Machinability Analysis and ANFIS modelling on Advanced Machining of Hybrid Metal Matrix Composites for Aerospace Applications. Issue 16 (10th December 2019)

• Record Type: Journal Article
• Authors: Manikandan, N.; Balasubramanian, K.; Palanisamy, D.; Gopal, P.M.; Arulkirubakaran, D.; Binoj, J.S.
• Is Part Of: Materials and manufacturing processes. Volume 34:Issue 16(2019)
• Page Start: 1866
• Page End: 1881
• Publication Date: 2019-12-10
• Subjects: Hybrid -- metal -- matrix -- composite -- material -- removal -- surface -- finish -- form -- orientation -- tolerance -- GRA -- ANFIS -- prediction; Manufacturing processes -- Periodicals; Materials -- Periodicals; Manufactured Materials; 670.5
• DOI: 10.1080/10426914.2019.1689264
• Languages: English
• ISSNs: 1042-6914

Abstract

ABSTRACT: Wire Electrical Discharge Machining (WEDM) is a competent method employed for machining intricate shapes in electrically conductive harder materials. Metal Matrix Composites (MMCs) possesses improved properties and considered as an alternate material for various engineering applications. An enduring issue with MMCs is that they are problematic for machining, because of abrasive nature of hard reinforcing phase. This current exploration is dealing the WEDM machinability analysis of LM6/SiC/Dunite Hybrid Metal Matrix Composites (HMMC) fabricated by two step stir casting process and to develop an artificial intelligence decision making model for WEDM the process parameters. The influence of input machining process variables namely Pulse ON, Pulse OFF, Flushing Pressure, Wire Feed and Servo Voltage against the desired output like Material Removal Rate (MRR), Surface finish, overcut, circularity error and perpendicularity error are investigated. Grey Relational Analysis (GRA) method is employed and grey relational coefficient values are given as input values for evolving the Adaptive Neuro Fuzzy Inference System (ANFIS) to predict the desired performance characteristic. A comparative study has been accomplished for validating the outcomes attained from evolved models and experimentation outcomes. The performance analysis of developed model proves that the developed model and this method are efficacious for predicting the desired performance measures.