Intelligent brain management system based on glass factory data
The intelligent brain management system has solved the problem of lagging production scheduling and order management in glass production, realizing production scheduling optimization and real-time feedback, and improving the accuracy and efficiency of production planning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU POLY GLASS NAT INSPECTION GLASS RES INST CO LTD
- Filing Date
- 2024-06-17
- Publication Date
- 2026-06-12
AI Technical Summary
In the existing glass production process, production scheduling and order management rely on manual operation, which makes it impossible to fully optimize business objectives, results in large time lags, and causes significant deviations between calculated results and actual results, making it impossible to provide real-time feedback and track progress.
The system employs a smart brain management system based on glass factory data, including a data acquisition and classification module, an intelligent production module, a production scheduling module, and an order placement and delivery module. It adjusts the production schedule through order priority calculation and quality feedback, and generates reports in conjunction with the data analysis module to achieve production optimization and intelligent scheduling.
It enables real-time feedback and tracking of production scheduling and order management, optimizes production progress display, and improves the accuracy and efficiency of production planning.
Smart Images
Figure CN118839883B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent Internet of Things (IoT) control technology, and in particular to a smart brain management system based on glass factory data. Background Technology
[0002] In the glass production process, from order entry to final production scheduling, information is involved at multiple stages. Decisions require a comprehensive consideration of various business objectives. Planning is the most crucial link in the manufacturing process; all information is aggregated at this stage to generate a time-based match between each order and each production line. Much production scheduling is still executed manually, sometimes only once a week. With massive amounts of information, this makes it impossible to fully optimize business objectives. Furthermore, time lags lead to mismatches between plans and execution. This necessitates significant human intervention or the neglect of certain constraints, resulting in a large discrepancy between the final calculated results and actual performance, failing to adequately meet actual needs.
[0003] Currently, the entire industry's management largely relies on MES systems, namely Manufacturing Execution Systems, which are production information management systems for the shop floor execution layer of manufacturing enterprises. MES systems can provide enterprises with management modules including manufacturing data management, production planning and scheduling management, production scheduling management, inventory management, quality management, human resource management, work center / equipment management, tooling management, procurement management, cost management, project dashboard management, production process control, lower-level data integration and analysis, and upper-level data integration and decomposition, creating a robust, reliable, comprehensive, and feasible manufacturing collaborative management platform for enterprises.
[0004] As the amount of data and production status increase in the MES system, it is no longer able to effectively reflect the management of production scheduling, layout, or subsequent orders. If problems are encountered, real-time feedback is not possible. How can we comprehensively display the progress and various situations encountered? Therefore, it is necessary to summarize the production line, orders, or other situations to better track them. Summary of the Invention
[0005] This invention addresses the shortcomings of existing technologies by providing a smart brain management system based on glass factory data.
[0006] To solve the above problems, the present invention provides the following technical solution:
[0007] A smart brain management system based on glass factory data includes a data acquisition and classification module, an intelligent production module, a production scheduling module, a data analysis module, and an order scheduling and delivery module.
[0008] The data acquisition and classification module acquires inventory data and a set of production orders, parses the set of production orders, and obtains relevant data for each order. The relevant data includes the unique identifier of the order, order purchase data, order production data, order work section type, materials required for each work section, and preset completion time for each work section.
[0009] The intelligent production module includes a scheduling unit and a scheduling optimization unit. The scheduling unit schedules all production orders based on the unique order identifier, order section type, and preset completion time for each section, combined with inventory data, order procurement data, and order production data. It also calculates the order priority of each order based on the unique order identifier and obtains the scheduling result based on the order priority. The scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set based on the order priority of each order or the quality feedback of the order section type, thereby obtaining the scheduling optimization result.
[0010] The production scheduling module includes a scheduling monitoring unit and a production scheduling unit. The scheduling monitoring unit is used to acquire the progress of each type of work section in real time for each time period and monitor the progress of each time period according to preset influencing factors. The production scheduling unit adjusts the progress of the corresponding work section type in the corresponding time period according to the production scheduling results and production scheduling optimization results to realize intelligent production scheduling.
