Metal-Based Additive Manufacturing Condition Monitoring: A Review on Machine Learning Based Approaches

The metal-based additive manufacturing (MAM) processes have great potential in wide industrial applications, for their capabilities in building dense metal parts with complex geometry and internal characteristics. However, various defects in the MAM process greatly affect the precision, mechanical properties and repeatability of final parts. These defects limit its application as a reliable manufacturing process, especially in the aerospace and medical industries where high quality and reliability are essential. MAM process monitoring provides a technical basis for avoiding and eliminating defects to improve the build quality. Based on of the nature of the MAM build defects, this article conducts a thorough investigation of monitoring methods, and proposes a machine learning (ML) framework for process condition monitoring. According to the structure of ML models, they are divided into shallow ML-based and deep learning-based methods. The state-of-the-art ML monitoring approaches, as well as the advantages and disadvantages of their algorithmic implementations, are discussed. Finally, the prospects of ML based process monitoring researches are summarized and advised.

I. Overview of MAM Process Monitoring System

Due to its excellent capabilities in building high performance metal components with complex structures, metal-based additive manufacturing (MAM) has been the most important additive manufacturing (AM) technologies, with broad application in aerospace, biomedical and automotive industries [1]–[3]. According to the material feeding methods, MAM processes are divided into two categories: direct energy deposition (DED) and powder bed fusion (PBF) [4]. The DED processes, such as laser metal deposition (LMD) and wire and arc additive manufacturing (WAMM) use synchronous feeding of powder or wire to fill the melt pool area with raw materials at the same time of high-energy beam scanning. The PBF uses a high-energy power source to build 3-D part layer-by-layer by fusing fine powders laid in the build chamber in advance. It mainly includes selective laser melting (SLM), selective laser sintering (SLS), and electron beam melting (EBM). Due to the coupling of complex physical-metallurgical processes in a very short time, the MAM build is prone to macro mechanical defects such as balling [5], [6], delamination, cracking [7], [8], powder bed defects [9], [10], and micro metallurgical defects such as porosity [11] and lack of fusion [12]. Table I gives the common defects in MAM. The mechanical properties of AM metal parts are deteriorated due to the internal defects of materials, which brings the main technical bottleneck in MAM applications. In order to produce high-quality parts, it is very important to monitor and control the defects of the build at the incipient stage. Adaptively adjusting process parameters can eliminate build defects, improve the build quality and the process stability. TABLE I Common Defects of MAM Process