This research investigates the use of Stir-squeeze-cast can be effectively machined using high-speed wire electric Hybrid Aluminium Matrix Composites (HAMCs), specifically AA 024, using nanoparticles of ceramic ( \(Al_{2} O_{3}\) , \(SiC\) , \(Si_{3} N_{4}\) , \(BN\) ). Because of the reinforcements’ natural hardness and abrasiveness, HAMCs are difficult to mill conventionally, despite their significant value in the industrial sector. By using a variety of machining variables, the study aims to create complicated surfaces with superior degradation properties and evaluate the erosion performance in terms of WWR (wire wear ratio) and MRR (material removal rate) for various profiles (curve, angular, and plane). A model for the WEDM process applied to HAMCs is presented, utilizing a Neural Network with Temporal Inductive Paths (TIPNN) optimized with the Starfish Optimization Algorithm (SOA). The process begins with the collection of a comprehensive dataset consisting of key machining variables such as drum speed \((D_{S} )\) , wire feed rate \((W_{FR} )\) , pulse voltage \((P_{V} )\) , pulse \((P)\) , and pulse current \((P_{I} )\) angular, and plane machining profiles. After pre-processing the data by normalizing inputs and outputs, handling missing values, and removing outliers, the TIPNN to capture the dynamic interactions between the input data, and a model is built and machining outcomes. The model’s performance is enhanced using SOA, a nature-inspired optimization technique that fine-tunes the network’s weights and adjusts machining variables to achieve optimal MRR and WWR. The proposed TIPNN-SOA model is assessed and contrasted with current techniques like genetic-integrated neural networks (HAMC), DS-EDM optimization strategies, and hybrid Grey-ANFIS techniques, demonstrating its superior performance in improving machining outcomes.