[0011] The data analysis module is used to manage and analyze order-related data, order procurement data, order production data, and inventory data, and generate reports.
[0012] The order scheduling and delivery module is used to ship all orders sequentially according to the production schedule completion status.
[0013] As one possible implementation, the production scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set according to the order priority of each order, including the following steps:
[0014] Monitor whether a target production order has been added. If so, obtain the relevant data of the target production order and its order priority.
[0015] Based on the relevant data of the target production order, determine whether the materials required for all order section types meet the demand, and sort the difference between the materials required for each order section type and the actual materials according to the preset sorting model to obtain the sorting results;
[0016] Based on the sorting results and order priority, a production schedule is created for all order segments of the new production orders, and adjustments and optimizations are made to the orders that are currently in production and those that have not yet been produced in the production order set.
[0017] As one possible implementation, the order priority is calculated using a priority calculation model, which is expressed as follows:
[0018]
[0019] Among them, V i ε represents the influencing factor of production orders. i λ represents the weight of the impact factor, Rev represents the adjusted score, and λ represents the weight of the impact factor. i Y represents the total score of priority, N represents the total number of production orders, and i represents the i-th production order.
[0020] As one possible implementation method, the order process includes at least one or more of the following: glass arrangement, glass cutting, glass edging, glass sorting, glass tempering, glass packaging, glass assembly, and others.
[0021] The quality feedback for each order section should at least include the quality feedback obtained based on the quality inspection results of glass cutting or glass edging.
[0022] As one possible implementation method, the quality inspection results of glass cutting or glass edging are obtained through the following steps:
[0023] A glass image set is obtained after the glass is cut or the glass is edged. The glass image set includes multi-angle bright field images of the glass and one-to-one corresponding multi-angle dark field images of the glass. The multi-angle bright field images and multi-angle dark field images of the glass are fused to obtain a multi-angle fused image of the glass. The glass background is a dark background.
[0024] The multi-angle fused image of the glass is preprocessed to obtain a binarized fused image. Based on the binarized fused image, it is determined whether there is a crack and the crack feature information is extracted to obtain crack feature information, wherein the crack feature information includes crack position relationship and crack length.
[0025] The crack feature information is analyzed, and the crack length is obtained based on the set of endpoints and the set of intersections. The existence of crack initiation points is then searched through the set of intersections.
[0026] If not, the location relationship of the cracks is used to determine whether they are isolated cracks on each piece of glass, and then the quality inspection results are obtained.
[0027] As one possible implementation, analyzing the crack feature information to obtain the crack direction and crack length includes the following steps:
[0028] Based on crack feature information, the largest circumcircle is fitted to obtain the main image of the crack and the main structure of the crack is extracted.
[0029] Crack analysis is performed on the main crack structure to obtain the set of endpoints and the set of intersections for each crack. The set of endpoints is the set of all crack endpoints, and the set of intersections is the set of points where cracks intersect.
[0030] Connect each intersection point to its corresponding endpoint along the path of each crack to form the main crack path. The length of the main crack path is the crack length.
[0031] Starting from each intersection point, traverse all intersection points to obtain the intersection point with the most main crack paths. Using the intersection point as the center, search for whether there is a rectangular connected region and determine whether the rectangular connected region exceeds a preset threshold. If the condition is met, the intersection point is the crack initiation point.
[0032] As one possible implementation, the method of determining whether a crack is an isolated crack on each piece of glass by the location relationship of the cracks includes the following steps:
[0033] The crack location relationship refers to the position of the crack on each piece of glass;
[0034] If the crack is located in the middle of each piece of glass and the edge of the crack does not extend beyond the edge of each piece of glass, then the crack is an isolated crack, and the current glass is judged to have a quality problem.
[0035] If a crack extends beyond the edge of each piece of glass, the positional relationship between the cracks in the previous or next piece of glass is used to determine whether there are related cracks. If so, the crack in the current glass is considered a non-isolated crack. The number of non-isolated cracks in the glass is statistically analyzed. If the number exceeds a preset threshold, the glass before cutting is considered to have quality problems.
