Multi-responses optimization of ultrasonic machining process

Abstract

The present work deals with multi-responses optimization of ultrasonic machining process using the imperialist competitive algorithm (ICA). Here, the process inputs were tool material, grit size and power rating of ultrasonic horn. Also, the main responses which should be simultaneously optimized were material removal rate (MRR), tool wear rate (TWR) and surface roughness (SR). Firstly, numbers of 24 experiments were conducted to collect data according to full factorial design. Then, obtained data were used to develop mapping relationship between inputs and responses based on adaptive neuro-fuzzy inference system (ANFIS). In order to maximize the MRR and minimize the TWR and SR simultaneously, the developed ANFIS models of MRR, TWR and SR were associated with ICA. Results indicated that the titanium tool, grit size of 423 and power rating of 328 W are the optimal solution which caused MRR of $$0.85\, \hbox {mg}/\hbox {min}$$0.85mg/min, TWR of $$0.22\,\hbox {mg}/\hbox {min}$$0.22mg/min and SR of $$0.65\,\upmu \hbox {m}$$0.65μm. Therefore, the ANFIS-ICA is a potential method which can be applied for optimization of multi-responses manufacturing problems in which characteristics are highly correlated to each other.