Enhancing EDM performance on TiN-Si3N4 using a hybrid computation intelligence algorithm (Grey-ANFIS)

Abstract

This study investigated the optimization of the electrical discharge machining (EDM) process for TiN-Si₃N₄ composites, a challenging and emerging field in materials engineering. To achieve superior machining efficiency and product quality, a hybrid computational intelligence algorithm, Grey-ANFIS (adaptive neuro-fuzzy inference system), was employed. First, comprehensive data on EDM process parameters and performance characteristics were collected. Then, Grey-ANFIS was used to model the complex relationships between the EDM process parameters (e.g., pulse on time, pulse off time, voltage, and current) and key performance indicators (e.g., material removal rate and electrode wear rate). The algorithm combined the adaptability of neural networks with the linguistic representation capabilities of fuzzy logic, making it well-suited for capturing the intricate, non-linear EDM process dynamics. The proposed approach enables the generation of precise predictive models that can accurately represent EDM process behavior. Subsequently, these models were employed to optimize EDM process parameters, thereby enhancing machining efficiency and product quality. A sensitivity analysis was also conducted herein to identify critical factors affecting the EDM process. The results demonstrated the efficacy of the Grey-ANFIS algorithm in achieving superior EDM process optimization for TiN-Si₃N₄ composites.