Optimizing kerf quality in high-speed WEDM of thin woven CFRP composites: a taguchi, WASPAS, and PSO approach

Abstract

Listen

Translate article

The process of machining CFRP composites presents unique challenges, particularly in the context of WEDM. The inherent properties of CFRP composites, such as their low electrical conductivity, anisotropic nature, and heterogeneous composition, require further research to enhance their machinability through WEDM techniques. This study examines the enhancement of kerf characteristics such as kerf width (Wk), delamination factor (DFK), and cutting speed (CSK) in thin woven 0°/90° CFRP composites using high-speed WEDM. A Taguchi L16 experimental analysis was employed to analyze the impact of key process parameters, including pulse-on (Pon), pulse-off (Poff), and input current (I), in conjunction with CFRP parameters such as the CFRP thickness (T) and cutting direction on the kerf characteristics. The CFRP thickness ranged from 0.5 to 2.0 mm, and the cutting directions studied were horizontal and inclined 30° cuts. A multiple-response optimization strategy using the CRITIC-WASPAS approach coupled with a particle swarm optimization (PSO) algorithm were applied to identify the ideal process combination for various CFRP thicknesses. The findings indicated that the CFRP thickness, pulse-off time, and input current are the most statistically significant factors influencing the overall kerf characteristics. The cutting direction has a negligible effect on the kerf width but has conflicting effects on the delamination factor and cutting speed. Specifically, a horizontal cut decreases delamination, whereas an inclined 30° cut is preferable for achieving higher cutting speeds. For precise kerf cutting, optimal process combinations were determined: Pon (30 µs), Poff (30 µs), and I (ranging from 4 to 5 A) for 0.5 mm CFRP thickness, and Pon (30 µs), Poff (15 µs), and varying input currents of 4 A, ranging from 4 to 3 A, and 3 A for CFRP thicknesses of 1.0, 1.5, and 2.0 mm, respectively.