Artificial Neural Networks for Optimizing Alumina Al<sub><b>2</b></sub>O<sub><b>3</b></sub> Particle and Droplet Behavior in 12kK Ar-H<sub><b>2</b></sub> Atmospheric Plasma Spray

This paper investigates the application of Direct Current Atmospheric Plasma Spraying (DC-APS) as a versatile thermal spray technique for the application of coatings with tailored properties to various substrates. The process uses a high-speed, high-temperature plasma jet to melt and propel the feedstock powder particles, making it particularly useful for improving the performance and durability of components in renewable energy systems such as solar cells, wind turbines, and fuel cells. The integration of nanostructured alumina (Al2O3) thin films into multilayer coatings is considered a promising advancement that improves mechanical strength, thermal stability, and environmental resistance. The study highlights the importance of understanding injection parameters and their impact on coating properties and uses simulation tools such as the Jets & Poudres (JP) code for in-depth analysis. Furthermore, the paper discusses the implementation of Artificial Neural Networks (ANN) to optimize the coating process by predicting flight characteristics and improving operating conditions. The results show that ANN models are effective in achieving highly accurate prediction values, highlighting the potential of AI in improving thermal spray technology.