ADAPTIVE PROCESS CONTROL USING MACHINE LEARNING FOR HIGH-QUALITY METAL ADDITIVE MANUFACTURING

Abstract

Metal Additive Manufacturing (MAM) has emerged as an advanced production technology capable of manufacturing lightweight, geometrically complex, and customized components for aerospace, biomedical, automotive, energy, and industrial applications. However, maintaining consistent product quality remains challenging because process parameters continuously vary due to thermal fluctuations, material inconsistencies, machine dynamics, and environmental conditions. Conventional process control techniques rely on predefined control rules and offline optimization, limiting their capability to respond to dynamic manufacturing conditions. Recent advances in Machine Learning (ML), Industrial Internet of Things (IIoT), Digital Twin technology, predictive analytics, and intelligent manufacturing enable adaptive process control through continuous analysis of real-time manufacturing information. This paper proposes an adaptive process control framework using Machine Learning for highquality Metal Additive Manufacturing by integrating intelligent sensor monitoring, predictive analytics, cloud manufacturing, and Digital Twin-based optimization. The proposed framework continuously predicts process deviations and dynamically adjusts manufacturing parameters to maintain product quality. Experimental analysis demonstrates significant improvements in manufacturing quality, process stability, defect reduction, operational efficiency, and enterprise scalability compared with conventional process control approaches. Keywords: Adaptive Process Control, Metal Additive Manufacturing, Machine Learning, Smart Manufacturing, Industrial Internet of Things, Predictive Analytics, Digital Twin, Quality Optimization, Industry 4.0, Intelligent Manufacturing.