A carbon fiber reinforced supercritical foaming material whole-process quality management and control system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FUJIAN XINRUI NEW MATERIALS TECHNOLOGY CO LTD
- Filing Date
- 2026-05-27
- Publication Date
- 2026-06-23
Smart Images
Figure CN122264641A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of production quality control technology, specifically to a full-process quality control system for carbon fiber reinforced supercritical foamed materials. Background Technology
[0002] Carbon fiber reinforced supercritical foamed material is a composite material prepared by supercritical fluid foaming of carbon fiber and polymer matrix. The uniformity of the internal pore structure of the finished product is affected by the dispersion state of the carbon fiber in the polymer melt. Different batches of carbon fiber delivered by the carbon fiber supplier exhibit batch-to-batch quality fluctuations in terms of monofilament diameter distribution and sizing agent content. These batch-to-batch quality fluctuations affect the uniformity of the pore distribution in the finished product during the foaming process.
[0003] In existing technologies, quality control methods in the production process of carbon fiber reinforced supercritical foamed materials include incoming carbon fiber inspection, intermediate product inspection during production, and sampling inspection of finished foamed products. Incoming carbon fiber inspection records the monofilament diameter and sizing agent content of each batch of carbon fiber. Intermediate product inspection involves collecting samples at fixed time intervals at each workstation and checking their appearance and dimensions. Finished product sampling inspection detects the internal pore distribution of the foamed product and determines its quality grade. The quality data generated at each of these stages are stored in different tables within the production management database.
[0004] In existing technologies, carbon fiber incoming inspection data is only used for batch acceptance judgment. No proactive correlation is established between carbon fiber batch quality fluctuation characteristics data and intermediate product inspection data, or finished product sampling inspection data, across batches. When a statistical correspondence exists between the quality fluctuation characteristics of a carbon fiber batch and historical foamed finished product quality deviations, the production scheduling order of the current carbon fiber batch is not adjusted based on this statistical correspondence before the material feeding stage, nor is the sampling frequency for the corresponding production task of the current carbon fiber batch differentiated. When non-conforming items are found in the sampling inspection results of foamed finished products, existing technologies do not link the non-conforming items to the carbon fiber supplier's batch code for traceability, nor do they perform any changes to the availability status of the remaining carbon fiber inventory in the same batch based on the traceability results. Therefore, the remaining carbon fiber inventory in the same batch can still be used in subsequent production tasks. The transmission of carbon fiber batch quality fluctuations to foamed finished product quality deviations cannot be effectively identified and blocked before the batch is consumed. The batch-level traceability chain for finished product quality deviations lacks a complete closed-loop record from carbon fiber batch quality prediction to production scheduling adjustment to changes in the status of remaining inventory. Summary of the Invention
[0005] The purpose of this invention is to provide a full-process quality control system for carbon fiber reinforced supercritical foamed materials, solving the following technical problems:
[0006] Existing technologies do not establish proactive correlation and closed-loop traceability between carbon fiber batch quality fluctuations and production scheduling and inventory status, resulting in the inability to prevent the transmission of batch quality fluctuations to finished product defects before material feeding and to block it after material feeding.
[0007] The objective of this invention can be achieved through the following technical solutions: A full-process quality control system for carbon fiber reinforced supercritical foamed materials includes: The task receiving module is used to obtain the production task order number recorded in the production task management database, as well as the associated carbon fiber supplier batch code, polymer matrix supplier batch code, and process formula code. The quality prediction module is used to retrieve historical delivery data records of the same supplier based on the carbon fiber supplier batch code, determine the historical batch with the most similar quality fluctuation characteristics, and extract the finished product quality rating level corresponding to the historical batch as the expected quality level reference value. The traceability construction module is used to retrieve material flow data records by production task order number, obtain the flow sequence and time information of this production task, and construct a material consumption traceability chain containing multiple sequential connection nodes. The scheduling shift module is used to compare the expected quality level reference value with the preset quality level warning value before the time information associated with the first flow link node. When the expected quality level reference value is lower than the preset quality level warning value, the production task order number is shifted backward in the daily scheduling queue. The inventory isolation module is used to obtain the finished product inspection result data of the current production task after the time information associated with the node of the final circulation link. When the result data contains non-conforming items and there is a scheduling queue movement record for the current task, it generates a status change instruction for the remaining inventory of the same batch. The closed-loop record module is used to integrate the batch quality prediction basis, scheduling queue movement record, finished product inspection conclusion and remaining inventory status change of this production task into a batch quality closed-loop record and push it to the production management terminal interface.
[0008] As a further aspect of the present invention: the process of determining the historical batch with the most similar quality fluctuation characteristics in the quality prediction module is as follows: Retrieve the carbon fiber monofilament diameter detection value sequence for each historical batch from the same supplier's historical delivery data records. Perform a piecewise linear fitting operation on each carbon fiber monofilament diameter detection value sequence. The division position of the segment interval is determined by the position where the fluctuation amplitude of three consecutive points exceeds the preset threshold. Extract the inflection point of the slope change rate of each segment fitting line and arrange them in chronological order to form a sequence of inflection points of quality fluctuation trend. Perform the same operation on the carbon fiber monofilament diameter detection value sequence corresponding to the current carbon fiber supplier batch code to obtain the current batch quality fluctuation trend inflection point sequence. Calculate the weighted deviation value between the current batch quality fluctuation trend inflection point sequence and the quality fluctuation trend inflection point sequence of each historical batch based on the difference in the distance and quantity of the inflection point positions. Select the historical batch with the smallest weighted deviation value as the historical batch with the most similar quality fluctuation characteristics.
[0009] As a further aspect of the present invention: the carbon fiber monofilament diameter detection value sequence is extracted from the incoming inspection data table corresponding to the carbon fiber supplier batch code. The incoming inspection data table is indexed with the carbon fiber supplier batch code as the primary key. The carbon fiber monofilament diameter detection value sequence is composed of multiple repeated measurement values arranged in the order of measurement. In the piecewise linear fitting operation, the preset threshold value is the third quartile of the fluctuation amplitude values of three adjacent points in the carbon fiber monofilament diameter detection value sequence of all historical delivered batches of the same supplier. The calculation process of the weighted deviation value is as follows: the time interval distance between each turning point in the current batch quality fluctuation trend turning point sequence and the nearest turning point in the corresponding time neighborhood in the quality fluctuation trend turning point sequence of each historical batch is accumulated to obtain the position deviation sub-value. The absolute value of the difference between the total number of turning points of the current batch and the historical batch is multiplied by a preset weight coefficient to obtain the quantity deviation sub-value. The position deviation sub-value and the quantity deviation sub-value are added together to obtain the weighted deviation value.
