US-20260126784-A1 - METHODS, SYSTEMS, AND STORAGE MEDIA FOR ASSEMBLY OPERATION CONTROL BASED ON INDUSTRIAL INTERNET OF THINGS

Abstract

Provide are a method, a system, and a storage medium for assembly operation control. The method includes: obtaining quality inspection information of a plurality of parts to be assembled; determining, based on the quality inspection information and assembly information, an assembly risk value of each of at least one assembly process through an assembly database; generating a first assembly parameter in response to determining that the assembly risk value satisfies a risk condition; obtaining assembly process data of an assembly operation based on a monitoring device; determining assembly quality of a completed process; in response to determining that the assembly quality does not satisfy a quality requirement, determining a second assembly parameter; generating quality warning information based on the assembly quality and the second assembly parameter; and generating, based on the completed process, quality update data for updating the assembly database.

Inventors

• Hanshu SHAO

Assignees

• CHENGDU QINCHUAN IOT TECHNOLOGY CO., LTD.

Dates

Publication Date

20260507

Application Date

20260107

Priority Date

20241125

Claims (20)

1. 1 . A system for assembly operation control, comprising an Industrial Internet of Things (IIoT) management platform, an IIoT sensing network platform, and an IIoT sensing and control platform, wherein the IIoT management platform is configured to: obtain, based on the IIoT sensing network platform, quality inspection information of a plurality of parts to be assembled through a quality inspection device of the IIoT sensing and control platform; generate a plurality of candidate assembly groups based on assembly information, wherein the assembly information includes at least one assembly process, and an assembly object, a position to be assembled, an interference relationship, an assembly position, and a connection relationship corresponding to each of the at least one assembly process; for each of the plurality of candidate assembly groups, determine, based on quality inspection information of a plurality of parts to be assembled in the candidate assembly groups, a plurality of reference assembly records through an assembly database, wherein the assembly database is configured in a data center of the IIoT management platform; determine, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group; determine an assembly risk value of each of at least one assembly process based on predicted assembly quality of the plurality of candidate assembly groups; determine that the assembly risk value satisfies a risk condition; generate a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition, wherein the first assembly parameter includes an assembly grouping parameter and an assembly sequence parameter; and send the first assembly parameter to an operation device of the IIoT sensing and control platform, and control the operation device to perform an assembly operation on the plurality of parts to be assembled based on the assembly grouping parameter and the assembly sequence parameter.
2. 2 . The system of claim 1 , wherein the IIoT management platform is further configured to: determine a set of parts based on a type of the plurality of parts to be assembled; determine a plurality of sub-sets of parts based on the quality inspection information of the plurality of parts to be assembled in the set of parts, a count of the plurality of sub-sets of parts being related to the quality inspection information of the plurality of parts to be assembled in the set of parts; and determine the plurality of candidate assembly groups based on the plurality of sub-sets of parts.
3. 3 . The system of claim 2 , wherein the count of the plurality of sub-sets of parts is further related to a reference fit tolerance range of a fit part.
4. 4 . The system of claim 1 , wherein the IIoT management platform is further configured to: for each of the plurality of reference assembly records, obtain a machining accuracy of a plurality of historical assembly parts in the reference assembly record; determine a matching degree between the reference assembly record and a candidate assembly group corresponding to the reference assembly record based on the machining accuracy and an impact value of the plurality of historical assembly parts on a finished product assembly quality; and determine the predicted assembly quality of the candidate assembly group based on the matching degree and historical assembly quality in the plurality of reference assembly records.
5. 5 . The system of claim 1 , wherein the IIoT management platform is further configured to: determine a fit part based on a historical assembly record of the plurality of parts to be assembled; and generate the first assembly parameter based on the fit part.
6. 6 . The system of claim 5 , wherein the IIoT management platform is further configured to: generate a plurality of candidate assembly sequences based on the assembly information; for each of the plurality of candidate assembly sequences, construct, based on a plurality of sets of parts, a plurality of sub-sets of parts, and the candidate assembly sequence, an assembly map corresponding to the candidate assembly sequence; determine, based on the assembly map corresponding to the candidate assembly sequence, a predicted fit tolerance corresponding to the candidate assembly sequence through a fit model, the fit model being a machine learning model; and determine the assembly sequence parameter based on a plurality of predicted fit tolerances corresponding to the plurality of candidate assembly sequences.
7. 7 . The system of claim 6 , wherein an input of the fit model includes assembly environment information.
8. 8 . The system of claim 6 , wherein the IIoT management platform is further configured to: in response to determining that the assembly quality satisfies a predetermined training condition, determine an incremental training set based on a historical assembly record corresponding to the assembly quality; and perform an incremental update on the fit model based on the incremental training set, wherein a learning rate of the fit model during the incremental update is lower than a learning rate of the fit model before the incremental update.
9. 9 . The system of claim 8 , wherein the IIoT management platform is further configured to: train the fit model based on labeled training samples, wherein the training samples include sample assembly maps, and labels include actual fit tolerances corresponding to the sample assembly maps.
