Use of Machine Learning to Improve Additive Manufacturing Processes

Abstract

Rapidly developing artificial intelligence (AI) can help machines and devices to perceive, analyze, and even make inferences in a similar way to human reasoning. The aim of this article is to present applications of AI methods, including machine learning (ML), in the design and supervision of processes used in the field of additive manufacturing techniques. This approach will allow specific tasks to be solved as if they were performed by a human expert in the field. The application of AI in the development of additive manufacturing technologies makes it possible to be assisted by the knowledge of experienced operators in the design and supervision of processes acquired automatically. This reduces the risk of human error and simplifies and automates the production of products and parts. AI in 3D technology creates a wide range of possibilities for generating 3D objects and enables a machine equipped with a vision system, used in ML processes, to analyze data similar to human thought processes. Incremental printing using such a printer allows the production of objects of ever-increasing quality from several materials simultaneously. The process itself is also precise and fast. An accuracy of 97.56% means that the model is precise and makes very few errors. The 3D printing system with artificial intelligence allows the device to adapt to, for example, different material properties, as the printer examines the 3D-printed surface and automatically adjusts the printing. AI/ML-based solutions similar to ours, once learning sets are modified or extended, are easily adaptable to other technologies, materials, or multi-material 3D printing. They also allow the creation of dedicated, ML solutions that adapt to the specifics of a production line, including as self-learning solutions as production progresses.

Keywords

• artificial intelligence (AI)
• machine learning (ML)
• additive manufacturing
• 3D printing