[0036] As one possible implementation method, adjustments and optimizations are made to the production order set for both in-production and out-of-production orders based on the quality feedback from different order segments, thereby obtaining optimized production scheduling results. This includes the following steps:
[0037] Receive the quality inspection results and determine whether the quality inspection results exceed the quality inspection score threshold. If so, inspect the glass in the glass cutting or glass edging section. If the glass is normal, inspect the process in the glass cutting or glass edging section to determine whether any abnormality has occurred.
[0038] If an anomaly occurs, the orders in production and those not yet produced in the production order set will be adjusted and optimized, and the orders in production and those not yet produced will be sorted according to urgency, importance and delivery time to obtain a second sorting result;
[0039] The second sorting results are filtered and sorted based on the glass cutting or glass edging process to obtain the third sorting results.
[0040] As one possible implementation method, intelligent production scheduling includes the following steps:
[0041] Production of the corresponding order will be executed first based on the third sorting result;
[0042] If the process abnormalities in the glass cutting or glass edging section have been eliminated, then production of the corresponding order will be executed according to the second sorting result.
[0043] As one possible implementation, the data analysis module includes a data management unit and a data intelligent analysis unit;
[0044] The data management unit manages and applies order-related data, order procurement data, order production data, and inventory data.
[0045] The data intelligent analysis unit analyzes order-related data, order procurement data, order production data, and inventory data, and feeds the analysis results back to the intelligent production module. It also generates reports based on the analysis results and sends them back to the corresponding client.
[0046] As one possible implementation, the intelligent production module further includes an intelligent film arrangement unit;
[0047] The intelligent scheduling unit is configured as follows:
[0048] The shape and size of the glass sheet are obtained, and the presence of defects in the glass sheet is determined based on the glass defect model. If defects exist, the location and size of the defects are marked, and the usable and unusable areas of the glass sheet are obtained based on the defect area.
[0049] Obtain the parameter information of the glass to be laid out, including at least the size information. Lay out the glass to be laid out using a glass laying model and in combination with the defect area. Specifically, a glass laying pre-training model is constructed based on a genetic algorithm, trained on a glass laying dataset, and optimized using an ant colony algorithm to obtain the glass laying model.
[0050] This invention, by adopting the above technical solutions, has significant technical effects:
[0051] The system and method of this invention can effectively reflect the management of production scheduling, layout, or subsequent orders. If problems are encountered, real-time feedback can be provided, and the progress and various situations encountered can be comprehensively displayed for better tracking. Attached Figure Description
[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0053] Figure 1 This is an overall schematic diagram of the system of the present invention;
[0054] Figure 2 This is a flowchart illustrating the method of the present invention;
[0055] Figures 3-6 This is a flowchart illustrating several embodiments of the present invention. Detailed Implementation
[0056] The present invention will be further described in detail below with reference to the embodiments. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.
[0057] Example 1:
[0058] A smart brain management system based on glass factory data, such as Figure 1 As shown, it includes a data acquisition and classification module 100, an intelligent production module 200, a production scheduling module 300, a data analysis module 400, and an order scheduling and delivery module 500.
[0059] A smart brain management system based on glass factory data includes a data acquisition and classification module, an intelligent production module, a production scheduling module, a data analysis module, and an order scheduling and delivery module.
[0060] The data acquisition and classification module acquires inventory data and a set of production orders, parses the set of production orders, and obtains relevant data for each order. The relevant data includes the unique identifier of the order, order purchase data, order production data, order work section type, materials required for each work section, and preset completion time for each work section.
[0061] The intelligent production module includes a scheduling unit and a scheduling optimization unit. The scheduling unit schedules all production orders based on the unique order identifier, order section type, and preset completion time for each section, combined with inventory data, order procurement data, and order production data. It also calculates the order priority of each order based on the unique order identifier and obtains the scheduling result based on the order priority. The scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set based on the order priority of each order or the quality feedback of the order section type, thereby obtaining the scheduling optimization result.
[0062] The production scheduling module includes a scheduling monitoring unit and a production scheduling unit. The scheduling monitoring unit is used to acquire the progress of each type of work section in real time for each time period and monitor the progress of each time period according to preset influencing factors. The production scheduling unit adjusts the progress of the corresponding work section type in the corresponding time period according to the production scheduling results and production scheduling optimization results to realize intelligent production scheduling.