10. 10 . The system of claim 1 , wherein the IIoT management platform is further configured to: obtain assembly process data of the assembly operation based on a monitoring device of the IIoT sensing and control platform; determine assembly quality of a completed process based on the assembly process data; in response to determining that the assembly quality does not satisfy a quality requirement, determine a second assembly parameter based on the assembly quality, send the second assembly parameter to a subsequent operation device of the IIoT sensing and control platform, and control the subsequent operation device to perform the assembly operation based on the second assembly parameter; generate quality warning information based on the assembly quality and the second assembly parameter, and send the quality warning information to an IIoT user platform based on an IIoT service platform; and generate, based on the completed process, quality update data for updating the assembly database.
11. 11 . A method for assembly operation control, the method being executed by an IIoT management platform of a system for assembly operation control, and the method comprising: obtaining, based on an IIoT sensing network platform, quality inspection information of a plurality of parts to be assembled through a quality inspection device of an IIoT sensing and control platform; generating a plurality of candidate assembly groups based on assembly information, wherein the assembly information includes at least one assembly process, and an assembly object, a position to be assembled, an interference relationship, an assembly position, and a connection relationship corresponding to each of the at least one assembly process; for each of the plurality of candidate assembly groups, determining, based on quality inspection information of a plurality of parts to be assembled in the candidate assembly groups, a plurality of reference assembly records through an assembly database, wherein the assembly database is configured in a data center of the IIoT management platform; determining, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group; determining an assembly risk value of each of at least one assembly process based on predicted assembly quality of the plurality of candidate assembly groups; determining that the assembly risk value satisfies a risk condition; generating a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition, wherein the first assembly parameter includes an assembly grouping parameter and an assembly sequence parameter; and sending the first assembly parameter to an operation device of the IIoT sensing and control platform, and controlling the operation device to perform an assembly operation on the plurality of parts to be assembled based on the assembly grouping parameter and the assembly sequence parameter.
12. 12 . The method of claim 11 , wherein the generating a plurality of candidate assembly groups based on assembly information includes: determining a set of parts based on a type of the plurality of parts to be assembled; determining a plurality of sub-sets of parts based on the quality inspection information of the plurality of parts to be assembled in the set of parts, a count of the plurality of sub-sets of parts being related to the quality inspection information of the plurality of parts to be assembled in the set of parts; and determining the plurality of candidate assembly groups based on the plurality of sub-sets of parts.
13. 13 . The method of claim 12 , wherein the count of the plurality of sub-sets of parts is further related to a reference fit tolerance range of a fit part.
14. 14 . The method of claim 11 , wherein the determining, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group includes: for each of the plurality of reference assembly records, obtaining a machining accuracy of a plurality of historical assembly parts in the reference assembly record; determining a matching degree between the reference assembly record and a candidate assembly group corresponding to the reference assembly record based on the machining accuracy and an impact value of the plurality of historical assembly parts on a finished product assembly quality; and determining the predicted assembly quality of the candidate assembly group based on the matching degree and historical assembly quality in the plurality of reference assembly records.
15. 15 . The method of claim 11 , wherein the generating a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition includes: determining a fit part based on a historical assembly record of the plurality of parts to be assembled; and generating the first assembly parameter based on the fit part.
16. 16 . The method of claim 15 , wherein the generating the first assembly parameter based on the fit part includes: generating a plurality of candidate assembly sequences based on the assembly information; for each of the plurality of candidate assembly sequences, constructing, based on a plurality of sets of parts, a plurality of sub-sets of parts, and the candidate assembly sequence, an assembly map corresponding to the candidate assembly sequence; determining, based on the assembly map corresponding to the candidate assembly sequence, a predicted fit tolerance corresponding to the candidate assembly sequence through a fit model, the fit model being a machine learning model; and determining the assembly sequence parameter based on a plurality of predicted fit tolerances corresponding to the plurality of candidate assembly sequences.
17. 17 . The method of claim 16 , wherein an input of the fit model includes assembly environment information.
18. 18 . The method of claim 16 , wherein the fit model is obtained through incremental training based on the assembly quality, the incremental training includes: in response to determining that the assembly quality satisfies a predetermined training condition, determining an incremental training set based on a historical assembly record corresponding to the assembly quality; and performing an incremental update on the fit model based on the incremental training set, wherein a learning rate of the fit model during the incremental update is lower than a learning rate of the fit model before the incremental update.
19. 19 . The method of claim 11 , further comprising: obtaining assembly process data of the assembly operation based on a monitoring device of the IIoT sensing and control platform; determining assembly quality of a completed process based on the assembly process data; in response to determining that the assembly quality does not satisfy a quality requirement, determining a second assembly parameter based on the assembly quality, sending the second assembly parameter to a subsequent operation device of the IIoT sensing and control platform, and controlling the subsequent operation device to perform the assembly operation based on the second assembly parameter; generating quality warning information based on the assembly quality and the second assembly parameter, and sending the quality warning information to an IIoT user platform based on an IIoT service platform; and generating, based on the completed process, quality update data for updating the assembly database.