[0010] As a further aspect of the present invention: in the traceability construction module, the process of retrieving material flow data records by production task order number and constructing a material consumption traceability chain containing multiple sequentially connected nodes is as follows: The production task management database stores a material flow data table, which includes fields for production task order number, material batch code, flow stage code, entry time, and exit time. Using the current production task order number as the search criterion, all corresponding records in the material flow data table are queried. The query results are sorted in ascending order by the entry time field value. The flow stage code field in each sorted record is then read sequentially and mapped to feeding stage nodes, mixing stage nodes, injection stage nodes, and molding stage nodes in the order of their first appearance. A node data structure is created for each node, containing the entry and exit time field values of the corresponding record. The nodes are then connected end-to-end according to the mapping order to form a material consumption traceability chain. A bidirectional reference index record between the flow stage nodes and the carbon fiber supplier batch code is established in the auxiliary index storage area of the material consumption traceability chain.
[0011] As a further aspect of the present invention: the method for determining the preset quality level warning value in the scheduling shift module is as follows: The system retrieves batch quality closed-loop records corresponding to completed production tasks from the production task management database. Completed production tasks are those marked as "completed" in the production task status field. The system extracts the expected quality level reference value and the finished product inspection conclusion value from each batch quality closed-loop record. Batch quality closed-loop records with unqualified finished product inspection conclusion values are selected to form a warning value calibration sample record set. The expected quality level reference value values of each record in the warning value calibration sample record set are arranged in descending order. The expected quality level reference value value corresponding to the preset percentile order in the arrangement sequence is extracted as the preset quality level warning value. At the end of each calendar month, the preset quality level warning value is re-executed based on the newly added batch quality closed-loop records of completed production tasks for that month, and the re-extracted values are updated and stored in the system configuration parameter storage area.
[0012] As a further aspect of the present invention: the process of moving the production task order number of the current day to the next stage in the scheduling queue in the scheduling shift module is as follows: Obtain the expected quality level reference values corresponding to all pending production task order numbers for the day. Sort all pending production task order numbers for the day in descending order of expected quality level reference values to generate an initial scheduling queue. Extract the current production task order number from the current position in the initial scheduling queue. Insert the extracted current production task order number into the position after all production task order numbers in the initial scheduling queue whose expected quality level reference values are lower than the preset quality level warning value, forming an adjusted scheduling queue. Record the original position number of the current production task order number in the initial scheduling queue and its new position number in the adjusted scheduling queue.
[0013] As a further aspect of the present invention: the specific process of generating the remaining inventory availability status change instruction for the same batch in the inventory isolation module is as follows: A data structure for changing the availability status of remaining inventory in the same batch is created. This data structure includes a carbon fiber supplier batch code field, an inventory availability status change operation code field, a change trigger source production task order number field, and an instruction execution status field. The carbon fiber supplier batch code corresponding to the current production task is written into the carbon fiber supplier batch code field, the restricted use status code is written into the inventory availability status change operation code field, the current production task order number is written into the change trigger source production task order number field, and the pending execution status code is written into the instruction execution status field. The completed data structure is written into the batch status change instruction queue of the inventory management system. The inventory management system polls the batch status change instruction queue and reads instruction records whose instruction execution status field has a pending execution status code. Based on the carbon fiber supplier batch code in the instruction record, the inventory record is located and the inventory availability status modification operation is performed.
[0014] As a further aspect of the present invention: the process of integrating the closed-loop recording module into a batch quality closed-loop record is as follows: A batch quality closed-loop record data structure is created, which includes a production task order number field, a carbon fiber supplier batch code field, a precursor historical batch identifier field, an expected quality grade reference value field, a scheduling queue original position field, a scheduling queue adjusted position field, a finished product inspection conclusion field, and a same batch remaining inventory availability status change instruction execution status field. Data corresponding to each operation step of this production task is extracted and written into each field. The precursor historical batch batch code is written into the precursor historical batch identifier field, the position number before scheduling queue movement is written into the scheduling queue original position field, the position number after scheduling queue movement is written into the scheduling queue adjusted position field, and the generation status of the same batch remaining inventory availability status change instruction is written into the same batch remaining inventory availability status change instruction execution status field. This data structure is stored in the batch quality closed-loop record table of the production task management database.
[0015] The beneficial effects of this invention are: This invention retrieves historical delivery data records corresponding to carbon fiber supplier batch codes before material input, extracts the carbon fiber monofilament diameter detection value sequence, and performs piecewise linear fitting and inflection point sequence extraction operations. It calculates the weighted deviation value between the current batch and the inflection point sequences of quality fluctuation trends of each historical batch, selects the historical batch with the smallest weighted deviation value as the preceding historical batch, and uses the finished product quality rating level corresponding to the preceding historical batch as the expected quality level reference value. When the expected quality level reference value is lower than the preset quality level warning value, the production task order number is moved backward in the daily scheduling queue. When the finished product inspection result shows a non-conforming item and the current production task order number has a scheduling queue movement record, a change instruction for the available status of the remaining inventory of the same batch is generated, changing the available status of the remaining carbon fiber inventory of the same batch to a restricted use status.
[0016] This invention establishes a proactive correlation between the batch quality fluctuation characteristics of carbon fiber and the production scheduling queue position. Before carbon fiber feeding, the scheduling operation is moved back for carbon fiber batches with high quality risk. At the same time, a feedback correlation is established between the finished product quality deviation information and the remaining inventory status of the same batch. This forms a batch-level quality closed-loop record from carbon fiber batch quality prediction to production scheduling adjustment and then to the change of remaining inventory status. This realizes the prevention before feeding and the blocking after feeding of the transmission of carbon fiber supplier batch quality fluctuations to foamed finished product quality deviations. Attached Figure Description
[0017] The invention will now be further described with reference to the accompanying drawings.
