Erosion Wear Response of Linz-Donawitz Slag Coatings: Parametric Appraisal and Prediction Using Imperialist Competitive Algorithm and Neural Computation

Slag, generated from basic oxygen furnace (BOF) or Linz-Donawitz (LD) converter, is one of the recyclable wastes in an integrated steel plant. The present work aims at utilization of waste LD slag to develop surface coatings by plasma spraying technique. This study reveals that LD slag can be gainfully used as a cost-effective wear-resistant coating material. A prediction model based on an artificial neural network (ANN) is also proposed to predict the erosion performance of these coatings. The 2.27% error shows that ANN successfully predicts the erosion wear rate of the coatings both within and beyond the experimental domain. In addition to it, a novel optimization algorithm called imperialist competitive algorithm (ICA) is used to obtain minimum erosion wear rate of 12.12 mg/kg. This algorithm is inspired by the imperialistic competition and has several advantages over other revolutionary algorithms like its simplicity, less computational time, and accuracy in predicting the results. A 2.39% error is noticed while comparing the erosion wear rate result of ICA with the experimental outcome.