A neural network hybrid diagnosis model for assembly machining accuracy based on physics guidance

By introducing physically guided neural network model and fuzzy hierarchical analysis method into the assembly process, a multi-level mechanism model is built, which solves the problem of dynamic identification and real-time monitoring of assembly line accuracy, and realizes efficient dynamic tracking and accurate identification of assembly process.

CN118862968BInactive Publication Date: 2024-12-20UNIV OF ELECTRONICS SCI & TECH OF CHINA
6 Cites 0 Cited by

Patent Information

Application Number
CN202411337513.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2024-12-20
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing assembly process diagnostic methods are difficult to directly use for dynamic identification and real-time monitoring of assembly line accuracy, especially due to the complex assembly line modeling and difficulty in data acquisition.

Method used

Using a neural network model based on physical guidance, the multi-level mechanism model of the assembly process is constructed, and the assembly process data is generated, and the fuzzy hierarchical analysis method and state space model are combined to construct a hybrid diagnosis of the convolutional neural network-long and short-term memory neural network architecture. Model, reduce the data volume requirement and realize dynamic prediction of product quality in the assembly process.

Benefits of technology

It realizes dynamic tracking and identification of assembly processes, reduces data acquisition volume, improves model accuracy, can be used for online product assembly quality identification, improves assembly accuracy and fault diagnosis accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention provides a hybrid diagnosis model of assembly processing accuracy based on physical guidance and belongs to the field of assembly processing and manufacturing. The hybrid diagnosis model proposed in the present invention first constructs a mechanism model of the required diagnosis assembly process line based on the fuzzy hierarchical analysis method and the state space model; then obtains the data set and preprocesses it; finally, constructs a dynamic prediction model of the assembly quality neural network guided by the mechanism model; this model can dynamically track and identify the assembly process, timely guide the model training process, reduce the amount of data required for the assembly process, ensure the model accuracy, and can be used for online product assembly quality accurate identification tasks. An assembly line fault diagnosis mechanism model that integrates the state space model and the fuzzy hierarchical analysis method is proposed, which can effectively reduce the amount of data collected during the assembly process. In addition, the method can also accurately identify the assembly process, thereby completing the multi-level judgment of the assembly.
Need to check novelty before this filing date? Find Prior Art