Performance degradation assessment method for linear motor feed systems driven by digital twins.

Area of Science:

• Engineering
• Artificial Intelligence
• Materials Science

Background:

• Traditional degradation assessment methods for linear motor feeding systems suffer from missing samples and low accuracy.
• Digital Twin technology offers advantages in complex equipment status evaluation and performance prediction.

Purpose of the Study:

• To introduce Digital Twin technology for enhanced performance degradation assessment in linear motor feeding systems.
• To address challenges of missing degradation samples and improve assessment accuracy.

Main Methods:

• A digital integrated model of mechanical, electrical, and control subsystems was established, analyzing multi-field coupling mechanisms.
• Gramian Angular Field (GAF) encoding converted time-series data to images, processed by AlexNet for degradation assessment.
• Simulations generated current datasets for normal and demagnetization degradation conditions.

Main Results:

• The proposed GAF-AlexNet model achieved 98.3% accuracy in assessing irreversible demagnetization degradation levels.
• Significant improvements were observed in the AUC index (10.3-15.2%) and average accuracy (10.5-14.7%) compared to traditional methods.
• The method effectively resolves the issue of insufficient real degradation samples.

Conclusions:

• The Digital Twin-integrated GAF-AlexNet model provides a powerful tool for performance degradation assessment in linear motor feeding systems.
• This approach enhances predictive maintenance and performance monitoring in industrial environments.
• The study successfully overcomes limitations of traditional methods and data scarcity.