Chalak Qazani, Mohamad Reza, Pourmostaghimi, Vahid, Moayyedian, Mehdi, and Pedrammehr, Siamak (2022) Estimation of tool--chip contact length using optimized machine learning in orthogonal cutting. Engineering Applications of Artificial Intelligence, 114. 105118.

PDF (Published Version) - Published Version Restricted to Repository staff only

Abstract

Tool–chip contact length has a significant effect on the various characteristics of metal cutting, including cutting pressures, chip formation, tool wear, tool life, and cutting temperatures. It should be added that there is a direct relationship between the tool–chip contact length and secondary shear zone thickness in the metal cutting process. The cutting force and shear zone temperature decrease by the reduction of tool–chip contact length. In addition, the tool–chip contact length affects the tool life and workpiece surface roughness. Lots of researchers have conducted extensive research to calculate the tool–chip contact length using mathematical or machine learning methods. The main objective of this study is to calculate the tool–chip contact length using a highly advanced machine learning method without any time-consuming and expensive experiments. However, an adaptive network-based fuzzy inference system (ANFIS) is not used yet in the prediction of the tool–chip contact length. In this study, we proposed the ANFIS to predict the tool–chip contact length for the first time in orthogonal cutting using depth of cut, feed-rate, and cutting speed as inputs of the proposed model. As the second contribution of this study, three evolutionary-based optimization techniques, including genetic algorithm, particle swarm optimization, and grey wolf optimization, as well as global-based Bayesian optimization, are employed to select the optimal hyperparameters of the proposed ANFIS model known as GA-ANFIS, PSO-ANFIS, GWO-ANFIS, and B-ANFIS, respectively. The proposed methods are designed and developed in MATLAB software to be compared with the previous method using genetic programming (GP). The outcomes of this research demonstrate that the GWO-ANFIS can decrease the mean square error between the actual and predicted tool–chip contact length of 15.60%, 3.67%, 89.75%, and 92.17% in comparison with those of GA-ANFIS, PSO-ANFIS, B-ANFIS, and GP, respectively. In addition, the fuzzy logic rule surface of the GWO-ANFIS shows 57.20%, 30.95%, and 11.85% dependency of tool–chip contact length to cutting speed, feed-rate, and depth of cut as the inputs of the orthogonal cutting process, respectively.

Item ID:	86736
Item Type:	Article (Research - C1)
ISSN:	1873-6769
Copyright Information:	© 2022 Elsevier Ltd. All rights reserved.
Date Deposited:	28 Aug 2025 00:16
FoR Codes:	40 ENGINEERING > 4014 Manufacturing engineering > 401406 Machining @ 40% 46 INFORMATION AND COMPUTING SCIENCES > 4611 Machine learning > 461199 Machine learning not elsewhere classified @ 60%
SEO Codes:	24 MANUFACTURING > 2412 Machinery and equipment > 241204 Industrial machinery and equipment @ 60% 22 INFORMATION AND COMMUNICATION SERVICES > 2204 Information systems, technologies and services > 220403 Artificial intelligence @ 40%
Downloads:	Total: 1 Last 12 Months: 1
	More Statistics