Parametric optimization for electrical discharge diamond grinding (EDDG) system using dual approach.

Area of Science:

• Manufacturing Engineering
• Materials Science
• Artificial Intelligence

Background:

• Electrically conductive materials are vital for many applications due to their strength and stiffness.
• Electrical Discharge Diamond Grinding (EDDG) is a key method for producing these materials.
• Traditional Artificial Neural Network (ANN) models often face performance issues due to suboptimal hidden layers and weights.

Purpose of the Study:

• To introduce and evaluate the Modified Ant Lion Optimization- Artificial Neural Network (MALO-ANN) technique.
• To enhance the performance and parametric optimization of the EDDG process.
• To investigate the impact of input factors on Material Removal Rate (MRR) and Surface Roughness (SR).

Main Methods:

• The study employed the Modified Ant Lion Optimization (MALO) algorithm to optimize the weights and hidden layers of an Artificial Neural Network (ANN).
• Input parameters including grit size, pulse-on/off duration, and current were systematically analyzed.
• The MALO-ANN model was applied to predict and optimize MRR and SR in the EDDG process.

Main Results:

• The MALO-ANN technique demonstrated significant improvements in the parametric optimization of EDDG.
• The optimized model achieved high accuracy, with an absolute error interval for MRR and SR ranging from 1.03% to 4.49%.
• A convergence rate of 89% was achieved, indicating enhanced efficiency and accuracy in EDDG operations.

Conclusions:

• The MALO-ANN approach offers a superior method for optimizing EDDG processes compared to conventional ANN models.
• This technique shows great potential for improving the efficiency and precision of manufacturing durable, electrically conductive materials.
• The study validates the effectiveness of MALO-ANN in achieving optimal material removal rate and surface roughness.