Multi-parameters optimization for electromagnetic acoustic transducers using surrogate-assisted particle swarm optimizer

Abstract

In ultrasonic testing field, the design and optimization of point-focusing shear vertical waves electromagnetic acoustic transducer (PFSV-EMAT) with high energy conversion efficiency have been a challenging task. In this work, a novel multi-parameter optimization method is described to improve PFSV-EMATs’ conversion performance, which integrates the newly proposed hybrid surrogate modeling (HSM) approach and particle swarm optimization (PSO) algorithm for efficient optimization. Based on the established finite element model of EMATs, the amplitudes of the displacement components at the observation point of a plate is the objective function to be maximized with five parameters pertaining to the cylindrical magnets, coaxial meander-line coils, and excitation signal, as design variables. Then, a brand-new HSM-assisted PSO algorithm is proposed and the superiority and rationality of the HSM strategy are proven. In addition, the multi-parameter optimization of PFSV-EMATs is conducted with the proposed HSM-PSO algorithm. Finally, experiments are carried out to verify the validation of optimization results, indicating that the increased signal amplitude of 12 times can be achieved by the optimized PFSV-EMAT, which has the improved signal amplitude and consistent point-focusing behavior, compared with the non-optimized one.