MULTI-SENSOR FUSION FOR EARLY DEFECT DETECTION IN METAL 3D PRINTING PROCESSES

Abstract

Metal 3D printing has become a critical manufacturing technology for producing complex components with improved design flexibility, reduced material waste, and accelerated production cycles. However, early detection of defects such as porosity, cracks, incomplete fusion, thermal distortion, and surface irregularities remains a major challenge affecting product reliability and manufacturing efficiency. This paper proposes a multi-sensor fusion framework for early defect detection in metal 3D printing processes by integrating Industrial Internet of Things (IIoT), machine learning, thermal imaging, optical monitoring, acoustic sensing, vibration analysis, Digital Twin technology, and cloud-based predictive analytics. The proposed methodology continuously collects heterogeneous manufacturing data, performs intelligent feature extraction, detects early defect signatures, and supports adaptive process control for improved production quality. Experimental evaluation demonstrates improvements in defect detection accuracy, process stability, manufacturing reliability, energy efficiency, and operational performance. The proposed framework provides a scalable solution for intelligent quality assurance in Industry 4.0- enabled additive manufacturing environments. Keywords— Multi-Sensor Fusion, Metal 3D Printing, Defect Detection, Machine Learning, IIoT, Digital Twin, Smart Manufacturing, Industry 4.0.