Geometric influence and multi-objective optimization of WEDM for hardened tool steels using ANN and NSGA-II

Area of Science:

• Materials Science and Engineering
• Manufacturing Technology

Background:

• Hardened AISI D2 and DC53 tool steels offer superior strength and wear resistance, crucial for tool and die manufacturing.
• Conventional machining of these steels is challenging due to abrasive metallic carbides, necessitating advanced techniques like WEDM.
• Geometric features of workpieces can influence WEDM process outcomes.

Purpose of the Study:

• To investigate the influence of WEDM parameters on cutting speed (CS) and material removal rate (MRR) for flat and curved profiles of D2 and DC53 steels.
• To systematically evaluate performance variations across different geometric attributes using a Taguchi L18 orthogonal array.
• To develop predictive models and optimize WEDM processes for enhanced efficiency.

Main Methods:

• Utilized a Taguchi L18 orthogonal array to design experiments for evaluating machining parameters (peak current, voltage, pulse duration) and material types (D2, DC53).
• Employed scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) analysis to understand process physics.
• Applied artificial neural networks (ANN) for predicting output parameters and analysis of variance (ANOVA) for statistical analysis.
• Optimized parameters using a non-dominated sorting genetic algorithm II (NSGA-II).

Main Results:

• WEDM parameter variations significantly impact CS and MRR for both flat and curved profiles.
• ANN models effectively predicted WEDM outcomes, correlating experimental and anticipated values.
• NSGA-II optimization led to substantial improvements: 63.17% (CS, flat), 75.36% (MRR, flat), 74.10% (CS, curved), and 73.38% (MRR, curved) compared to unoptimized settings.

Conclusions:

• The study successfully identified optimal WEDM parameters for hardened D2 and DC53 steels, enhancing machining efficiency.
• The integration of Taguchi methods, ANN, and NSGA-II provides a robust framework for optimizing complex manufacturing processes.
• Achieved significant performance gains demonstrate the potential for improved productivity in tool and die manufacturing using optimized WEDM.