[0063] The data analysis module is used to manage and analyze order-related data, order procurement data, order production data, and inventory data, and generate reports.
[0064] The order scheduling and delivery module is used to ship all orders sequentially according to the production schedule completion status.
[0065] This invention can be understood as an upgrade and improvement of the MES system. Through the system of this invention, the management status of production scheduling, layout or subsequent orders can be well reflected. If problems are encountered, real-time feedback can be provided. The progress and various situations encountered can be fully displayed to achieve better tracking.
[0066] Compared to previous digital dashboard products or general MES systems, the system of this invention schedules production based on multiple dimensions such as order placement time, delivery time, and production process, enabling online data flow. Furthermore, the system emphasizes production scheduling optimization and adjustment, allowing real-time monitoring of glass status in core processes, particularly glass cutting and edging. It conducts comprehensive quality checks on the glass, providing feedback on quality and using this feedback to optimize and adjust the production schedule. The system also acquires real-time progress data for each process type at each time period and monitors progress based on preset influencing factors. Additionally, it manages and analyzes order-related data, order procurement data, order production data, and inventory data, generating reports. Finally, it can ship all orders sequentially based on production schedule completion status.
[0067] In one embodiment, the production scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set according to the order priority of each order, including the following steps:
[0068] Monitor whether a target production order has been added. If so, obtain the relevant data of the target production order and its order priority.
[0069] Based on the relevant data of the target production order, determine whether the materials required for all order section types meet the demand, and sort the difference between the materials required for each order section type and the actual materials according to the preset sorting model to obtain the sorting results;
[0070] Based on the sorting results and order priority, a production schedule is created for all order segments of the new production orders, and adjustments and optimizations are made to the orders that are currently in production and those that have not yet been produced in the production order set.
[0071] Specifically, the order priority is calculated using a priority calculation model, which is expressed as follows:
[0072]
[0073] Among them, V i Factors influencing production orders, such as customer waiting time, expected future revenue, and order value, are represented by ε. i λ represents the weight of the impact factor, Rev represents the adjusted score, and λ represents the weight of the impact factor. i Let λ represent the adjustment weights, Y represent the total priority score, N represent the total number of production orders, and i represent the i-th production order. Rev1 is the first adjustment score, which can be assigned based on the importance of each production order as measured by an artificial intelligence algorithm model; Rev2 is the second adjustment score, which can be assigned based on the importance of the corresponding production order. Furthermore, the first and second adjustment weights must satisfy λ1. i +λ2 i =1.
[0074] In one embodiment, the order process includes at least one or more of the following: glass arrangement, glass cutting, glass edging, glass sorting, glass tempering, glass packaging, glass assembly, and others.
[0075] The quality feedback for each order section should at least include the quality feedback obtained based on the quality inspection results of glass cutting or glass edging.
[0076] The quality inspection results of glass cutting or glass edge grinding are obtained through the following steps:
[0077] A glass image set is obtained after the glass is cut or the glass is edged. The glass image set includes multi-angle bright field images of the glass and one-to-one corresponding multi-angle dark field images of the glass. The multi-angle bright field images and multi-angle dark field images of the glass are fused to obtain a multi-angle fused image of the glass. The glass background is a dark background.
[0078] The multi-angle fused image of the glass is preprocessed to obtain a binarized fused image. Based on the binarized fused image, it is determined whether there is a crack and the crack feature information is extracted to obtain crack feature information, wherein the crack feature information includes crack position relationship and crack length.
[0079] The crack feature information is analyzed, and the crack length is obtained based on the set of endpoints and the set of intersections. The existence of crack initiation points is then searched through the set of intersections.
[0080] If not, the location relationship of the cracks is used to determine whether they are isolated cracks on each piece of glass, and then the quality inspection results are obtained.
[0081] Specifically, the analysis of the crack feature information to obtain the crack direction and crack length includes the following steps:
[0082] Based on crack feature information, the largest circumcircle is fitted to obtain the main image of the crack and the main structure of the crack is extracted.
