The aim of the present study is to investigate the slurry erosion behavior of uncoated and WC-10Co-4Cr coated SS-317 steel, with a focus on the influence of various parameters on erosion rates. Additionally, it examines the predictive accuracy of the Artificial Neural Network (ANN) model for forecasting erosion behavior and compares its performance with traditional methods such as Multiple Linear Regression (MLR).
The research employed Design of Experiments (Taguchi and ANOVA) to systematically analyze the effects of rotational speed, concentration, particle size and time duration, on the erosion rates. Erosion behavior was modeled using ANN, with the model’s predictions being rigorously compared against experimental results to evaluate its accuracy. Additionally, erosion mechanisms were explored to gain insights into the wear processes and material performance.
The results showed that WC-coated SS-317 (986 HV) exhibited 22% higher resistance to erosion mass loss than uncoated SS-317, owing to its higher hardness. The particle size was the most influential parameter (confirmed using ANOVA) as it contributed 45.5% to the erosion wear rate for coated SS-317, with time duration contributing the least. The ANN model demonstrated high prediction accuracy, with error rates of 0–4.5% for uncoated and 0–1.2% for coated SS-317, outperforming MLR by 20-fold.
This study highlights the exceptional erosion resistance of WC-10Co-4Cr coatings, ideal for high-abrasion environments like thermal plants and ash-handling systems. Utilizing Taguchi, ANN and ANOVA approaches, it provides valuable insights into material performance and reliable erosion wear predictions, enhancing durability and efficiency in industrial applications.