[0018] Figure 1This is a schematic diagram of the modules of the present invention. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0020] Please see Figure 1 As shown, this invention is a full-process quality control system for carbon fiber reinforced supercritical foamed materials, comprising: The task receiving module is used to obtain the production task order number recorded in the production task management database, as well as the associated carbon fiber supplier batch code, polymer matrix supplier batch code, and process formula code. The quality prediction module is used to retrieve historical delivery data records of the same supplier based on the carbon fiber supplier batch code, determine the historical batch with the most similar quality fluctuation characteristics, and extract the finished product quality rating level corresponding to the historical batch as the expected quality level reference value. The traceability construction module is used to retrieve material flow data records by production task order number, obtain the flow sequence and time information of this production task, and construct a material consumption traceability chain containing multiple sequential connection nodes. The scheduling shift module is used to compare the expected quality level reference value with the preset quality level warning value before the time information associated with the first flow link node. When the expected quality level reference value is lower than the preset quality level warning value, the production task order number is shifted backward in the daily scheduling queue. The inventory isolation module is used to obtain the finished product inspection result data of the current production task after the time information associated with the node of the final circulation link. When the result data contains non-conforming items and there is a scheduling queue movement record for the current task, it generates a status change instruction for the remaining inventory of the same batch. The closed-loop record module is used to integrate the batch quality prediction basis, scheduling queue movement record, finished product inspection conclusion and remaining inventory status change of this production task into a batch quality closed-loop record and push it to the production management terminal interface.
[0021] In a preferred embodiment of the present invention, the process of determining the historical batch with the most similar quality fluctuation characteristics in the quality prediction module is as follows: The production task management database stores an incoming inspection data table. This table uses the carbon fiber supplier batch code as the primary key to establish an index structure. It records the sequence of carbon fiber monofilament diameter measurements obtained after each carbon fiber delivery. The sequence consists of 30 diameter values obtained from microscopic measurements of 30 carbon fiber monofilament samples taken from a single delivery batch, arranged in chronological order of measurement. All 30 diameter values are in micrometers. For example, for a carbon fiber supplier batch with batch code CF20241015, the sequence of carbon fiber monofilament diameter measurements is 7.12, 7.08, 7.15, 7.11, 7.09, up to the 30th value of 7.13.
[0022] Once the production order number is received, the industrial control computer parses the supplier's unique code portion of the carbon fiber supplier's batch code associated with that production order number. The supplier's unique code consists of the first six characters of the carbon fiber supplier's batch code. Using the supplier's unique code as the search criterion, the industrial control computer retrieves the carbon fiber monofilament diameter test value sequence for all historical batches delivered by the same supplier from the incoming inspection data table in the production task management database. The historical delivery data records for the same supplier contain incoming inspection data for all batches since the first delivery from that supplier.
[0023] For each historical batch of carbon fiber monofilament diameter detection numerical sequence obtained from the retrieval, a piecewise linear fitting operation is performed. The piecewise linear fitting operation first scans all three adjacent values in the carbon fiber monofilament diameter detection numerical sequence, calculates the absolute value of the difference between the first and second values plus the absolute value of the difference between the second and third values, and uses the sum as the fluctuation amplitude value of that three adjacent values. For example, if the 5th, 6th, and 7th values in the numerical sequence are 7.21, 7.08, and 7.33 respectively, then the fluctuation amplitude value is the absolute value of 7.21 minus 7.08 (0.13) plus the absolute value of 7.08 minus 7.33 (0.25), totaling 0.38. The preset fluctuation amplitude threshold is set to the third and fourth quartile of the fluctuation amplitude values of all three adjacent values in the carbon fiber monofilament diameter detection numerical sequence of all historical delivery batches from the same supplier. The third and fourth quartile is obtained by sorting all three adjacent fluctuation amplitude values from smallest to largest and taking the value at the 75th percentile. For example, if a supplier has 1450 consecutive fluctuation values for all three adjacent points in all historical delivery batches, and the 1088th value after sorting is 0.52, then the preset fluctuation amplitude threshold is 0.52 micrometers.
[0024] When the fluctuation range of three adjacent data points exceeds 0.52 micrometers, the midpoint of these three data points is marked as the division point of the segmented interval. The carbon fiber monofilament diameter detection data sequence is divided into multiple continuous segmented intervals by the above division points, with each segmented interval containing at least four consecutive data points. Least squares linear fitting is performed on all data points within each segmented interval to obtain the corresponding fitted line. The expression for the fitted line is y = a multiplied by x + b, where x is the measurement sequence number, y is the fitted value of the carbon fiber monofilament diameter, and a is the slope of the fitted line. The slope change rate is calculated for each adjacent segmented interval. The slope change rate is the difference between the slope 'a' of the fitted line of the subsequent segmented interval and the slope 'a' of the fitted line of the previous segmented interval, divided by the measurement sequence number interval between the center points of the two segmented intervals. When the absolute value of the slope change rate is greater than 0.05, the starting point of the subsequent segmented interval is recorded as a slope change rate inflection point.
[0025] The inflection points of all slope change rates extracted from the numerical sequence of carbon fiber monofilament diameter measurements for each historical batch are arranged in chronological order according to the original time sequence of the numerical sequence, forming the inflection point sequence of the quality fluctuation trend for that historical batch. For example, if a historical batch has 5 inflection points extracted, located at positions 8, 12, 19, 24, and 27 of the measurement sequence, then the inflection point sequence of the quality fluctuation trend for that historical batch is 8, 12, 19, 24, and 27.
[0026] The same operational procedure is performed on the carbon fiber monofilament diameter detection value sequence corresponding to the current carbon fiber supplier batch code: piecewise linear fitting, extraction of slope change rate inflection points, and arranging in chronological order to obtain the current batch quality fluctuation trend inflection point sequence. For example, if the current batch of carbon fiber monofilament diameter detection value sequence contains 30 values, and after the above operation, the inflection points are extracted to be located at the 6th, 14th, and 22nd measurement serial numbers, then the current batch quality fluctuation trend inflection point sequence is 6, 14, and 22.
[0027] Subsequently, the weighted deviation values between the current batch's quality fluctuation trend inflection point sequence and each historical batch's quality fluctuation trend inflection point sequence are calculated sequentially. The weighted deviation value is obtained by adding the positional deviation component value and the quantity deviation component value.
[0028] The calculation method for the positional deviation sub-value is as follows: Iterate through the positional values of each turning point in the current batch's quality fluctuation trend turning point sequence. For each current batch turning point, find the turning point in the historical batch quality fluctuation trend turning point sequence that is closest to the current batch's turning point in terms of time position. The closeness in time position is measured by the absolute value of the difference between the corresponding measurement sequence numbers of the two turning points. This absolute value is taken as the time interval distance of the current batch's turning point. The time interval distances corresponding to all turning points in the current batch's quality fluctuation trend turning point sequence are summed, and the summed result is the positional deviation sub-value. For example, when matching the current batch turning point sequence 6, 14, 22 with a historical batch turning point sequence 8, 12, 19, 24, 27, the distance between turning point 6 and historical sequence 8 is 2, the distance between turning point 14 and 12 is 2, and the distance between turning point 22 and 24 is 2. The summed result is 6, so the positional deviation sub-value is 6.