[0083] Crack analysis is performed on the main crack structure to obtain the set of endpoints and the set of intersections for each crack. The set of endpoints is the set of all crack endpoints, and the set of intersections is the set of points where cracks intersect.
[0084] Connect each intersection point to its corresponding endpoint along the path of each crack to form the main crack path. The length of the main crack path is the crack length.
[0085] Starting from each intersection point, traverse all intersection points to obtain the intersection point with the most main crack paths. Using the intersection point as the center, search for whether there is a rectangular connected region and determine whether the rectangular connected region exceeds a preset threshold. If the condition is met, the intersection point is the crack initiation point.
[0086] Specifically, in one embodiment, determining whether a crack is an isolated crack on each piece of glass based on its location relationship includes the following steps:
[0087] The crack location relationship refers to the position of the crack on each piece of glass;
[0088] If the crack is located in the middle of each piece of glass and the edge of the crack does not extend beyond the edge of each piece of glass, then the crack is an isolated crack, and the current glass is judged to have a quality problem.
[0089] If a crack extends beyond the edge of each piece of glass, the positional relationship between the cracks in the previous or next piece of glass is used to determine whether there are related cracks. If so, the crack in the current glass is considered a non-isolated crack. The number of non-isolated cracks in the glass is statistically analyzed. If the number exceeds a preset threshold, the glass before cutting is considered to have quality problems.
[0090] The production scheduling optimization results are obtained by adjusting and optimizing the production order set based on the quality feedback of the order section type, including the following steps:
[0091] Receive the quality inspection results and determine whether the quality inspection results exceed the quality inspection score threshold. If so, inspect the glass in the glass cutting or glass edging section. If the glass is normal, inspect the process in the glass cutting or glass edging section to determine whether any abnormality has occurred.
[0092] If an anomaly occurs, the orders in production and those not yet produced in the production order set will be adjusted and optimized, and the orders in production and those not yet produced will be sorted according to urgency, importance and delivery time to obtain a second sorting result;
[0093] The second sorting results are filtered and sorted based on the glass cutting or glass edging process to obtain the third sorting results.
[0094] After obtaining the first sorting result, the second sorting result, and the third sorting result, intelligent production scheduling is implemented, including the following steps:
[0095] Production of the corresponding order will be executed first based on the third sorting result;
[0096] If the process abnormalities in the glass cutting or glass edging section have been eliminated, then production of the corresponding order will be executed according to the second sorting result.
[0097] In one embodiment, the data analysis module includes a data management unit and a data intelligent analysis unit;
[0098] The data management unit manages and applies order-related data, order procurement data, order production data, and inventory data; the data intelligent analysis unit analyzes the order-related data, order procurement data, order production data, and inventory data, feeds the analysis results back to the intelligent production module, and generates reports from the analysis results and feeds them back to the corresponding client.
[0099] In addition, to enable better film arrangement, the intelligent production module also includes an intelligent film arrangement unit; the intelligent film arrangement unit is configured as follows:
[0100] The shape and size of the glass sheet are obtained, and the presence of defects in the glass sheet is determined based on the glass defect model. If defects exist, the location and size of the defects are marked, and the usable and unusable areas of the glass sheet are obtained based on the defect area.
[0101] Obtain the parameter information of the glass to be laid out, including at least the size information. Lay out the glass to be laid out using a glass laying model and in combination with the defect area. Specifically, a glass laying pre-training model is constructed based on a genetic algorithm, trained on a glass laying dataset, and optimized using an ant colony algorithm to obtain the glass laying model.
[0102] In this embodiment, scheduling is combined with artificial intelligence. The optimal scheduling solution is found through artificial intelligence, and then scheduling production is executed.
[0103] Example 2:
[0104] A smart brain management method based on glass factory data, implemented using a management system, includes a data acquisition and classification module, an intelligent production module, a production scheduling module, a data analysis module, and an order scheduling and delivery module, such as... Figure 2 As shown, it includes the following steps:
[0105] S100: The data acquisition and classification module acquires inventory data and production order set, parses the production order set, and obtains relevant data for each order. The relevant data includes the order's unique identifier, order purchase data, order production data, order work section type, materials required for each work section, and preset completion time for each work section.
