consistency mark is formed when both the offset direction of the ignition response detection data and the offset direction of the deviation risk segment in the corresponding process node are upward offsets. A consistency mark is formed when both the offset direction of the ignition response detection data and the offset direction of the deviation risk segment in the corresponding process node are downward offsets. An inconsistency mark is formed when the offset direction of the ignition response detection data is upward and the offset direction of the deviation risk segment in the corresponding process node is downward. An inconsistency mark is formed when the offset direction of the ignition response detection data is downward and the offset direction of the deviation risk segment in the corresponding process node is upward. When there is no offset record for the ignition response detection data, a terminal no offset mark is formed. When the offset direction of the deviation risk segment in the corresponding process node cannot be determined, an offset direction pending determination mark is formed. When there are interval markers in the ignition response detection data, the interval markers are retained simultaneously in addition to the consistency marker, inconsistency marker, terminal no offset marker, or offset direction pending determination marker.

[0064] When establishing terminal detection offset consistency based on consistent and inconsistent markers, consistent markers, inconsistent markers, terminal no-offset markers, and offset direction pending determination markers are statistically analyzed within the same batch of risk sets. Terminal detection offset consistency is formed according to the marker composition. If a consistent marker exists within the same batch of risk sets and no inconsistent markers exist, the terminal detection offset consistency is recorded as direction consistent. If an inconsistent marker exists within the same batch of risk sets and no consistent marker exists, the terminal detection offset consistency is recorded as direction inconsistent. If both consistent and inconsistent markers exist within the same batch of risk sets, the terminal detection offset consistency is recorded as direction mixed. If only terminal no-offset markers exist within the same batch of risk sets, the terminal detection offset consistency is recorded as terminal no offset. If only offset direction pending determination markers exist within the same batch of risk sets, the terminal detection offset consistency is recorded as pending determination. If both terminal no-offset markers and offset direction pending determination markers exist within the same batch of risk sets, and neither consistent nor inconsistent markers exist, the terminal detection offset consistency is recorded as pending review. When ignition response detection data interval markers exist within the same batch of risk sets, the ignition response detection data interval markers are synchronously written into the terminal detection offset consistency.

[0065] Batch coupling risk results are generated based on the sequential transmission direction, continuous workstation span, and terminal detection offset consistency. Batch coupling risk results refer to the record of whether there is a cross-process continuous association between the deviation risk segment and the electronic control module, material status, and assembly process within the same production batch of the same electronic ignition module. When generating batch coupling risk results, the number of attribution categories in the sequential transmission direction is read first, then the number of continuous process nodes covered by the continuous workstation span is read, and finally the terminal detection offset consistency is read. The number of attribution categories is the number of categories in the electronic control module, material status, and assembly process where deviation risk segments have already appeared. The number of continuous process nodes is the number of process nodes covered from the starting process node to the ending process node in the continuous workstation span. When the number of attribution categories is not less than 2, the continuous workstation span covers at least 2 continuous process nodes, and the terminal detection offset consistency is consistent in direction or mixed in direction, the batch coupling risk result is recorded as having cross-process coupling risk. When the number of attribution categories is 1 or the continuous workstation span covers a single process node, the batch coupling risk result is recorded as a single-process risk. When the terminal detection offset consistency is inconsistent in direction, has no terminal offset, is pending judgment, or is pending review, the batch coupling risk result is recorded as a risk requiring review. The batch coupling risk result retains the production batch identifier, sequential transmission direction, continuous workstation span, terminal detection offset consistency, consistency mark, inconsistency mark, no terminal offset mark, offset direction pending judgment mark, ignition response detection data interval mark, and the corresponding deviation risk segment.

[0066] Based on the batch coupling risk results, the re-mining and calibration processes are determined. After executing the calibration process, batch quality data is generated again, including: The system reads the production batch identifier and locates the same production batch of the same electronic ignition module according to the batch identifier; then it reads the sequential transmission direction, continuous workstation span, and terminal detection offset consistency; and reads the relevant process nodes, start acquisition time, end acquisition time, and segment source record in the corresponding deviation risk segment. Each data item in the batch coupling risk result retains its correspondence with the deviation risk segment, without changing the original record content of the deviation risk segment, and only determines the production positions that need to be re-acquired and corrected based on the batch coupling risk result.