[0029] The calculation method for the quantity deviation sub-item value is as follows: Calculate the total number of turning points in the current batch's quality fluctuation trend turning point sequence, and calculate the total number of turning points in the historical batch's quality fluctuation trend turning point sequence. Calculate the absolute value of the difference between the two totals, and multiply this absolute value by a preset quantity weighting coefficient. In this embodiment, the preset quantity weighting coefficient is set to 1.5. For example, if the current batch has 3 turning points and a historical batch has 5 turning points, the absolute value of the difference is 2. Therefore, the quantity deviation sub-item value is 2 multiplied by 1.5, which equals 3.0.
[0030] Add the positional deviation value of 6 to the quantity deviation value of 3.0 to obtain the weighted deviation value of 9.0 between the historical batch and the current batch.
[0031] Following the above calculation process, a weighted deviation value is calculated for all historical batches from the same supplier. The historical batch with the smallest weighted deviation value is selected as the historical batch with the most similar quality fluctuation characteristics, and the batch code of this historical batch is recorded as the identifier of the preceding historical batch.
[0032] The industrial control computer reads the finished product quality rating corresponding to the previous historical batch identifier from the finished product inspection data table in the production task management database. The finished product quality rating is stored in the finished product inspection data table, which is indexed using the subcontracted production task number as the primary key. The finished product quality rating value is one of three grades: first-class, second-class, or third-class. The read finished product quality rating is the expected quality rating reference value for this production task.
[0033] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
[0034] In another preferred embodiment of the present invention, the process of retrieving material flow data records by production task order number and constructing a material consumption traceability chain containing multiple sequentially connected nodes in the traceability construction module is as follows: The production task management database stores a material flow data table. This table records detailed time and material information for each production task order number as it sequentially passes through each stage of the carbon fiber reinforced supercritical foam material production process. The data structure of the material flow data table includes five fields: production task order number, material batch code, stage code, entry time, and exit time. The production task order number field is a 18-character fixed-length string. The first eight characters represent the year, month, and day, and the last ten characters represent the daily sequence number. For example, 20241120000000001234 represents the 1234th production task on November 20, 2024. The material batch code field is a variable-length string, storing the batch code of either the carbon fiber supplier or the polymer matrix supplier. The encoding rule for the carbon fiber supplier batch code is a 6-digit unique supplier code plus an 8-digit arrival date plus a 4-digit serial number. For example, FIBER1202411150003 indicates that the supplier code is FIBER12, the goods arrived on November 15, 2024, and it is the 3rd batch of that day. The circulation stage code field is a 2-digit fixed-length string, with values including TL for feeding, HH for mixing, ZR for injection, and CX for molding. The entry time and exit time fields are both date and time types, accurate to the second, in a format of 14 consecutive digits: year, month, day, hour, minute, second. For example, 20241120084530 represents November 20, 2024, 8:45:30 AM.
[0035] When a production task with production order number 20241120000000001234 is received, the industrial control computer performs a structured query in the material flow data table using this production task order number as the search condition. The search statement condition is that the production task order number field equals 20241120000000001234. There are four records in the material flow data table that match this production task order number, corresponding to the feeding stage, mixing stage, injection stage, and molding stage, respectively. The industrial control computer sorts the above four records in ascending order by the value of the entry time field. The sorted record order is as follows: the entry time of the first record is 20241120084530, and the circulation stage code is TL; the entry time of the second record is 20241120085115, and the circulation stage code is HH; the entry time of the third record is 20241120090640, and the circulation stage code is ZR; the entry time of the fourth record is 20241120091820, and the circulation stage code is CX.
[0036] The industrial control computer sequentially reads the flow stage code field from each of the sorted records. The first flow stage code read is TL, which the industrial control computer maps to the feeding stage node. The second flow stage code read is HH, which the industrial control computer maps to the mixing stage node. The third flow stage code read is ZR, which the industrial control computer maps to the injection stage node. The fourth flow stage code read is CX, which the industrial control computer maps to the forming stage node. The result of the above mapping operations is a sequential arrangement of four nodes: feeding stage node, mixing stage node, injection stage node, and forming stage node.
[0037] The industrial control computer creates node data structures for each of the four nodes. Each node data structure is an in-memory data container containing five attributes: node name, node type, entry time, exit time, and associated batch code. For the feeding node, the node name is assigned "Feeding," the node type is assigned "TL," the entry time is assigned "20241120084530," the exit time is assigned "20241120085045," and the associated batch code is assigned "FIBER1202411150003." For the mixed-process node, the node name is assigned "Mixed-process," the node type is assigned "HH," the entry time is assigned "20241120085115," the exit time is assigned "20241120090610," and the associated batch code is assigned "FIBER1202411150003." For the injection stage node, the node name attribute is assigned as "Injection Stage", the node type attribute is assigned as "ZR", the entry time attribute is assigned as "20241120090640", the exit time attribute is assigned as "20241120091750", and the associated batch code attribute is assigned as "FIBER1202411150003". For the molding stage node, the node name attribute is assigned as "Molding Stage", the node type attribute is assigned as "CX", the entry time attribute is assigned as "20241120091820", the exit time attribute is assigned as "20241120093210", and the associated batch code attribute is assigned as "FIBER1202411150003".
[0038] The industrial control computer connects the feeding, mixing, injection, and molding nodes in a mapping order to form a material consumption traceability chain. The connection method involves adding pointer variables to the data structures of the feeding, mixing, injection, and molding nodes, pointing to the storage addresses of the mixing, injection, and molding node data structures respectively. This creates a unidirectional linked list structure starting from the feeding stage, passing through the mixing and injection stages, and ending at the molding stage; this unidirectional linked list structure is the material consumption traceability chain.
[0039] After the material consumption traceability chain is constructed, the industrial control computer establishes a bidirectional reference index record between the circulation links and the carbon fiber supplier batch codes in the auxiliary index storage area of the material consumption traceability chain. The auxiliary index storage area is an independently allocated memory region that internally maintains a hash index table. The key of the hash index table is the carbon fiber supplier batch code string, and the value is an array of pointers to the corresponding nodes in the material consumption traceability chain. For the carbon fiber supplier batch code FIBER1202411150003, the industrial control computer calculates the hash value of the string, looks up the corresponding slot in the hash index table, and stores the pointers of all four nodes associated with the carbon fiber supplier batch code into the pointer array in sequence. The order in which the pointers are stored in the pointer array is consistent with the order in which the nodes appear in the material consumption traceability chain. At the same time, the carbon fiber supplier batch code is stored in the associated batch code attribute of each node. When it is necessary to query the corresponding carbon fiber supplier batch code from any node, the associated batch code attribute of that node can be read directly. After the bidirectional reference index record is established, during the subsequent integration of quality closed-loop records, all nodes in the material consumption traceability chain associated with the carbon fiber supplier batch code can be located within a constant time based on the carbon fiber supplier batch code. Alternatively, the corresponding carbon fiber supplier batch code can be obtained from any node in the material consumption traceability chain.