[0106] S200, the intelligent production module includes a scheduling unit and a scheduling optimization unit. The scheduling unit schedules all production orders based on the unique order identifier, the type of work segment, and the preset completion time for each work segment, combined with inventory data, order procurement data, and order production data. It also calculates the order priority of each order based on the unique order identifier and obtains the scheduling result based on the order priority. The scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set based on the order priority of each order or the quality feedback of the type of work segment, thereby obtaining the scheduling optimization result.
[0107] S300, the production scheduling module includes a scheduling monitoring unit and a production scheduling unit. The scheduling monitoring unit is used to obtain the progress of each type of work section in real time in each time period and monitor the progress of each time period according to preset influencing factors. The production scheduling unit adjusts the progress of the corresponding work section in the corresponding time period according to the production scheduling results and production scheduling optimization results to realize intelligent production scheduling.
[0108] The S500 data analysis module manages and analyzes order-related data, order procurement data, order production data, and inventory data, and generates reports.
[0109] The S500 order scheduling and delivery module delivers all orders sequentially based on the completion status of the production schedule.
[0110] In one embodiment, such as Figure 3 As shown, it also includes the following steps:
[0111] S310. Monitor whether a target production order has been added. If so, obtain the relevant data of the target production order and the order priority of the target production order.
[0112] S320. Based on the relevant data of the target production order, determine whether the materials required for all order section types meet the demand, and sort the difference between the materials required for each order section type and the actual materials according to the preset sorting model to obtain the sorting result;
[0113] S330. Based on the sorting results and order priority, schedule production for all order segments of new production orders, and adjust and optimize the orders that are being produced and those that have not yet been produced in the production order set.
[0114] In one embodiment, the order process includes at least one or more of the following: glass cutting, glass edging, glass sorting, glass tempering, glass packaging, glass assembly, and others. Therefore, it also includes quality inspection of the glass after cutting or edging. That is, after quality inspection, the production schedule will be adjusted or optimized, such as... Figure 4 As shown, specifically:
[0115] S340. Obtain a glass image set after the glass is cut or the glass is edged, wherein the glass image set includes glass multi-angle bright field images and corresponding glass multi-angle dark field images, and fuse the glass multi-angle bright field images and glass multi-angle dark field images to obtain a glass multi-angle fused image, wherein the glass background is a dark background.
[0116] S350. Preprocess the multi-angle fused image of the glass to obtain a binary fused image. Based on the binary fused image, determine whether there is a crack and extract it to obtain crack feature information, wherein the crack feature information includes the crack position relationship and crack length.
[0117] S360. Analyze the crack feature information, obtain the crack length based on the set of endpoints and the set of intersections, and find out whether there is a crack initiation point through the set of intersections.
[0118] S370. If not, determine whether it is an isolated crack on each piece of glass by the relationship between the crack positions.
[0119] In step S360, the crack feature information is analyzed to obtain the crack direction and crack length, such as... Figure 5 As shown, it includes the following steps:
[0120] S361. Based on crack feature information, fit the largest circumcircle to obtain the main image of the crack and extract the main structure of the crack;
[0121] S362. Analyze the crack situation of the main crack structure to obtain the set of endpoints and the set of intersections of each crack, wherein the set of endpoints is the set of all crack endpoints and the set of intersections is the set of intersections between cracks.
[0122] S363. Connect each intersection point with its corresponding endpoint along the path of each crack to form the main crack path. The length of the main crack path is the crack length.
[0123] S364. Starting from each intersection point, traverse all intersection points to obtain the intersection point with the most main crack paths. Using the intersection point as the center, search for whether there is a rectangular connected region and determine whether the rectangular connected region exceeds a preset threshold. If the condition is met, the intersection point is the crack occurrence point.
[0124] In one embodiment, step S370 involves determining whether a crack is an isolated crack on each piece of glass based on the location relationship of the cracks. Figure 6 As shown, it includes the following steps:
[0125] S371, The crack location relationship refers to the position of the crack on each piece of glass;
[0126] S372. If the crack is located in the middle of each piece of glass and the edge of the crack does not extend beyond the edge of each piece of glass, then the crack is an isolated crack, and it is determined that the current glass has a quality problem.
