Studies on magnetorheological properties of carbonyl iron/Fe3O4 powder based soft-magnetic fluids with artificial intelligence for industrial applications

Abstract

For mechanical devices such as haptic gadgets, mechanical dampers, and brake systems to function properly, magnetorheological fluids (MRF)—also known as smart fluids—are crucial. An innovative method for optimizing the preparation process of magnetorheological fluid that is both cost-effective and data-driven in nature is investigated in this article. In this MRF preparation experiment, silicone oil is the carrier liquid and carbonyl iron and Fe3O4 powders are carefully selected as magnetic particles to prepare soft magnetic fluids. Sodium dodecyl benzene sulfonate, oleic acid, polyethylene glycol, graphite, and diatomite were added to improve the MRF's physical qualities. Then, the MRF properties were examined with artificial intelligence method and an enhanced grey wolf optimization algorithm is developed for improving the hyperparameters of the least square support vector machine. In addition, a new procedure for updating individual positions is suggested to better preserve population variety and increase solution quality, and a better updating technique is created for the convergence coefficient vector to avoid local optima. The results show that the suggested modified least square support vector machine is the best algorithm to predict the MRF properties.