A review of energy efficiency and Machine learning analysis for additive manufacturing of direct laser metal deposition

Abstract

Additive Manufacturing (AM) is an innovative industrial process that utilizes layering of materials to create unique products with an unprecedented level of flexibility. To maximize the benefits of AM, it is crucial to understand the precise utilization of laser energy. This review provides a concise overview of the Directed Laser Metal Deposition (DLMD) process, which includes an analysis of input parameters and the effective use of laser power, along with the impact of laser synchronous preheating processes. Various input parameters such as travel speed, hatch spacing, powder feed rate, and the standard of distance influence the calculation of volumetric energy density (VED). VED is important for Clad bed geometry, including dilution (D), height (H), width (W) of AM wall, and clad bead stability (ΔH). In the field of manufacturing, Machine Learning (ML) is a commonly used subset of Artificial Intelligence (AI) to improve the quality of processes, expedite production, and reduce costs. Specifically, in Metal Additive Manufacturing (MAM), ML techniques are leveraged to optimize various process parameters, resulting in the production of high-quality, defect-free products. This review provides a detailed explanation of the step-by-step implementation process of ML and its classification within a fundamental structure.