SENSOR-DRIVEN TOOL WEAR PREDICTION AND ADAPTIVE CONTROL IN SMART MACHINING SYSTEMS

Abstract

Tool wear is one of the most significant factors affecting machining quality, dimensional accuracy, production efficiency, and manufacturing cost in modern machining operations. Conventional tool wear assessment relies on periodic manual inspection, which often results in unexpected tool failures and production interruptions. This paper proposes a sensordriven tool wear prediction and adaptive control framework integrating intelligent sensors, Industrial Internet of Things (IIoT), machine learning, predictive analytics, cloud manufacturing, and real-time process monitoring for smart machining systems. The proposed methodology continuously acquires machining signals including cutting force, vibration, acoustic emission, temperature, spindle power, and tool displacement to predict tool wear and automatically optimize machining parameters. Adaptive process control dynamically adjusts machining conditions to extend tool life while maintaining product quality and operational stability. Experimental evaluation demonstrates improvements in prediction accuracy, machining reliability, tool utilization, production efficiency, predictive maintenance, and manufacturing sustainability. The proposed framework provides an intelligent and scalable solution for smart machining environments.