Abstract: Jet erosion is normally used to study the relative erosion behaviour of different material at moderate solid concentration, velocity, and particle size and impact angle. A slurry pot is then fabricated by inserting propeller from the bottom of the cylinder and is rotated at the speed required for uniform distribution. The test specimens are then mounted on test fixture which is fixed and has a provision to move in different angular position to find out the wear for different angles. Erosion of materials is occurs due to the impact of high velocity of slurry which is comes out from the nozzle and impacting on the test specimen. Different experiments were conducted for repeatability test by using parameters such as particle size distribution, solid-liquid concentration, variable angles (from150 to 900) and time variables. The results were obtained for Cast Iron (brittle material) by using above specified parameters. In this project the experimental results are validates by using Adaptive Neuro Fuzzy Inference System (ANFIS) and Regression analysis. It is observed that the erosion wear prediction of cast iron material with Adaptive Neuro-Fuzzy Inference system (ANFIS) using triangular membership function is better than Regression Analysis.

Keywords: Jet Erosion tester, Nozzle, Slurry erosion, Repeatability, Adaptive Neuro-Fuzzy Inference System, Regression.