1. The prediction of surface roughness and tool vibration by using metaheuristic-based ANFIS during dry turning of Al alloy (AA6013)
- Author
-
Mehmet Ali Guvenc, Hasan Huseyin Bilgic, Mustafa Cakir, Selcuk Mistikoglu, Mühendislik ve Doğa Bilimleri Fakültesi -- Makina Mühendisliği Bölümü, Havacılık ve Uzay Bilimleri Fakültesi -- Havacılık ve Uzay Mühendisliği Bölümü, Güvenç, Mehmet Ali, Çakır, Mustafa, and Mıstıkoğlu, Selçuk
- Subjects
Optimization ,Engineering & Materials Science - Manufacturing - Tool Wear ,Signals ,Swarm optimization ,Turning ,Tool vibration ,Chatter suppression ,Fuzzy neural networks ,Aerospace Engineering ,ACO-ANFIS ,Surface Roughness ,Industrial and Manufacturing Engineering ,Tool vibrations ,Ant colony optimization ,Particle swarm ,Adaptive neuro-based fuzzy inference system ,Wear ,Extension ,MLRM ,Particle swarm optimization-adaptive neuro-based fuzzy inference system ,ACO-adaptive neuro-based fuzzy inference system ,GA-ANFIS ,Mechanical Engineering ,Applied Mathematics ,General Engineering ,Genetic algorithm-adaptive neuro-based fuzzy inference system ,Fuzzy systems ,Genetic algorithms ,Fuzzy inference systems ,Aluminum alloys ,Fuzzy inference ,Inconel (Trademark) ,Particle swarm optimization (PSO) ,Automotive Engineering ,Carbide Tools ,Strength ,Regression analysis ,PSO-ANFIS ,Forecasting - Abstract
In this article, the adaptive neuro-based fuzzy inference system (ANFIS) model is developed to estimate the surface roughness (Ra) and tool vibrations (Acc) of AA6013 aluminum alloy during dry turning. Turning experiments were carried out with seven different cutting speeds, five different feed rates and seven different depth of cuts. These three different cutting parameters were tested with each other in different variations. ANFIS model is optimized using the genetic algorithm (GA), particle swarm optimization (PSO) and ant colony optimization. Performance of the developed model is compared with that of multi-linear regression model, which is one of the conventional prediction approaches. At the end of the study, it is revealed that the GA-ANFIS with an R-value of 0.946 is seen as the best model among the proposed approaches in the estimation of Acc. The PSO-ANFIS with an R-value of 0.916 is seen as the best model among the proposed approaches in the estimation of Ra. GA-ANFIS model for Acc prediction and PSO-ANFIS model for Ra prediction are the best approaches among the models discussed in the study. Moreover, the relationship between Acc and Ra values was examined and an empirical model was proposed.
- Published
- 2022
- Full Text
- View/download PDF