MODELING AND MULTI-OBJECTIVE OPTIMIZATION FOR APSed THERMAL BARRIER COATINGS ON IN718

Abstract

Experimental results based on [Formula: see text] orthogonal arrays (OAs) of Air Plasma Spraying (APS) parameters for deposition of thermal barrier coatings (TBCs) onto bond coated Inconel 718 (IN718) superalloys were used for mathematical models, using artificial neural network (ANN), for hardness and roughness. Thereafter, it was optimized from ANN-coupling Genetic Algorithm (GA). The developed ANN-based models showed optimal level of responses as 6.61 and 1209.8, for roughness and hardness, respectively. The average percentage prediction error (APPE)/mean error percentage (MEP) and root mean square error (RMSE) for the roughness were found as 0.07834% and 0.00754%, respectively, while these APPE/MEP and RMSE for hardness were found as 0.456% and 0.00765%, respectively.