New Materials, Compounds and Applications

ISSN Print: 2521-7194

ISSN Online: 2523-4773

New Materials, Compounds and Applications is an open access, strictly peer reviewed journal that is devoted to publication of the reviews and full-length papers recording original research results on, or techniques for, studying the relationship between structure, properties of materials and compounds and their applications. Materials include metals, ceramics, glasses, polymers, energy materials, electrical materials, composite materials, fibers, nanostructured materials, nanocomposites, and biological and biomedical materials.

Submission Current Issue Archive Special Issues

Multi Objective Optimization and Prediction During Machining of Mg-TiO₂ Nanocomposite using Grey Relational Analysis (GRA) and Artificial Neural Network (ANN) Techniques

Published: 04-08-2026

https://doi.org/10.62476/nmca102319

Abstract

Magnesium-based metal matrix composites (Mg-MMCs) provide corrosion resistance, elevated-temperature strength and lightweight characteristics for a range of applications. This research integrates Taguchi-based Grey Relational Analysis (GRA) with Artificial Neural Networks (ANN) to assess, optimize and predict the machinability of Mg-TiO₂ nanocomposites. An L9 orthogonal array was employed to investigate the effects of TiO₂ nanoparticulate weight fraction, cutting speed and feed rate on cutting force, temperature and surface roughness. GRA identified 1.5 wt.% TiO₂, 1000 rpm and 5 mm/min as the optimal settings for machinability. ANOVA indicated that cutting speed substantially influenced force and temperature, whereas feed rate predominantly affected surface roughness. The ANN model achieved correlation coefficients exceeding 0.99, accurately forecasting performance metrics with minimal errors. GRA-based multi-response optimization on an L9 design is integrated with an ANN to translate a discrete optimum for Mg-TiO₂ machining. The ANN estimates cutting force and temperature to support data informed selection of process set points. The integrated GRA-ANN framework thus offers a dependable approach for optimizing machining parameters and forecasting machining performance in Mg-TiO₂nanocomposites.