[0127] S373. If the crack extends beyond the edge of each piece of glass, then based on the positional relationship of the cracks in the previous or next piece of glass, determine whether there are related cracks. If so, the crack in the current glass is a non-isolated crack. Then, count the number of non-isolated cracks in the glass. If the count exceeds a preset threshold, then the glass before cutting is judged to have quality problems.
[0128] Various changes and modifications made without departing from the spirit and scope of this invention, and all equivalent technical solutions, also fall within the scope of this invention.
[0129] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0130] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0131] This invention is described with reference to flowchart illustrations and / or block diagrams of the method, terminal device (system), and computer program product according to the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0132] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0134] It should be noted that:
[0135] The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.
[0136] Furthermore, it should be noted that the shapes and names of the parts and components described in the specific embodiments described in this specification may differ. All equivalent or simple variations made to the structure, features, and principles described in this patent concept are included within the protection scope of this patent. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to replace them, as long as they do not depart from the structure of this invention or exceed the scope defined in these claims, they should all fall within the protection scope of this invention.
Claims
1. A smart brain management system based on glass factory data, characterized in that, It includes a data acquisition and classification module, an intelligent production module, a production scheduling module, a data analysis module, and an order scheduling and delivery module; The data acquisition and classification module acquires inventory data and a set of production orders, parses the set of production orders, and obtains relevant data for each order. The relevant data includes the unique identifier of the order, order purchase data, order production data, order work section type, materials required for each work section, and preset completion time for each work section. The intelligent production module includes a scheduling unit and a scheduling optimization unit. The scheduling unit schedules all production orders based on the unique order identifier, order section type, and preset completion time for each section, combined with inventory data, order procurement data, and order production data. It also calculates the order priority of each order based on the unique order identifier and obtains the scheduling result based on the order priority. The scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set based on the order priority of each order or the quality feedback of the order section type, thereby obtaining the scheduling optimization result. The quality feedback for each order section type should at least include quality feedback obtained based on the quality inspection results of glass cutting or glass edging. The quality inspection results for glass cutting or glass edge grinding are obtained through the following steps: A glass image set is obtained after the glass is cut or the glass is edged. The glass image set includes multi-angle bright field images of the glass and one-to-one corresponding multi-angle dark field images of the glass. The multi-angle bright field images and multi-angle dark field images of the glass are fused to obtain a multi-angle fused image of the glass. The glass background is a dark background. The multi-angle fused image of the glass is preprocessed to obtain a binarized fused image. Based on the binarized fused image, it is determined whether there is a crack and the crack feature information is extracted to obtain crack feature information, wherein the crack feature information includes crack position relationship and crack length. The crack feature information is analyzed, and the crack length is obtained based on the set of endpoints and the set of intersections. The existence of crack initiation points is then searched through the set of intersections. If not, the location relationship of the cracks is used to determine whether they are isolated cracks on each piece of glass, and then the quality inspection results are obtained. The analysis of the crack feature information to obtain the crack direction and crack length includes the following steps: Based on crack feature information, the largest circumcircle is fitted to obtain the main image of the crack and the main structure of the crack is extracted. Crack analysis is performed on the main crack structure to obtain the set of endpoints and the set of intersections for each crack. The set of endpoints is the set of all crack endpoints, and the set of intersections is the set of points where cracks intersect. Connect each intersection point to its corresponding endpoint along the path of each crack to form the main crack path. The length of the main crack path is the crack length. Starting from each intersection point, traverse all intersection points to obtain the intersection point with the most main crack paths. Using the intersection point as the center, search for whether there is a rectangular connected region and determine whether the rectangular connected region exceeds a preset threshold. If the condition is met, the intersection point is the crack initiation point. The production scheduling module includes a scheduling monitoring unit and a production scheduling unit. The scheduling monitoring unit is used to acquire the progress of each type of work section in real time for each time period and monitor the progress of each time period according to preset influencing factors. The production scheduling unit adjusts the progress of the corresponding work section type in the corresponding time period according to the production scheduling results and production scheduling optimization results to realize intelligent production scheduling. The data analysis module is used to manage and analyze order-related data, order procurement data, order production data, and inventory data, and generate reports. The order scheduling and delivery module is used to ship all orders sequentially according to the production schedule completion status.
2. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, The production scheduling optimization unit adjusts and optimizes the orders in production and those not yet produced in the production order set according to the order priority of each order, including the following steps: Monitor whether a target production order has been added. If so, obtain the relevant data of the target production order and its order priority. Based on the relevant data of the target production order, determine whether the materials required for all order section types meet the demand, and sort the difference between the materials required for each order section type and the actual materials according to the preset sorting model to obtain the sorting results; Based on the sorting results and order priority, a production schedule is created for all order segments of the new production orders, and adjustments and optimizations are made to the orders that are currently in production and those that have not yet been produced in the production order set.
3. The intelligent brain management system based on glass factory data according to claim 1 or 2, characterized in that, The order priority is calculated using a priority calculation model, which is expressed as follows: in, Indicating the factors affecting production orders, This indicates the weight of the influencing factor. This indicates an adjustment to the score. This indicates an adjustment of the weights. The total score indicates the priority. This indicates the total number of production orders. Indicates the first One production order.
4. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, The order process includes at least one or more of the following: glass arrangement, glass cutting, glass edging, glass racking, glass tempering, glass packaging, glass assembly, and others.
5. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, The method for determining whether a crack is an isolated crack on each piece of glass by analyzing the location relationship of the cracks includes the following steps: The crack location relationship refers to the position of the crack on each piece of glass; If the crack is located in the middle of each piece of glass and the edge of the crack does not extend beyond the edge of each piece of glass, then the crack is an isolated crack, and the current glass is judged to have a quality problem. If a crack extends beyond the edge of each piece of glass, the positional relationship between the cracks in the previous or next piece of glass is used to determine whether there are related cracks. If so, the crack in the current glass is considered a non-isolated crack. The number of non-isolated cracks in the glass is statistically analyzed. If the number exceeds a preset threshold, the glass before cutting is considered to have quality problems.
6. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, Based on the quality feedback from different order stages, the production order set is adjusted and optimized for both in-production and out-of-production orders to obtain production scheduling optimization results. This includes the following steps: Receive the quality inspection results and determine whether the quality inspection results exceed the quality inspection score threshold. If so, inspect the glass in the glass cutting or glass edging section. If the glass is normal, inspect the process in the glass cutting or glass edging section to determine whether any abnormality has occurred. If an anomaly occurs, the orders in production and those not yet produced in the production order set will be adjusted and optimized, and the orders in production and those not yet produced will be sorted according to urgency, importance and delivery time to obtain a second sorting result; The second sorting results are filtered and sorted based on the glass cutting or glass edging process to obtain the third sorting results.
7. The intelligent brain management system based on glass factory data according to claim 6, characterized in that, To achieve intelligent production scheduling, the following steps are included: Production of the corresponding order will be executed first based on the third sorting result; If the process abnormalities in the glass cutting or glass edging section have been eliminated, then production of the corresponding order will be executed according to the second sorting result.
8. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, The data analysis module includes a data management unit and a data intelligent analysis unit; The data management unit manages and applies order-related data, order procurement data, order production data, and inventory data. The data intelligent analysis unit analyzes order-related data, order procurement data, order production data, and inventory data, and feeds the analysis results back to the intelligent production module. It also generates reports based on the analysis results and sends them back to the corresponding client.
9. The intelligent brain management system based on glass factory data according to claim 1, characterized in that, The intelligent production module also includes an intelligent film scheduling unit; The intelligent scheduling unit is configured as follows: The shape and size of the glass sheet are obtained, and the presence of defects in the glass sheet is determined based on the glass defect model. If defects exist, the location and size of the defects are marked, and the usable and unusable areas of the glass sheet are obtained based on the defect area. Obtain the parameter information of the glass to be laid out, including at least the size information. Lay out the glass to be laid out using a glass laying model and in combination with the defect area. Specifically, a glass laying pre-training model is constructed based on a genetic algorithm, trained on a glass laying dataset, and optimized using an ant colony algorithm to obtain the glass laying model.