Implementation of Adaptive Neuro Fuzzy Inference System in Determining the Influence of Bearing Clearance in Mild Steel Turning

• Author(s): Amiebenomo Sebastian Oaihimire ; Ighodalo Osagie
• Paper ID: 1705041
• Page: 236-251
• Published Date: 18-09-2023
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 7 Issue 3 September-2023

Abstract

The metal cutting industry constantly seeks to optimize machining techniques to guarantee highly accurate parts while minimizing costs. However, the complexity of the relationship between process parameters and performance measures such as metal removal rate (MRR), surface finish, chip flow pattern, specific energy consumption, and tool life presents a significant challenge for manufacturers. The paper introduces an innovative solution for determining the influence of bearing clearance on mild steel turning operations using the Adaptive Neuro Fuzzy Inference System (ANFIS) optimization model. By assimilating insights from training data onto a fuzzy inference system, ANFIS effectively maps solutions to complex problems. The proposed approach accurately predicts surface roughness, MRR, and tool wear for different sets of cutting parameters, offering a viable approach for improving product quality and profitability while reducing associated manufacturing costs. By applying substrative clustering with values of radius of parameter equal to 0.1, 0.2, and 0.3 respectively, the initial membership function of the independent variables and fuzzy rules were developed. Training was done by using an initial step size of 0.1, the value of MAPE obtained was 3.123% and correlation coefficient (R) of 0.9072. The results obtained indicate that the ANFIS model predicts surface roughness with high accuracy, with an average error of 0.17?m. The study also found that increasing bearing clearance results in a decrease in surface roughness, as seen with a reduction from 3.96?m to 2.22 ?m. Furthermore, the model predicted MRR with an average error of 0.6%, and it revealed that increasing bearing clearance results in a significant increase in MRR, ranging from 2.22 m3/min to 5.98 m3/min at a clearance level of 0.010mm. The findings contribute to the body of knowledge in manufacturing engineering, offering valuable insights into the relationship between bearing clearance, machining parameters, and performance measures.

Keywords

Accurate machining techniques, ANFIS model, Bearing clearance, Surface roughness in mild steel, Manufacturing engineering, Optimization modeling

Citations

IRE Journals:
Amiebenomo Sebastian Oaihimire , Ighodalo Osagie "Implementation of Adaptive Neuro Fuzzy Inference System in Determining the Influence of Bearing Clearance in Mild Steel Turning" Iconic Research And Engineering Journals Volume 7 Issue 3 2023 Page 236-251

IEEE:
Amiebenomo Sebastian Oaihimire , Ighodalo Osagie "Implementation of Adaptive Neuro Fuzzy Inference System in Determining the Influence of Bearing Clearance in Mild Steel Turning" Iconic Research And Engineering Journals, 7(3)