20. 20 . A non-transitory computer-readable storage medium storing computer instructions, wherein when reading the computer instructions in the storage medium, a computer implements a method for assembly operation control, the method being executed by an IIoT management platform of a system for assembly operation control, and the method comprising: obtaining, based on an IIoT sensing network platform, quality inspection information of a plurality of parts to be assembled through a quality inspection device of an IIoT sensing and control platform; generating a plurality of candidate assembly groups based on assembly information, wherein the assembly information includes at least one assembly process, and an assembly object, a position to be assembled, an interference relationship, an assembly position, and a connection relationship corresponding to each of the at least one assembly process; for each of the plurality of candidate assembly groups, determining, based on quality inspection information of a plurality of parts to be assembled in the candidate assembly groups, a plurality of reference assembly records through an assembly database, wherein the assembly database is configured in a data center of the IIoT management platform; determining, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group; determining an assembly risk value of each of at least one assembly process based on predicted assembly quality of the plurality of candidate assembly groups; determining that the assembly risk value satisfies a risk condition; generating a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition, wherein the first assembly parameter includes an assembly grouping parameter and an assembly sequence parameter; and sending the first assembly parameter to an operation device of the IIoT sensing and control platform, and controlling the operation device to perform an assembly operation on the plurality of parts to be assembled based on the assembly grouping parameter and the assembly sequence parameter.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a Continuation of U.S. Patent Application No. 19/001,376, filed on December 24, 2024, which claims priority to Chinese Patent Application No. 202411686939.8, filed on November 25, 2024, the entire content of each of which is hereby incorporated by reference. TECHNICAL FIELD The present disclosure relates to the field of assembly manufacturing, and in particular relates to a method, a system, and a storage medium for assembly operation control. BACKGROUND In the modern manufacturing industry, the assembly production line is a key link in realizing efficient, mass production of products. In the actual process of parts assembly, due to factors such as fluctuations in the quality of parts, assembly quality is difficult to control. Existing assembly operation control methods are still deficient in the prediction and warning of abnormal risks. Therefore, it is desired to provide a system, a method, and a storage medium for assembly operation control to efficiently analyze and predict abnormal risks that may occur in the process of assembling, and give timely warnings or safety recommendations to realize efficient and safe product production. SUMMARY One or more embodiments of the present disclosure provide a system for assembly operation control, comprising an Industrial Internet of Things (IIoT) management platform, an IIoT sensing network platform, and an IIoT sensing and control platform. The IIoT management platform is configured to: obtain, based on the IIoT sensing network platform, quality inspection information of a plurality of parts to be assembled through a quality inspection device of the IIoT sensing and control platform; generate a plurality of candidate assembly groups based on assembly information, wherein the assembly information includes at least one assembly process, and an assembly object, a position to be assembled, an interference relationship, an assembly position, and a connection relationship corresponding to each of the at least one assembly process; and for each of the plurality of candidate assembly groups: determine, based on quality inspection information of a plurality of parts to be assembled in the candidate assembly groups, a plurality of reference assembly records through an assembly database, wherein the assembly database is configured in a data center of the IIoT management platform; determine, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group; determine an assembly risk value of each of at least one assembly process based on predicted assembly quality of the plurality of candidate assembly groups. The IIoT management platform is further configured to: determine that the assembly risk value satisfies a risk condition; generate a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition, wherein the first assembly parameter includes an assembly grouping parameter and an assembly sequence parameter; and send the first assembly parameter to an operation device of the IIoT sensing and control platform, and control the operation device to perform an assembly operation on the plurality of parts to be assembled based on the assembly grouping parameter and the assembly sequence parameter. One or more embodiments of the present disclosure provide a method for assembly operation control, the method being executed by an IIoT management platform of a system for assembly operation control. The method comprises: obtaining, based on an IIoT sensing network platform, quality inspection information of a plurality of parts to be assembled through a quality inspection device of an IIoT sensing and control platform; generating a plurality of candidate assembly groups based on assembly information, wherein the assembly information includes at least one assembly process, and an assembly object, a position to be assembled, an interference relationship, an assembly position, and a connection relationship corresponding to each of the at least one assembly process; for each of the plurality of candidate assembly groups: determining, based on quality inspection information of a plurality of parts to be assembled in the candidate assembly groups, a plurality of reference assembly records through an assembly database, wherein the assembly database is configured in a data center of the IIoT management platform; determining, based on the plurality of reference assembly records, predicted assembly quality of the candidate assembly group; determining an assembly risk value of each of at least one assembly process based on predicted assembly quality of the plurality of candidate assembly groups. The method further comprises: determining that the assembly risk value satisfies a risk condition; generating a first assembly parameter in response to determining that the assembly risk value satisfies the risk condition, wherein the first assemb