Experimental and artificial neural network-based slurry erosion behavior evaluation of cast iron

• Published in: International journal on interactive design and manufacturing Vol. 18; no. 9; pp. 6739 - 6749
• Main Authors: Karthik, S., Sharath, B. N., Madhu, P., Madhu, K. S., Prem Kumar, B. G., Verma, Akarsh
• Format: Journal Article
• Language: English
• Published: Paris Springer Paris 01.11.2024; Springer Nature B.V
• Subjects: 20th century; Artificial intelligence; Artificial neural networks; CAE) and Design; Cast iron; Catalytic cracking; Composite materials; Computer-Aided Engineering (CAD; Electronics and Microelectronics; Engineering; Engineering Design; Experimental methods; Flow velocity; Hydraulic turbines; Impingement; Industrial Design; Instrumentation; Machine learning; Machinery; Mechanical Engineering; Neural networks; Original Paper; Parameter identification; Parameter modification; Petroleum refineries; Slurry pipelines; Solid surfaces; Turbines; Impingement angle; Artificial neural network; Kinetic energy; Slurry erosion; Cast iron
• DOI: 10.1007/s12008-023-01618-9
• ISSN: 1955-2513

Slurry erosion is the surface degradation process occurring due to the mechanical interactions between the solid surface and the particles in the presence of a fluid medium. Due to the slurry erosion, machinery components in the industrial fields (like the hydraulic turbines, slurry pipelines, etc.) conditions undergo severe damage such that the machine parts couldn’t be repaired and should be replaced with the new ones sooner. The present research work described a slurry jet erosion test rig based on the modified venturi devices. Slurry erosion tests were conducted by varying the parameters such as the impingement angle, velocity, concentration, silica sand as the erosive particles with size, and with cast iron as the test coupon. Artificial neural network methodology based on machine learning and artificial intelligence was adopted to identify the dominating parameter among the chosen test parameters. The experimental and the artificial neural networks method results confirmed that the impingement angle was the most significant parameter for causing the material removal amongst the other parameters. Specifically, from the artificial neural network prediction, it was found that the contribution of parameters for erosion prediction was impingement angle (highest) > velocity > concentration > erosive particle size (lowest). It was also reported that with an increase in the velocity the erosion of the test coupon also increased. This investigation may help the materials scientists to accelerate their studies on cast iron material domain.