Non-contact Inspection of Electrically Discharged Materials Using Machine Learning

Electrical discharge machining (EDM) is a very popular non-conventional machining technique for electrically conductive materials. In this process, surface roughness is an important characteristic of EDM-machined surfaces. This is because it affects the performance and durability of the machined parts. However, the conventional contact-based inspection methods for measuring surface roughness are time-consuming and can damage the machined surface. In this paper, we propose a non-contact method for inspecting electrically discharged materials using machine learning. We use three regression algorithms to predict the surface roughness of EDM-machined surfaces based on non-contact measurements using a self-made augmented dataset. Out of all three, K-nearest neighbors (KNN) was found to be the best performing algorithm, with a mean squared error (MSE) of 0.00157 and R2-score of 0.99.