Deep learning-based digital twin for intelligent predictive maintenance of rapier loom.

This article focuses on an intelligent predictive maintenance approach for rapier looms driven by the integration of digital twin technology and deep learning. It presents a comprehensive digital twin system architecture for the full operating cycle of rapier weaving machines, enabling real-time monitoring and virtual-real interaction. The study proposes an improved Whale Optimization Algorithm (IWOA) optimized Back Propagation (BP) neural network model for evaluating and predicting process parameters related to loom efficiency, alongside a Bidirectional Long Short-Term Memory (BiLSTM) network-based health state assessment model that constructs a health index to predict spindle degradation and remaining life. Experimental results demonstrate that combining digital twins with deep learning enhances predictive maintenance accuracy, supports dynamic model updating through twin data, and facilitates timely maintenance decisions to improve operational safety and reduce costs in textile manufacturing.