[0040] In another preferred embodiment of the present invention, the method for determining the preset quality level warning value in the scheduling shift module is as follows: The production task management database stores a batch quality closed-loop record table, which stores the batch quality closed-loop records generated for each completed production task. A completed production task is defined as one whose production task status field is marked as completed. This status field is automatically updated to completed by the industrial control computer after the departure time of the forming stage node in the material consumption traceability chain is recorded. The batch quality closed-loop record table contains 12 fields, including production task order number, carbon fiber supplier batch code, expected quality grade reference value, finished product inspection conclusion, original position in the scheduling queue, and adjusted position in the scheduling queue. The expected quality grade reference value field takes the values of first-class, second-class, or third-class. The finished product inspection conclusion field takes the values of qualified or unqualified.
[0041] At the end of each calendar month, the industrial control computer performs a recalculation and update operation of the preset quality level warning value. The recalculation operation first retrieves all batch quality closed-loop records from the batch quality closed-loop record table where the production task status field is "completed" and the completion date belongs to the current month. From each batch quality closed-loop record, the expected quality level reference value field value and the finished product inspection conclusion field value are extracted. Then, batch quality closed-loop records with a finished product inspection conclusion field value of "unqualified" are selected, and all selected records form the warning value calibration sample record set. For example, if at the end of a calendar month, 350 completed production tasks are retrieved, and 28 of them have a finished product inspection conclusion of "unqualified," then the warning value calibration sample record set includes these 28 records.
[0042] The industrial control computer converts the expected quality grade reference value corresponding to each record in the warning value calibration sample record set into numerical data for sorting and comparison. The conversion rule is: first-grade product corresponds to a value of 3, second-grade product corresponds to a value of 2, and third-grade product corresponds to a value of 1. After conversion, 28 values are obtained, for example, including 8 first-grade products corresponding to the value 3 (8 in total), 12 second-grade products corresponding to the value 2 (12 in total), and 8 third-grade products corresponding to the value 1 (8 in total). The industrial control computer arranges the above 28 values in descending order, and the sorting result is 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1.
[0043] In this embodiment, the preset percentile is set to the 80th percentile. The position corresponding to the 80th percentile is calculated by multiplying the total number of records in the warning value calibration sample record set by 0.8. 28 multiplied by 0.8 equals 22.4, which is rounded up to the 23rd position. The value corresponding to the 23rd position in the permutation sequence is 1. Reversing the value 1, we can convert it to the corresponding quality grade of third-class product. Therefore, the preset quality grade warning value is determined to be third-class product.
[0044] The industrial control computer writes the determined preset quality level warning value (Grade III) into the preset quality level warning value configuration item in the system configuration parameter storage area, overwriting the old value in that item. The system configuration parameter storage area is a key-value pair configuration table in the production task management database, where the key is the preset quality level warning value and the value is Grade III. All scheduling shift judgment operations in the following calendar month will use the updated Grade III as the preset quality level warning value.
[0045] The process of moving the production task order number backward in the daily scheduling queue in the scheduling shift module is as follows.
[0046] At the start of each production day, the industrial control computer reads all pending production task order numbers from the production task management database. These are the production task order numbers whose production task status field is marked as "awaiting material input" and whose planned production date is equal to the current date. The industrial control computer then retrieves the expected quality level reference value for each pending production task order number. This expected quality level reference value is calculated by the quality prediction module when the production task order number is received and stored in the production task basic information table of the production task management database.
[0047] The industrial control computer combines all pending production task numbers and their corresponding expected quality level reference values into a sorted tuple list. Each sorted tuple contains two parts: the production task number string and the expected quality level reference value. The sorted tuple list is ordered from highest to lowest according to the expected quality level reference value, with the value 3 (for first-class products) at the highest position, the value 2 (for second-class products) in the middle position, and the value 1 (for third-class products) at the lowest position. For example, if there are 5 pending production tasks on a given day, with production task numbers and expected quality level reference values of Task A (second-class product), Task B (first-class product), Task C (third-class product), Task D (first-class product), and Task E (second-class product), the sorted order would be: Task B (value 3), Task D (value 3), Task A (value 2), Task E (value 2), and Task C (value 1). This sorted sequence is the initial scheduling queue, where each position corresponds to a position number: Task B is the 1st position, Task D is the 2nd position, Task A is the 3rd position, Task E is the 4th position, and Task C is the 5th position.
[0048] When the expected quality level reference value corresponding to the current production task order number is lower than the preset quality level warning value, the industrial control computer triggers a scheduling shift operation. For example, if the current production task order number is Task A, its expected quality level reference value is Grade 2, and the preset quality level warning value is Grade 3, the value 2 corresponding to Grade 2 is not lower than the value 1 corresponding to Grade 3, therefore no scheduling shift operation is triggered. If the current production task order number is Task C, its expected quality level reference value is Grade 3, and the value 1 corresponding to Grade 3 is lower than the value 1 corresponding to Grade 3, the comparison result is equal. In this case, according to the preset rule, when the expected quality level reference value equals the preset quality level warning value, a scheduling shift operation is also triggered.
[0049] The scheduling shift operation first records the original position number of the current production task order in the initial scheduling queue. The original position number of task C is the 5th position. The industrial control computer extracts the sorted tuple corresponding to task C from its current position in the initial scheduling queue. After extraction, the initial scheduling queue has 4 tuples remaining, in the following order: task B value 3, task D value 3, task A value 2, and task E value 2.
[0050] The industrial control computer iterates through all tuples in the remaining queue, searching for production task numbers whose expected quality level reference value is lower than the preset quality level warning value. A value lower than the preset quality level warning value is defined as a value less than 1. In the remaining queue, task B (value 3), task D (value 3), task A (value 2), and task E (value 2) are all not less than 1. Therefore, there are no production task numbers in the queue with expected quality level reference values lower than the preset quality level warning value. In this case, the industrial control computer inserts the extracted task C into the end of the queue. After insertion, the queue order is task B (value 3), task D (value 3), task A (value 2), task E (value 2), and task C (value 1). This sequence is the adjusted scheduling queue. The industrial control computer records task C's new position in the adjusted scheduling queue as the 5th position.
