One of my favorite process schemes in L-PBF, which you can see below. It seems so "easy" to change the L-PBF process parameters. Laser power, scan speed, hatch distance and layer thickness are a good starting point for printing your first parts. However, you still need to adapt your process to your customer application, and that's what makes additive manufacturing so exciting.
Even if you think of your printer as a black box, I think you will always need process engineers to tailor your process to the customer's needs.
I'm looking forward to the next steps in process engineering. From my perspective, AI will be the next stage to assist process engineers modify L-PBF parameters or change them directly in a running process (laser control systems). It will help us modify process parameters even faster (e.g. process simulations) and give us initial trends about process behavior without having to run multiple build jobs with density cubes across the build plate.
What are your thoughts on AI in L-PBF? Should we give AI a chance to accelerate our process development? Are you already using AI to analyze your monitored process data (e.g. machine data, powder bed images)?
Comments