Genetic algorithm-based multi-objective optimization of an exciter for magnetic induction tomography


Abstract

Purpose - The purpose of this paper is to present the methodology of designing an exciter for magnetic induction tomography (MIT). The design of the exciter must satisfy following requirements: maximize MIT system sensitivity and minimize harmful influence on electronic MIT equipment. Design/methodology/approach - Two objective functions are considered, namely: a magnetic flux density in the protected regions and a module of the eddy current density vector in the object under test in the vicinity of a sensor. The paper demonstrates a multi-objective optimization technique (based on the e-constrained method) which, by coupling the finite-element method with a genetic algorithm (GA), supports the design of the exciter. Findings - It is possible to design in a relatively simple way an exciter for MIT under the given assumptions. Originality/value - The paper is of value in presenting a detailed description of the multi-objective optimization procedure.