Back propagation model for prediction of deposition parameters in plasma sprayed WC-based coatings

The deposition parameters frequently have a significant impact on the characteristics of plasma spray coating. Due to the intricate chemical and thermodynamics processes, it is challenging to study and create a complete model of plasma spray process. The use of WC-based coatings, which offers exceptional wear resistance, in turbomachinery components such as blades, vanes, extend their service life and lowers need of maintenance. The objective of this investigation is to develop a predictive model for mechanical properties of plasma deposited WC20Cr₃C₂7Ni coatings using a back propagation neural network. Additionally, the study also analyses the impact of deposition parameters irrespective of the intermediate process. The change in porosity, nano-hardness, and sliding wear rate of coatings under various powder feed rate, stand-off distance, and powder gas N₂ flow rate was predicted using back propagation neural network algorithm. The developed model accurately predicted the characteristics of WC-based coatings, evidenced by a comparison between predicted and experimental results that shows similar trends. In order to specifically evaluate each input variable’s relative importance for improving prediction accuracy, the mean impact value analysis was used.