Intelligent control method and system for horizontal lathe based on digital twin

By constructing a working state map and a spatiotemporal neural network for a horizontal lathe, the problem of low accuracy in the intelligent control strategy of horizontal lathes in existing technologies is solved. This enables real-time self-evolution of the horizontal lathe and accurate prediction of future time periods, thereby improving the accuracy and safety of intelligent control.

CN122308258APending Publication Date: 2026-06-30YANGZHOU CHUNFENG MARINE MASCH MFG CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
YANGZHOU CHUNFENG MARINE MASCH MFG CO LTD
Filing Date
2026-04-21
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing digital twin monitoring technology lacks topological correlation modeling of the overall machine tool structure in horizontal lathes. This makes it susceptible to interference from complex environmental noise when facing transient changes during the cutting process, making it difficult to achieve effective denoising and mapping from high-dimensional physical space to low-dimensional potential feature space, thus affecting the accuracy of the intelligent control strategy of horizontal lathes.

Method used

Multiple working data of a horizontal lathe are collected, and a working state map is constructed by combining a three-dimensional model. Through a spatiotemporal graph neural network and a dynamic adjacency matrix update mechanism, a real-time self-evolving working state map is constructed, which is then denoised and the working evolution trajectory for future time periods is predicted. By combining a causal inference network, key factors and safe feasible regions are determined, and intelligent control strategies are output.

Benefits of technology

This improves the accuracy of the horizontal lathe's work evolution trajectory and the precision of its intelligent control strategy over future time periods, fully considers the safe and feasible domain and work constraints, and enhances the intelligent control capabilities of the horizontal lathe.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122308258A_ABST
    Figure CN122308258A_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent control method and system for a horizontal lathe based on digital twins. The invention relates to the technical field of digital twins. Based on the dynamic identification of the working state map, the transient changes of the horizontal lathe during operation are determined to identify the feature mapping result from a high-dimensional physical space to a low-dimensional potential feature space. This feature mapping result is input into the digital twin space, and a long short-term memory network is introduced into the digital twin space to improve the accuracy of the horizontal lathe's working evolution trajectory in future time periods. Dynamic compensation data and the current working state of the horizontal lathe are input into a causal inference network to identify multiple key factors causing abnormalities in the horizontal lathe. Dynamic inference is performed on these key factors and the current working history of the horizontal lathe. Combined with the working constraints of the horizontal lathe, an intelligent control strategy for the horizontal lathe is determined, improving the accuracy of the intelligent control strategy.
Need to check novelty before this filing date? Find Prior Art