[0051] If multiple production tasks with an expected quality level reference value of Grade 3 exist simultaneously in the daily production schedule, for example, besides Task C, there is also Task F with an expected quality level reference value of Grade 3. The initial scheduling queue is: Task B (value 3), Task D (value 3), Task A (value 2), Task E (value 2), Task C (value 1), Task F (value 1). When Task C triggers a scheduling shift, its original position is 5th. After Task C is removed, the queue is: Task B (value 3), Task D (value 3), Task A (value 2), Task E (value 2), Task F (value 1). Production task orders with expected quality level reference values lower than the preset quality level warning value in the queue have no order numbers. Task C is inserted at the end of the queue, and the queue becomes: Task B (value 3), Task D (value 3), Task A (value 2), Task E (value 2), Task F (value 1), Task C (value 1), Task C (value 1), and Task C's new position is 6th.
[0052] The industrial control computer writes the original position number 5 and the new position number 6 of this production task order number into the corresponding original position field and adjusted position field of the scheduling queue in the production task basic information table of the production task management database. The final order of the adjusted scheduling queue will be read sequentially by the production execution terminal, which will then execute the carbon fiber feeding operation corresponding to each production task order number according to the order of the adjusted scheduling queue.
[0053] In another preferred embodiment of the present invention, the specific process of generating the available status change instruction for the remaining inventory in the same batch in the inventory isolation module is as follows: Once the departure time of the foaming and molding stage is recorded, the industrial control computer retrieves the finished product sampling inspection result data corresponding to this production task. The finished product sampling inspection result data is stored in the finished product inspection data table of the production task management database. This table includes fields for production task order number, inspection item name, inspection item standard value range, inspection item measured value, and single-item judgment conclusion. Using the current production task order number as the search condition, the industrial control computer queries the finished product inspection data table for the values of the single-item judgment conclusion fields for all inspection items. If the value of any single-item judgment conclusion field for any inspection item is unqualified, the industrial control computer determines that the finished product inspection result data for this production task contains unqualified items. The industrial control computer then retrieves the scheduling queue movement record for this production task order number recorded in the scheduling shift module. This scheduling queue movement record is stored in the production task basic information table of the production task management database, containing the original position field and the adjusted position field of the scheduling queue. When the original position field value of the scheduling queue is not equal to the adjusted position field value of the scheduling queue, the industrial control computer determines that there is a scheduling queue movement record for the current production task order number.
[0054] After determining that the finished product inspection results contain non-conforming items and that the production task order number has a scheduling queue movement record, the industrial control computer executes the generation operation of a change instruction for the available status of the remaining inventory in the same batch. The industrial control computer first creates a data structure for the change instruction in the available status of the remaining inventory in the same batch in memory. This data structure is a data object containing five attributes. The first attribute is the carbon fiber supplier batch code, a 18-bit fixed-length string. The second attribute is the inventory availability status change operation code, a 2-bit fixed-length string with values of 01 (change from available to restricted use), 02 (change from restricted use to available), and 03 (change from restricted use to frozen). The third attribute is the change trigger source production task order number, also an 18-bit fixed-length string. The fourth attribute is the instruction execution status, a 1-bit fixed-length string with values of 0 (pending execution), 1 (in execution), 2 (execution successful), and 3 (execution failed). The fifth attribute is the instruction creation time, a date and time type, accurate to the second.
[0055] The industrial control computer reads the carbon fiber supplier batch code field value from the production task basic information table corresponding to this production task, and writes the read carbon fiber supplier batch code string into the carbon fiber supplier batch code attribute of the remaining inventory availability status change instruction data structure for the same batch. For example, if the carbon fiber supplier batch code associated with this production task is FIBER1202411150003, then this attribute is assigned the value FIBER1202411150003. The industrial control computer assigns the inventory availability status change operation code attribute to 01, where 01 represents changing the carbon fiber batch from an available status to a restricted usage status. The industrial control computer assigns the change trigger source production task order number attribute to the current production task order number, for example, 20241120000000001234. The industrial control computer assigns the instruction execution status attribute to 0, where 0 represents a pending execution status. The industrial control computer assigns the instruction creation time attribute to the current date and time value, for example, 20241120104530.
[0056] The industrial control computer writes the completed batch remaining inventory availability status change instruction data structure into the batch status change instruction queue list of the inventory management database. The batch status change instruction queue list is a first-in, first-out message queue list, with a table structure containing seven fields: instruction serial number, carbon fiber supplier batch code, change operation code, trigger source order number, instruction status, creation time, and last processing time. The instruction serial number field is an auto-incrementing integer primary key, automatically generated by the database. The industrial control computer performs a data insertion operation, writing the carbon fiber supplier batch code attribute value into the carbon fiber supplier batch code field, the inventory availability status change operation code attribute value into the change operation code field, the change trigger source production task order number attribute value into the trigger source order number field, the instruction execution status attribute value into the instruction status field, and the instruction creation time attribute value into the creation time field. After the insertion operation is completed, the instruction record appears in the batch status change instruction queue list awaiting processing.
[0057] The inventory management database has a built-in timed polling task that executes every 30 seconds. Each time the polling task executes, it performs a query operation on the batch status change instruction queue list, with the query condition being that the instruction status field equals 0. The query operation returns all instruction records with a pending execution status, sorted in ascending order by the instruction serial number field. The polling task then sequentially reads each pending instruction record, retrieving the values from the carbon fiber supplier batch code field and the trigger source order number field.
[0058] For each pending instruction record, the periodic polling task locates the corresponding inventory record in the inventory master table of the inventory management database based on the carbon fiber supplier batch code field. The inventory master table is indexed using the carbon fiber supplier batch code as the primary key and includes fields for carbon fiber supplier batch code, inventory quantity, inventory availability status, storage location code, and last change time. The inventory availability status field can be set to available, restricted, or frozen. For example, in the inventory record with carbon fiber supplier batch code FIBER1202411150003, the inventory quantity field has a value of 850 kg, the inventory availability status field has a value of available, the storage location code field has a value of A zone 03 row 12, and the last change time field has a value of 20241115093000.
[0059] The periodic polling task modifies the inventory availability status field of this inventory record from available to restricted use, and simultaneously updates the last change time field to the current time. This modification is performed by executing a database update statement. The update statement's condition is that the carbon fiber supplier batch code field equals FIBER1202411150003, and the update content is to set the inventory availability status field to restricted use and the last change time field to the current date and time. After the update statement executes successfully, the inventory management database returns one affected row.