[0067] The re-collection process refers to the range of process nodes within the same production batch of the same electronic ignition module that require the re-collection of quality-related data according to the process node sequence. The re-collection process includes the starting process node, the ending process node, and the continuous process nodes between the starting and ending process nodes. The starting process node is determined based on the sequential transfer direction. When the first category in the sequential transfer direction is the electronic control module, the detection position of the electronic control module is determined as the starting process node. When the first category in the sequential transfer direction is the material state, the entry position of the input material is determined as the starting process node. When the first category in the sequential transfer direction is the assembly process, the assembly operation position is determined as the starting process node. When there are parallel records in the sequential transmission direction, the corresponding category and process node of the parallel records are read, and the earliest process node in the parallel records is determined as the starting process node of the re-mining process according to the production sequence of electronic control module detection position, conductive connection position, input production material entry position, assembly operation position and ignition response verification position.

[0068] The end node of the remining process is determined based on the continuous workstation span. When the continuous workstation span record is a single workstation span, the workstation corresponding to the single workstation span is determined as the end node of the remining process. When the continuous workstation span record is a range of consecutive workstations, the end node within the range of consecutive workstations is read and determined as the end node of the remining process. When there is a span interruption marker in the continuous workstation span, the last consecutive workstation before the span interruption marker is determined as the end node of the remining process, and the span interruption marker is written into the remining process record. When there are repeated occurrences in the continuous workstation span, the number of repeated occurrences, the corresponding acquisition time, and the corresponding fragment source record are written into the remining process record. When the starting node of the remining process is later than the ending node of the remining process, the starting node and ending node of the remining process are swapped according to the workstation sequence, and an order adjustment marker is written into the remining process record.

[0069] After determining the starting and ending process nodes of the re-mining process, the re-mining process node range is generated according to the production sequence of electronic control module detection position, conductive connection position, material input position, assembly operation position, and ignition response verification position. The re-mining process node range includes the starting and ending process nodes of the re-mining process, as well as the continuous process nodes between them. If the re-mining process node range does not cover the ignition response verification position, the ignition response verification position is written as a verification acquisition process node into the re-mining process record to re-acquire ignition response detection data. If the re-mining process node range already covers the ignition response verification position, it is not written again. The re-mining process record includes at least the production batch identifier, the starting process node of the re-mining process, the ending process node of the re-mining process, the verification acquisition process node, the sequential transfer direction, the continuous workstation span, and the corresponding deviation risk segment.

[0070] A calibration process refers to a process node within the scope of a re-collection process that requires quality confirmation, data verification, or process status restoration. Quality confirmation involves rereading the inspection content for the data category corresponding to the calibration process and comparing the reread inspection content with the preset range of the corresponding process node. Data verification involves re-establishing the correspondence between the original records, acquisition time, data category, and source location records corresponding to the calibration process. Process status restoration involves restoring the inspection conditions, material status records, or assembly process records corresponding to the calibration process to the allowable record range recorded before the start of the production task. The allowable record range is determined by the preset range of the corresponding process node or stable records for the same production task.

[0071] Based on the consistency of terminal detection offset, a correction process is determined within the re-mining process. When the terminal detection offset consistency record shows consistent direction, the deviation risk segment forming the consistency mark is read, and the process node corresponding to the deviation risk segment forming the consistency mark is determined as the correction process. When the terminal detection offset consistency record shows mixed direction, deviation risk segments forming both consistency and inconsistency marks are read, and the process nodes corresponding to both the deviation risk segments forming the consistency mark and the deviation risk segments forming the inconsistency mark are jointly determined as the correction process. When the terminal detection offset consistency record shows inconsistent direction, the process node corresponding to the deviation risk segment forming the inconsistency mark is determined as the correction process, and the ignition response verification position is written into the verification object of the correction process. When the terminal detection offset consistency record shows no terminal offset, the process node corresponding to the deviation risk segment is determined as the correction process, but the ignition response verification position is not determined as the correction process. When the terminal detection offset consistency record shows pending judgment or pending verification, the process node corresponding to the deviation risk segment with the offset direction pending judgment mark or the ignition response detection data interval mark is determined as the correction process, and the ignition response verification position is written into the verification object of the correction process.

