An assessment of erosive wear of hydro-turbine steel using statistical modelling and optimisation

Abstract

The current study pertains to the influence of chosen process parameters on erosive wear of F6NM stainless steel. Response surface methodology was used to plan experiments. Response surface method with face centred composite design has been adopted to develop a regression model. Development of erosive wear model was based on five factors, which included sediment concentration (A), particle size (B), angle of impact (C), test duration (D) and rotational speed of slurry (E). A mathematical model was developed to predict the deterioration through wear on F6NM stainless steel and the appropriateness of the model was certified using analysis of variance. A robust correlation is attained between the model predicted and experimentally obtained values for weight loss and the percentage of error is 12%. On the basis of mathematical model, single objective optimisation of parameters has been performed with genetic algorithm (GA) technique and this method yields reduction of 34.78% for material wear.