This study adopts two computational intelligence techniques namely least-squares support vector machine (LS-SVM) and group method of data handling (GMDH) type polynomial neural network to model the plasma spray coating process. The coating qualities were evaluated by determining its thickness, porosity and micro-hardness. Four parameters including primary gas flow rate, stand-off distance, powder flow rate and arc current that affecting the coating properties were chosen as input variables in model development. The performances of the developed models were evaluated by calculating the deviations between the predicted and the actual values based on the performance indices of mean absolute error (MAE), root mean square error (RMSE) and coefficient of determination (R2). The results demonstrate thehigh capability of the LS-SVM and GMDH-type polynomial neural network for prediction of thickness and micro-hardness values with lowest MAE, RMSE and highest R2. Due to the high non-linearity behavior of porosity in coating process, the results obtained by these models for porosity are not very well.
An Experimental Investigation into the Amalgamated Al<sub>2</sub>O<sub>3</sub>-40% TiO<sub>2</sub> Atmospheric Plasma Spray Coating Process on EN24 Substrate and Parameter Optimization Using TLBO