[0060] After a successful modification operation in the main inventory table, the scheduled polling task updates the instruction status field of the corresponding instruction record in the batch status change instruction queue from 0 to 2, indicating successful execution. If a database connection error or update conflict occurs during the modification operation, the scheduled polling task updates the instruction status field from 0 to 3, indicating execution failure, and simultaneously writes the error information to the instruction execution log table in the inventory management database for subsequent troubleshooting. At this point, the entire process of generating and executing the instruction to change the available status of the remaining inventory in the same batch is complete. The available status of the remaining 850 kg of carbon fiber corresponding to the supplier batch code FIBER1202411150003 has changed from available to restricted use. This batch of carbon fiber will be restricted from use in subsequent production tasks or will require additional review before it can be used.
[0061] In another preferred embodiment of the present invention, the process of integrating the closed-loop recording module into a batch quality closed-loop record is as follows: Once the status field of the instruction for changing the availability of remaining inventory in the same batch is updated to 2 (indicating successful execution), the industrial control computer initiates the integration operation of the batch quality closed-loop record. The industrial control computer first creates a batch quality closed-loop record data structure in memory. This data structure is a data object containing nine attributes. The first attribute is the production task order number, a 18-digit fixed-length string. The second attribute is the carbon fiber supplier batch code, also an 18-digit fixed-length string. The third attribute is the predecessor historical batch identifier, also an 18-digit fixed-length string. The fourth attribute is the expected quality grade reference value, a variable-length string with values ranging from first-class, second-class, to third-class. The fifth attribute is the original position in the scheduling queue, an integer. The sixth attribute is the adjusted position in the scheduling queue, an integer. The seventh attribute is the finished product inspection conclusion, a variable-length string with values ranging from qualified to unqualified. The 8th attribute is the execution status attribute of the remaining inventory availability status change instruction in the same batch. The data type is a 1-digit fixed-length string, and the value meaning is consistent with the instruction status field in the batch status change instruction queue list. The 9th attribute is the closed-loop record generation time attribute, and the data type is date and time, accurate to the second.
[0062] The industrial control computer extracts the corresponding data from each operational step of this production task and writes it into the aforementioned attributes one by one. The production task order number attribute is assigned the value 20241120000000001234. The carbon fiber supplier batch code attribute is assigned the value FIBER1202411150003.
[0063] The data for the predecessor historical batch identifier attribute originates from the calculation results of the quality prediction module during the current production task receiving phase. When determining the historical batch with the most similar quality fluctuation characteristics, the quality prediction module stores the batch code of the selected historical batch in the memory cache of the industrial control computer, with the current production task order number as the cache key. The industrial control computer reads the predecessor historical batch identifier value from the cache using the current production task order number as the key, for example, FIBER1202409100015, and writes this value into the predecessor historical batch identifier attribute.
[0064] The data for the expected quality grade reference value attribute also comes from the calculation results of the quality prediction module. The quality prediction module reads the finished product quality assessment grade corresponding to the predecessor historical batch identifier FIBER1202409100015 from the finished product inspection data table. The grade read is Grade 1, and the industrial control computer writes Grade 1 into the expected quality grade reference value attribute.
[0065] The data for the original and adjusted position attributes of the scheduling queue originates from the position information recorded by the scheduling shift module during the daily scheduling queue adjustment process. The industrial control computer reads the original position field value (5) and the adjusted position field value (6) corresponding to the current production task order number from the production task basic information table in the production task management database. It writes 5 into the original position attribute of the scheduling queue and 6 into the adjusted position attribute of the scheduling queue.
[0066] The data for the finished product inspection conclusion attribute comes from the summary of the finished product sampling inspection results for this production task. The industrial control computer queries the finished product inspection data table for all individual judgment conclusions corresponding to the production task order number. When all individual judgment conclusions are qualified, the finished product inspection conclusion attribute is assigned the value of qualified; when any individual judgment conclusion is unqualified, the finished product inspection conclusion attribute is assigned the value of unqualified. In this production task, the individual judgment conclusion for the porosity distribution uniformity item is unqualified; therefore, the finished product inspection conclusion attribute is assigned the value of unqualified.
[0067] The data for the execution status attribute of the remaining inventory availability status change instruction in the same batch comes from the instruction status field of the batch status change instruction queue list in the inventory isolation module. The industrial control computer uses the production task order number as the search condition to query the corresponding record in the trigger source order number field of the batch status change instruction queue list, reads the instruction status field value of 2 from that record, and writes 2 into the execution status attribute of the remaining inventory availability status change instruction in the same batch.
[0068] The closed-loop record generation time attribute is assigned the current date and time value, such as 20241120104820.
[0069] The industrial control computer stores the completed batch quality closed-loop record data structure in the batch quality closed-loop record table of the production task management database. The batch quality closed-loop record table is a persistent storage table with a structure containing 10 fields: record serial number, production task order number, carbon fiber supplier batch code, predecessor historical batch identifier, expected quality level reference value, original position in the scheduling queue, adjusted position in the scheduling queue, finished product inspection conclusion, inventory change instruction status, and record generation time. The industrial control computer performs a data insertion operation, writing the nine attribute values of the data structure into the corresponding nine fields. The record serial number field is automatically generated by the database as an auto-incrementing integer. After the insertion operation is completed, the batch quality closed-loop record for this production task is persistently saved in the database. It can be retrieved and queried subsequently using the production task order number or carbon fiber supplier batch code for quality control event retrospective analysis.
[0070] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.
Claims
1. A full-process quality control system for carbon fiber reinforced supercritical foamed materials, characterized in that, include: The task receiving module is used to obtain the production task order number recorded in the production task management database, as well as the associated carbon fiber supplier batch code, polymer matrix supplier batch code, and process formula code. The quality prediction module is used to retrieve historical delivery data records of the same supplier based on the carbon fiber supplier batch code, determine the historical batch with the most similar quality fluctuation characteristics, and extract the finished product quality rating level corresponding to the historical batch as the expected quality level reference value. The traceability construction module is used to retrieve material flow data records by production task order number, obtain the flow sequence and time information of this production task, and construct a material consumption traceability chain containing multiple sequential connection nodes. The scheduling shift module is used to compare the expected quality level reference value with the preset quality level warning value before the time information associated with the first flow link node. When the expected quality level reference value is lower than the preset quality level warning value, the production task order number is shifted backward in the daily scheduling queue. The inventory isolation module is used to obtain the finished product inspection result data of the current production task after the time information associated with the node of the final circulation link. When the result data contains non-conforming items and there is a scheduling queue movement record for the current task, it generates a status change instruction for the remaining inventory of the same batch. The closed-loop record module is used to integrate the batch quality prediction basis, scheduling queue movement record, finished product inspection conclusion and remaining inventory status change of this production task into a batch quality closed-loop record and push it to the production management terminal interface.
2. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 1, characterized in that, In the quality prediction module, the process of determining the historical batch with the most similar quality fluctuation characteristics is as follows: Retrieve the carbon fiber monofilament diameter detection value sequence for each historical batch from the same supplier's historical delivery data records. Perform a piecewise linear fitting operation on each carbon fiber monofilament diameter detection value sequence. The division position of the segment interval is determined by the position where the fluctuation amplitude of three consecutive points exceeds the preset threshold. Extract the inflection point of the slope change rate of each segment fitting line and arrange them in chronological order to form a sequence of inflection points of quality fluctuation trend. Perform the same operation on the carbon fiber monofilament diameter detection value sequence corresponding to the current carbon fiber supplier batch code to obtain the current batch quality fluctuation trend inflection point sequence. Calculate the weighted deviation value between the current batch quality fluctuation trend inflection point sequence and the quality fluctuation trend inflection point sequence of each historical batch based on the difference in the distance and quantity of the inflection point positions. Select the historical batch with the smallest weighted deviation value as the historical batch with the most similar quality fluctuation characteristics.
3. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 2, characterized in that, The carbon fiber monofilament diameter detection value sequence is extracted from the incoming inspection data table corresponding to the carbon fiber supplier batch code. The incoming inspection data table is indexed with the carbon fiber supplier batch code as the primary key. The carbon fiber monofilament diameter detection value sequence is composed of multiple repeated measurement values arranged in the order of measurement. In the piecewise linear fitting operation, the preset threshold value is the third and fourth quartile of the fluctuation amplitude values of three adjacent points in the carbon fiber monofilament diameter detection value sequence of all historical delivered batches of the same supplier. The calculation process of the weighted deviation value is as follows: the time interval distance between each turning point in the current batch quality fluctuation trend turning point sequence and the nearest turning point in the corresponding time neighborhood in the quality fluctuation trend turning point sequence of each historical batch is accumulated to obtain the position deviation sub-value. The absolute value of the difference between the total number of turning points of the current batch and the historical batch is multiplied by the preset weight coefficient to obtain the quantity deviation sub-value. The position deviation sub-value and the quantity deviation sub-value are added to obtain the weighted deviation value.
4. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 1, characterized in that, In the traceability construction module, the process of retrieving material flow data records by production task order number and constructing a material consumption traceability chain containing multiple sequentially connected nodes is as follows: The production task management database stores a material flow data table. The material flow data table includes fields for production task order number, material batch code, flow link code, entry time, and exit time. Using the production task order number as the search condition, all corresponding records in the material flow data table are queried. The query results are sorted in ascending order by the value of the entry time field. The flow link code field in each sorted record is read in sequence and mapped to the feeding link node, mixing link node, injection link node, and molding link node in the order of their first appearance. Create a node data structure for each node, and write the entry time field value and exit time field value of the corresponding record into the node data structure. Connect the first and last nodes in the mapping order to form a material consumption traceability chain. In the auxiliary index storage area of the material consumption traceability chain, establish a bidirectional reference index record between the circulation link node and the carbon fiber supplier batch code.
5. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 1, characterized in that, In the scheduling shift module, the preset quality level warning value is determined as follows: The system retrieves batch quality closed-loop records corresponding to completed production tasks from the production task management database. Completed production tasks are those marked as "completed" in the production task status field. The system extracts the expected quality level reference value and the finished product inspection conclusion value from each batch quality closed-loop record. Batch quality closed-loop records with unqualified finished product inspection conclusion values are selected to form a warning value calibration sample record set. The expected quality level reference value values of each record in the warning value calibration sample record set are arranged in descending order. The expected quality level reference value value corresponding to the preset percentile order in the arrangement sequence is extracted as the preset quality level warning value. At the end of each calendar month, the preset quality level warning value is re-executed based on the newly added batch quality closed-loop records of completed production tasks for that month, and the re-extracted values are updated and stored in the system configuration parameter storage area.
6. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 5, characterized in that, The process of moving the current production task order number backward in the daily scheduling queue in the scheduling shift module is as follows: Obtain the expected quality level reference values corresponding to all pending production task order numbers for the day. Sort all pending production task order numbers for the day in descending order of expected quality level reference values to generate an initial scheduling queue. Extract the current production task order number from the current position in the initial scheduling queue. Insert the extracted current production task order number into the position after all production task order numbers in the initial scheduling queue whose expected quality level reference values are lower than the preset quality level warning value, forming an adjusted scheduling queue. Record the original position number of the current production task order number in the initial scheduling queue and its new position number in the adjusted scheduling queue.
7. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 1, characterized in that, In the inventory isolation module, the specific process for generating the remaining inventory availability status change instruction for the same batch is as follows: A data structure for changing the availability status of remaining inventory in the same batch is created. This data structure includes a carbon fiber supplier batch code field, an inventory availability status change operation code field, a change trigger source production task order number field, and an instruction execution status field. The carbon fiber supplier batch code corresponding to the current production task is written into the carbon fiber supplier batch code field, the restricted use status code is written into the inventory availability status change operation code field, the current production task order number is written into the change trigger source production task order number field, and the pending execution status code is written into the instruction execution status field. The completed data structure is written into the batch status change instruction queue of the inventory management system. The inventory management system polls the batch status change instruction queue and reads instruction records whose instruction execution status field has a pending execution status code. Based on the carbon fiber supplier batch code in the instruction record, the inventory record is located and the inventory availability status modification operation is performed.
8. The whole-process quality control system for carbon fiber reinforced supercritical foamed materials according to claim 1, characterized in that, In the closed-loop recording module, the process of integrating into batch quality closed-loop records is as follows: A batch quality closed-loop record data structure is created, which includes a production task order number field, a carbon fiber supplier batch code field, a precursor historical batch identifier field, an expected quality grade reference value field, a scheduling queue original position field, a scheduling queue adjusted position field, a finished product inspection conclusion field, and a same batch remaining inventory availability status change instruction execution status field. Data corresponding to each operation step of this production task is extracted and written into each field. The precursor historical batch batch code is written into the precursor historical batch identifier field, the position number before scheduling queue movement is written into the scheduling queue original position field, the position number after scheduling queue movement is written into the scheduling queue adjusted position field, and the generation status of the same batch remaining inventory availability status change instruction is written into the same batch remaining inventory availability status change instruction execution status field. This data structure is stored in the batch quality closed-loop record table of the production task management database.