Modelling and Multi-Objective Optimization during ECDM of Silicon Carbide Reinforced Epoxy Composites

Abstract

The silicon carbide (SiC) reinforced epoxy composites has evidenced their reputation as advanced composite material in domain of strength and durable nature. But, presence of SiC particles as reinforcement in epoxy composites wanes the machinability by increasing tool wear in conventional drilling process. For the reduction in tool wear and improvement in machinability, it becomes necessary to use non-conventional machining method which can elude direct tool-composite interaction. In domain of non-conventional machining methods, the electrochemical discharge machining (ECDM) process is emerging as potential contender for machining of nonconductive, hard and brittle fibrous materials. In this paper, artificial neural network-based hybrid model was generated for prediction and optimizing material removal rate (MRR) and taper during ECDM of SiC reinforced epoxy composites. The input parameters considered for the process were selected as voltage, electrolyte concentration, inter-electrode gap and duty factor whereas MRR and taper were obtained as response parameters. The experimentation was performed based on response surface methodology (RSM) using central composite design. RSM and ANN models were used to predict the MRR and taper during ECDM. The input parameters were optimized for higher MRR and minimum taper using desirability function. The obtained results for optimal solution were validated with the neural network model. The validation exemplifies the influence that machining methods focussed on artificial intelligence have on optimizing progressions.