[0072] When determining the calibration process, if the deviation risk segment originates from the inflection point of the charge / discharge curve, the detection location of the electronic control module is designated as the calibration process, and the calibration process record includes the charge / discharge curve data requiring reconfirmation, the detection stage, the detection value type, and the acquisition time range. If the deviation risk segment originates from the connection conduction fluctuation segment, the conductive connection location is designated as the calibration process, and the calibration process record includes the connection conduction data requiring reconfirmation, conduction resistance record, conduction voltage drop record, conduction status record, and acquisition time range. If the deviation risk segment originates from the material state gradual change segment, the entry location of the input production material is designated as the calibration process, and the calibration process record includes the material state data requiring reconfirmation, the batch number of the input production material, the storage environment temperature record, the storage environment humidity record, the particle state record, the appearance state record, and the acquisition time range. If the deviation risk segment originates from the assembly process reversal segment, the assembly operation location is designated as the calibration process, and the calibration process record includes the assembly process data requiring reconfirmation, the assembly pressure record, the assembly displacement record, the assembly temperature record, the assembly speed record, the operation status record, and the acquisition time range. When the same deviation risk segment originates from multiple segment source records, multiple correction procedures are determined according to the multiple segment source records, and the correspondence between the multiple segment source records is retained in the correction procedure record.

[0073] During the calibration process, the corresponding process nodes are checked for quality verification, data review, or process status restoration according to the calibration process record. When the electronic control module detection location is used as a calibration process, no-load and continuous readings are re-executed to regenerate charge / discharge curve data, and the correlation between the acquisition times of the charge / discharge curve data before and after calibration is maintained. When the conductive connection location is used as a calibration process, the on-resistance record, on-voltage drop record, and on-state record are re-read, and the correlation between the acquisition times of the connection on-state data before and after calibration is maintained. When the material input location is used as a calibration process, the batch number record, storage environment temperature record, storage environment humidity record, particle state record, and appearance state record of the input materials are re-read, and the correlation between the acquisition times of the material state data before and after calibration is maintained. When the assembly operation location is used as a calibration process, the assembly pressure record, assembly displacement record, assembly temperature record, assembly speed record, and operation status record are re-read, and the correlation between the acquisition times of the assembly process data before and after calibration is maintained. When the ignition response verification location is used as the verification object of the calibration process, the verification start record, response occurrence record, response duration record, and response status record are reread.

[0074] After the calibration process, charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response detection data are re-acquired during the re-acquisition process. When re-acquiring charge / discharge curve data, voltage and current records are read at the electronic control module detection location, and the data is saved according to the detection stage, detection value type, and acquisition time. When re-acquiring connection continuity data, conduction resistance, conduction voltage drop, and conduction status records are read at the conductive connection location, and the connection continuity data is saved according to the acquisition time. When re-acquiring material state data, the batch number, ambient temperature, humidity, particle state, and appearance of the input materials are read at the material entry location, and the material state data is saved according to the acquisition time. When re-acquiring assembly process data, assembly pressure, assembly displacement, assembly temperature, assembly speed, and operation status records are read at the assembly operation location, and the assembly process data is saved according to the acquisition time. When re-collecting ignition response detection data, the ignition response verification start record, response occurrence record, response duration record, and response status record are read at the ignition response verification location, and the ignition response detection data is saved according to the collection time. If the scope of the re-collected process node does not cover the process node corresponding to a certain data category, the original record of the same data category under the same production batch identifier in the batch quality data is retained, and a non-re-collected mark is written into the regenerated batch quality data.

[0075] Establish a correspondence between the re-acquired data and the production batch identifier, process node, and acquisition time. Reuse the production batch identifier and add a re-acquisition batch mark to the re-acquired data. The re-acquisition batch mark distinguishes the batch quality data from the regenerated batch quality data; it does not replace the production batch identifier. Write the re-acquired charge / discharge curve data, connection continuity data, material state data, assembly process data, and ignition response test data into the re-acquisition data record set. Each data entry in the re-acquisition data record set retains the production batch identifier, re-acquisition batch mark, process node, acquisition time, data category, source location record, and test content. If multiple data entries of different data categories are written at the same acquisition time, each entry is retained separately. If re-acquisition fails at the same process node, a re-acquisition failure mark is written to the re-acquisition data record set, while retaining the production batch identifier, re-acquisition batch mark, process node, acquisition time, and data category.

