Machine learning applications for quality improvement in laser powder bed fusion: A state-of-the-art review

Abstract

As one of the most popular additive manufacturing methods, laser powder bed fusion (L-PBF) builds 3D components with complex geometries layer by layer using alloy powders. This technique has found widespread adoption in various industrial applications, including biomedical and aerospace fields. However, L-PBF encounters challenges related to poor process repeatability and inconsistency in fabricated part quality, which hinder its broader adoption. Various quality improvement methods have been proposed to address these challenges and achieve high-quality, reliable parts. Given the abundance of parameters and the intricate phenomena that occur during the process, machine learning (ML) methods play a critical role in enhancing the quality of L-PBF, providing an optimum solution for improving the quality of manufactured parts. This review paper begins with a comprehensive and straightforward introduction to ML, focusing primarily on different learning approaches. Subsequently, the paper explores different ML methods applied to parameter optimization and in situ monitoring, both contributing to enhanced quality control. In parameter optimization, ML is employed to extract relationships between input parameters and key factors such as melt pool characteristics, porosity, and mechanical properties. Shifting the focus to in situ monitoring, the paper introduces the application of ML in analyzing various sensor data generated throughout the L-PBF process. Accomplished tasks include segmentation, regression, and classification of quality measurement. In summary, this review underscores the critical role of machine learning in addressing challenges associated with L-PBF, providing an optimal solution for quality enhancement.