AI-Driven Optimization of Ring Spinning: Adaptive Spacer Adjustment for Enhanced Yarn Quality and Production Efficiency

This study investigates the optimization of ring frame parameters, particularly focusing on the effects of spacer size on yarn quality-specifically strength, evenness, imperfections, and hairiness. Additionally, a novel AI-Driven Adaptive Spacer Adjustment System is introduced, which dynamically adjusts spacer size during the spinning process based on real-time sensor data. This combined analysis includes trials on 40 Ne and 60 Ne yarn counts with six different spacer sizes (2.75 mm to 4 mm) and an evaluation of machine efficiency improvements for 9 combed wool and 24 carded counts. The AI system uses a combination of optical sensors, reinforcement learning (PPO), support vector regression (SVR), and convolutional neural networks (CNN) to continuously monitor and optimize yarn quality and process efficiency. The results show significant improvements in yarn quality and productivity through both conventional spacer size optimization and the integration of AI-driven automation. The AI system offers real-time adjustments that enhance yarn quality and reduce production inefficiencies. This research provides a transformative approach to spinning process optimization by combining cutting-edge AI technology with traditional manufacturing techniques.