[0076] When generating regenerated batch quality data, the production batch identifier is used as the affiliation condition to group the re-collected data record sets corresponding to the same production batch of the same electronic ignition module into the same re-collected batch data set. Then, the process node is used as the classification condition to divide the same re-collected batch data set into re-collected record sets for electronic control module detection positions, conductive connection positions, material input positions, assembly operation positions, and ignition response verification positions. Finally, the acquisition time is used as the sorting condition to arrange the data in each re-collected record set in chronological order. Data with the same acquisition time in the re-collected record set is processed using a parallel storage method. Data with missing acquisition times is marked as having a missing acquisition time. Data with empty detection content is marked as having missing detection content. The same re-collected batch data set is then connected according to the process node order to form the regenerated batch quality data.

[0077] The regenerated batch quality data includes production batch identifier, re-acquisition batch marker, process node, acquisition time, data category, source location record, testing content, calibration process record, re-acquisition process record, and batch coupling risk results. The regenerated batch quality data also retains the correspondence with the original batch quality data; specifically, each re-acquisition data record in the regenerated batch quality data saves the production batch identifier, process node, data category, and acquisition time of the corresponding original data record. If a re-acquisition data record cannot find a corresponding original data record, a new re-acquisition record marker is written into the regenerated batch quality data. If the original data record was not re-acquisitioned, a non-re-acquisition marker is retained in the regenerated batch quality data.

Claims

1. An electronic ignition module production system, characterized in that, include: Quality acquisition module: Collects charge and discharge curve data, connection and continuity data, material status data, assembly process data and ignition response detection data of the same production batch of electronic ignition modules to form batch quality data; Breakpoint merging module: Merges batch quality data according to process nodes, marks the collected breakpoint data, retains adjacent data before and after exceeding the limit, and distinguishes between short-term impact segment and stable offset segment; Deviation extraction module: Based on the short-term impact segment and the stable offset segment, extract the inflection point of the charge and discharge curve, connect the conduction fluctuation segment, the material state gradual change segment and the assembly process reversal segment to form a local deviation segment; Risk matching module: Matches local deviation segments with the rework locations, number of reworks, and re-inspection results in historical fault rework records to form deviation risk segments; Coupling determination module: Based on the sequential transmission direction of deviation risk segments among electronic control module, material state and assembly process, continuous work station span and terminal detection offset consistency, batch coupling risk results are generated; Re-mining and calibration module: Based on the batch coupling risk results, determine the re-mining process and calibration process, and generate batch quality data again after executing the calibration process.

2. The electronic ignition module production system according to claim 1, characterized in that, Data on charge / discharge curves, connectivity data, material status data, assembly process data, and ignition response detection data of the same production batch of electronic ignition modules are collected to form batch quality data, specifically: When the same production batch of electronic ignition modules enters the process node, the following data are collected respectively: charge and discharge curve data formed by the electronic control module, connection and conduction data formed by the conductive connection position, material state data formed by the input production materials, assembly process data formed by the assembly operation position, and ignition response detection data formed by the ignition response verification position. Establish a correspondence between production batch identifiers, process nodes, and data collection times to form batch quality data.

3. The electronic ignition module production system according to claim 2, characterized in that, Batch quality data is merged according to process nodes, data collection breakpoints are marked, adjacent data before and after exceeding limits are retained, and short-term impact segments and stable offset segments are distinguished. Specifically: Batch quality data are merged according to production batch identifier and process node, and the merged batch quality data are arranged according to the time of collection. Data that is missing at the acquisition time or whose acquisition sequence is interrupted is marked as acquisition breakpoint data; Data that deviates from the preset range of the corresponding process node is considered out-of-limit data, and adjacent data before and after the out-of-limit data are retained. Based on the continuous state of the out-of-limit data and the adjacent data before and after the out-of-limit data, the short-term impact segment and the stable offset segment are distinguished.

4. The electronic ignition module production system according to claim 3, characterized in that, Based on the continuous state of the out-of-limit data and the adjacent data before and after the out-of-limit data, the short-term impact period and the stable offset period are distinguished, specifically as follows: Connect the out-of-limit data with the adjacent data before and after the out-of-limit data according to the collection time to form the out-of-limit neighborhood data; In the out-of-limit neighborhood data, when the data deviates from the preset range of the corresponding process node only appears within the adjacent collection time and the adjacent data before and after the out-of-limit data returns to the preset range of the corresponding process node, the out-of-limit neighborhood data is classified into the short-term impact period. When data deviating from the preset range of the corresponding process node maintains the same offset direction continuously along the acquisition time, the data in the out-of-limit neighborhood will be classified into the stable offset segment.

