LAUSR

Abstract Thermal spray comprises a group of coating processes for coating manufacturing in which metallic or nonmetallic materials are deposited in a molten or semi-molten condition. Most often, the coating properties are significantly influenced by the operating parameters. However, obtaining a comprehensive modeling or analytical analysis of the thermal spray process is too difficult to be practical due to the complex chemical and thermodynamic reactions. Accordingly, the present study aims at applying an artificial neural network (ANN) model to predict the HVOF sprayed Cr3C2−25NiCr coatings and analyze the influence of operating parameters regardless of the intermediate process. The process parameters, which were automatically recorded by the homemade HVOF spray system during the spray process, were served as the inputs for the ANN model. Then, the porosity, microhardness and wear rate of coatings were measured and considered as targets for the ANN model. After configuring and training procedure of the model, the predicted results were compared to the results of experimental data. The good consistency found between these results permits to verify the reliability and accuracy of the trained ANN model. Additionally, the mean impact value (MIV) analysis was conducted to quantitatively explore the relative significance of each input variable for improving the effective prediction.