Prediction of Particle Properties in Plasma Spraying Based on Machine Learning

Abstract Thermal spraying processes include complex nonlinear interdependencies among process parameters, in-flight particle properties and coating structure. Therefore, employing computer-aided methods is essential to quantify these relationships subsequently enhance the reproducibility. Typically, classic modeling approaches are pursued understand interactions. While able capture very systems, increasingly sophisticated models have drawback of requiring considerable calculation time. In this study, two different Machine Learning (ML) methods, Residual Neural Network (ResNet) Support Vector (SVM), were used estimate in plasma a much faster manner. To end, data sets comprising parameters such as electrical current gas flow well velocities, temperatures positions been extracted from CFD simulation jet. Furthermore, Design Experiments (DOE) Central Composite (CCD) Latin Hypercube Sampling (LHS), employed cover set representative for training ML models. The results show that developed trends precisely dramatically than computation-intensive simulations.