Plasma spraying is a key industrial coating process that exhibits intricate nonlinear interactions among process parameters. This complexity makes accurate predictions of particle properties, which greatly affect process behavior, very challenging. Specifically, particle velocities and temperatures profoundly impact coating quality and process efficiency. Conventional methods often require empirical correlations and extensive parameter tuning due to their limited ability to capture the underlying physics within this intricate system. This study introduces Physics-Informed Neural Networks (PINNs) as a solution. By seamlessly integrating known physical laws and constraints directly into the model architecture, PINNs offer the potential to learn the underlying physics of the system. For comparison, Artificial Neural Networks (ANNs) are also developed. Computational Fluid Dynamics (CFD) simulations of a plasma generator and plasma jet model provide data to train both ANN and PINN models. The study reveals an improvement in particle velocity prediction through the proposed PINN model, demonstrating its capability to handle complex relationships. However, challenges arise in predicting particle temperature, warranting further investigation. The developed models can aid in optimizing the plasma spraying process by predicting essential particle properties and guiding necessary process adjustments to enhance coating quality.

Thermal Spraying of tungsten carbide is usually performed to optimize surface hardness. However, often there are additional coating or process requirements, e.g., low porosity, low roughness, or high deposition efficiency, that must be met to satisfy the customer or increase the process efficiency. Such optimizations are difficult because various influencing factors, e.g., lambda (λ), standoff-distance, or powder feed rate, have different effects on the individual coating and process properties. In this study, several trials were performed and analyzed (technically & databased) to understand how each influencing factor affects these coating and process properties. It could be determined that there is no general best performance parameter but that only individual solutions depending on prioritized importance of the properties can be obtained. Using data, one way of solving the challenge is the performance of multi-objective optimization. With the help of this method, the challenges of optimization can be solved using test data. The simultaneous optimization can be visualized in a 2D/3D graph. A boundary line can be observed where different process and coating characteristics are optimized. Applying such databased methods can help to work more efficiently as well as more sustainable in development and application.