Work

RSM and machine learning-based optimization of vitrified grinding wheels from furnace slag–ceramic waste blends

Year: 2026

Type: article

Abstract: The increasing cost, limited availability, and environmental concerns associated with conventional grinding wheel materials have necessitated the exploration of alternative, sustainable options. In this study, a novel abrasive grinding wheel was developed using furnace slag and ceramic wastes, employing the vitrified bonding method. The aim was to formulate a cost-effective and high-performance grinding wheel while promoting the recycling of industrial by-products. A D-optimal mixture design was utilized to generate experimental formulations, and the grinding wheels were produced with fixed proportions of silicon carbide (95 wt. %) and sulfur (1 wt. %) along with varying percentages of furnace slag and ceramics. Mechanical tests, including hardness, compressive strength, and wear rate, were conducted alongside thermal analysis using thermogravimetric analysis, microstructural characterization using scanning electron microscopy, and elemental composition determination via X-ray fluorescence. Response Surface Methodology (RSM) and a machine learning model based on Random Forest with Bayesian optimization were employed for multi-objective optimization of the input factors. The RSM yielded an optimal formulation of 1.53 wt. % furnace slag and 2.47 wt. % ceramics, while the ML model predicted optimal values at 1.567 and 2.538 wt. %, respectively. The grinding wheel produced using the ML-optimized formulation demonstrated excellent agreement between predicted and experimental values: hardness of 93.5 HV, compressive strength of 3.65 MPa, and wear rate of 1.54 m3/N m, outperforming the control wheel in wear resistance. Statistical evaluation revealed that the ML model provided superior prediction accuracy across all performance indices compared to RSM. The study concludes that integrating industrial waste materials with AI-driven optimization offers a viable pathway toward the sustainable and high-performance development of grinding wheels. Future work is recommended to explore extended curing protocols and alternative waste materials for broader industrial applications.

Source: AIP Advances

DOI: 10.1063/5.0307084

Authors: I. E. Ekengwu, O. C. Okafor, K. C. Owuama, P. N. Atanmo

Institutions: Nnamdi Azikiwe University, Imo State University, Chukwuemeka Odumegwu Ojukwu University

Language: English

Topic: Recycling and utilization of industrial and municipal waste in materials production

Subfield: Building and Construction

Field: Engineering

Domain: Physical Sciences

SDGs: Responsible consumption and production, Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $1,350