What does this research mean for the field?

Integrating generalized receptance coupling substructure analysis (GRCSA) with artificial intelligence algorithms creates a dynamic modeling method that reduces the prediction error of frequency response functions in precision motion platforms to approximately 2.5%. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to improve the prediction of dynamic characteristics for precision motion platforms under high-speed conditions by combining advanced analytical methods with AI.

June 18, 2026Open Access

Prediction of Dynamic Characteristics and Control Parameter Optimization for Precision Motion Stages by Integrating Generalized Receptance Coupling Substructure Analysis and Machine Learning

Key Points

This research aims to improve the prediction of dynamic characteristics for precision motion platforms under high-speed conditions by combining advanced analytical methods with AI.
Integrated generalized receptance coupling substructure analysis with AI algorithms for modeling.
Established a nonlinear prediction model for dynamic parameters based on frequency response functions.
Optimized interface control parameters to reduce prediction errors in dynamic characteristic modeling.
Reduced prediction error of frequency response functions from 3.50% to 2.47%.
Achieved an overall prediction error of about 2.5% under various control parameters.

Abstract

To address the complex dynamic behavior of four-axis precision motion platforms under high-speed and high-acceleration conditions, as well as the difficulty of traditional modeling methods in balancing accuracy and efficiency, this paper proposes a data/model-driven dynamic modeling and analysis method that integrates generalized receptance coupling substructure analysis (GRCSA) with artificial intelligence (AI) algorithms. Based on the GRCSA theory, the initial analytical framework of the dynamic model of the precision motion platform is established, and the frequency response functions (FRFs) of the substructure and interface are preliminarily obtained. On this basis, the nonlinear prediction model of the dynamic parameters of the interface driving direction is established by using the AI algorithm, enabling fast and accurate prediction of the dynamic characteristics of the interface under different servo control parameters in the guide rail driving direction. Finally, based on the data/model-driven dynamic modeling and analysis method, the interface control parameters are optimized. The interface and substructure parameters are modified to reduce the prediction error of the FRFs from 3.50% to 2.47%. This method can achieve the prediction error of the dynamic characteristics of the interface under different control parameters of about 2.5%.

Read Full Paperexternally

Demander à l'IA

Bookmark

View Full Paper