One of the main challenges in additive manufacturing is the control in-process to achieve the manufacturing of “first-time-right” parts. However, difficulties are multiples. The first is to collect properly the data needed to monitor the process (temperatures, images, …) and then to analyze those data to be able to define some control models and at the end to embed them in the loop to correct or control the on-going process.
Different ways are investigated in the state-of-the art for monitoring and data analysis using traditional Machine Learning algorithms [1]. But they are still too simple for a Multiphysics process such as additive manufacturing or too time consuming for analysis which leads to a non-possible real-time control.
So, in this research topic, the scientific objective is to develop a more adapted analysis model to support the implementation of closed-loop control of DED processes.
[1] C. Wang, X.P. Tan, S.B. Tor, C.S. Lim, "Machine learning in additive manufacturing: State-of-the-art and perspectives" Additive Manufacturing, Volume 36, 2020, 101538, ISSN 2214-8604,

Additive manufacturing is often presented as a greener process than conventional manufacturing for mechanical parts manufacturing. The analysis is generally done through the angle of material consumption, material waste. But as additive manufacturing process can be a long process and as it is using high energy means such as Lasers or electric arc generator it becomes important to assess the global energy use during the process and for different process parameters [2]. This is why one of the main goals of this research line is to implement energy data measurement (linked with the previous research line in ADVANCED MATERIALS AND PROCESSES GROUPS). Indeed, the objective is to study the completeness of the data used for the control models in the frame of an environment assessment tool. This tool could be a multi-criteria reference based on a simplified model for LCA (Life Cycle Analysis) of the additive manufacturing process. For this, a simplified database of carbon impact and energy consumption flows will be developed. Other pollutant emissions on deployed AM technologies as well as the energy necessary due to the resource’s consumption can also be studied to alert of a risk of impact transfer.
[2] Peng, T., Kellens, K., Tang, R., Chen, C., & Chen, G. (2018). Sustainability of additive manufacturing: An overview on its energy demand and environmental impact. Additive Manufacturing, 21, 694-704.