Tabu search coupled with deterministic strategies for the optimal design of MRI devices

Article type: Research Article

Authors: Fanni, A.^a | Giacinto, G.^a | Marchesi, M.^a | Serri, A.^a

Affiliations: [a] Dipartimento di Ingegneria Elettrica ed Elettronica, Università di Cagliari, Piazza d’Armi, 09123 Cagliari, Italy

Abstract: A tabu search based strategy has been developed to achieve design optimisation of electromagnetic structures. Tabu Search (TS) is a metaheuristic that guides the search for the optimal solution making use of flexible memory systems which exploit the history of the search. It consists of systematically prohibiting some solutions to prevent cycling and, at the same time, avoiding to be trapped in local minima, and deeply explores the search space to minimise the objective function. Since the variables to be optimised are continuous in nature, we developed a continuous tabu search algorithm in which the range of variation of each variable is discretised by division into sub-ranges. In this way, it is possible to identify a finite alphabet and apply tabu search strategies. Best performances have been reached by coupling tabu search with other optimisation procedures, such as Golden Search (GS) and a non-linear Simplex (SX) method. The algorithm is compared with more traditional methods on a Magnetic Resonance Imaging (MRI) design problem, showing a good performance improvement.

DOI: 10.3233/JAEM-1999-126

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 10, no. 1, pp. 21-31, 1999