Sensitivity analysis of drill wear and optimization using Adaptive Neuro fuzzy –genetic algorithm technique toward sustainable machining. (20th January 2018)

• Record Type: Journal Article
• Authors: Saw, Lip Huat; Ho, Li Wen; Yew, Ming Chian; Yusof, Farazila; Pambudi, Nugroho Agung; Ng, Tan Ching; Yew, Ming Kun
• Is Part Of: Journal of cleaner production. Volume 172(2018)
• Page Start: 3289
• Page End: 3298
• Publication Date: 2018-01-20
• Subjects: Drilling -- Tool wear -- Tool condition monitoring -- ANFIS -- Multi-objective optimization; Factory and trade waste -- Management -- Periodicals; Manufactures -- Environmental aspects -- Periodicals; Déchets industriels -- Gestion -- Périodiques; Usines -- Aspect de l'environnement -- Périodiques; 628.5
• DOI: 10.1016/j.jclepro.2017.10.303
• Languages: English
• ISSNs: 0959-6526

Abstract

Abstract: Machining processes have an important place in the manufacturing industry and it indeed contributed to the economic growth of a country. About 75% of machining processes involved drilling operation. Tool wear is a common phenomenon in the machining operation and significantly affects the product dimension accuracy, machining efficiency, manufacturing downtime, surface roughness and economic loss. Hence, an intelligent tool condition monitoring system is needed to maximize tool life and reduce machine downtime due to the tool replacement. In this study, experiments were conducted to investigate the influence of different drilling parameters on average drilling torque and thrust force. Effects of spindle rotational speed, feed rate and diameter of drill on tool wear were determined through Adaptive Neuro Fuzzy Inference System (ANFIS). Next, genetic algorithm (GA) was used to identify the optimal drilling parameter for different diameters of drill. Experimental results agreed well with the GA prediction results with a relative error of 3%. Hence, the results showed that ANFIS-GA is a faster and more accurate alternative to the existing methods for tool wear prediction. Highlights: We investigated the drill wear using cutting force signal. We conducted sensitivity analysis of the drill wear on mild steel. We used Adaptive Neuro Fuzzy Inference system to predict drill wear. We optimized the drilling parameters using genetic algorithm method.