5. The electronic ignition module production system according to claim 4, characterized in that, Based on the short-term impact segment and the stable offset segment, the inflection point of the charge-discharge curve is extracted, and the conduction fluctuation segment, the material state gradual change segment, and the assembly process reversal segment are connected to form a local deviation segment, specifically: Using the short-term impact segment and the stable offset segment as positioning objects, the location where the change direction changes is extracted from the charge-discharge curve data to form the inflection point of the charge-discharge curve; Extract the continuous reciprocating range of the corresponding short-term impact segment and stable offset segment from the connection and conduction data to form the connection and conduction fluctuation segment; Extract the continuous unidirectional change interval of the corresponding stable offset segment from the material state data to form the material state gradual change segment; Extract the intervals where the direction of process change changes from the assembly process data to form the assembly process reversal segment; The inflection points of the charge / discharge curves, the connection and conduction fluctuation segments, the material state gradual change segments, and the assembly process reversal segments are collected according to the process nodes and the time of data collection to form local deviation segments.

6. The electronic ignition module production system according to claim 5, characterized in that, Local deviation segments are matched with the rework locations, rework times, and re-inspection results in historical fault rework records to form deviation risk segments, specifically: Read local deviation segments according to process nodes and acquisition times; Establish a correspondence between the inflection point of the charge / discharge curve, the connection and conduction fluctuation segment, the material state gradual change segment, and the assembly process reversal segment in the local deviation segment and the rework location in the historical fault rework record. Based on the corresponding relationship, read the number of reworks and the results of re-inspection, and mark the local deviation segments with consistent rework parts, duplicate rework records, and unqualified re-inspection results as deviation risk segments.

7. The electronic ignition module production system according to claim 6, characterized in that, Based on the corresponding relationship, the number of reworks and the re-inspection results are read. Local deviation segments where the reworked parts are consistent, the number of reworks is repeated, and the re-inspection results are unqualified are marked as deviation risk segments, specifically: Establish a correspondence between the location of the local deviation segment and the rework part in the historical fault rework record according to the process node and the time of data collection; When the location of the local deviation segment is at the process node corresponding to the rework part or the process node adjacent to the rework part, read the number of reworks and the re-inspection results; Local deviation segments with duplicate rework records and unqualified re-inspection results are classified as deviation risk segments.

8. The electronic ignition module production system according to claim 7, characterized in that, Based on the sequential transmission direction, continuous workstation span, and consistency of terminal detection offset of deviation risk segments among electronic control modules, material states, and assembly processes, batch coupling risk results are formed, specifically as follows: The deviation risk segments are arranged according to the process node and the time of data collection, and the order of occurrence of the deviation risk segments among the electronic control module, material status and assembly process is identified to form the direction of sequential transmission. The span of continuous workstations is determined based on the consecutive process nodes crossed by the deviation risk segment; The offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node is compared with the offset direction of the deviation risk segment in the corresponding process node to form the terminal detection offset consistency. Based on the sequential transmission direction, continuous workstation span, and consistency of terminal detection offset, batch coupling risk results are generated.

9. The electronic ignition module production system according to claim 8, characterized in that, The offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node is compared with the offset direction of the deviation risk segment in the corresponding process node to form the terminal detection offset consistency, specifically: Read the offset direction of the ignition response detection data corresponding to the deviation risk segment relative to the preset range of the corresponding process node, and read the offset direction of the deviation risk segment in the corresponding process node; When the offset direction of the ignition response detection data and the offset direction of the deviation risk segment in the corresponding process node are on the same side of the preset range of the corresponding process node, a consistent mark is formed. When the offset direction of the ignition response detection data is opposite to the offset direction of the deviation risk segment in the corresponding process node, an inconsistency mark is formed; Terminal detection offset consistency is formed based on consistent and inconsistent markers.

10. The electronic ignition module production system according to claim 9, characterized in that, Based on the batch coupling risk results, the re-mining process and the calibration process are determined. After the calibration process is executed, batch quality data is generated again, specifically as follows: The starting node of the remining process is determined based on the direction of sequential transfer, and the ending node of the remining process is determined based on the span of the continuous work station. Based on the consistency of terminal detection offset, a correction process is determined within the re-sampling process; After the calibration process is performed, charge-discharge curve data, connection continuity data, material status data, assembly process data, and ignition response test data are re-acquired in the re-collection process. Correspondence is established according to the production batch identifier, process node, and collection time to form the regenerated batch quality data.

Images