Optimization design of high-frequency ultrasonic transducer based on ANFIS and particle swarm optimization algorithm

Abstract

In this study, an optimization design method based on adaptive neuro-fuzzy inference system (ANFIS) and modified particle swarm optimization (PSO) algorithm is proposed to enhance the performance of high-frequency ultrasonic transducers (UTs). The optimization indexes include center frequency (CF) and −6 dB bandwidth (BW), and the design parameters (DPs) include thickness (TP) and size (SP) of piezoelectric layer, thicknesses of matching layers (Ag-epoxy and Parylene C). Based on simulated results, an ANFIS model is established to characterize the effect of DPs on the performance indexes (PIs). Optimization targets are set as CF of 30 MHz and BW of 65%, the modified PSO algorithm is adopted to optimize the design parameters of high-frequency UT. The optimized TP and SP are 110 μm and 5.2 mm. The optimized thicknesses of Ag-epoxy (TAE) and Parylene C (TPC) are 16 μm and 20 μm, respectively. Simulations and experiments of high-frequency UT are conducted based on the optimized DPs, the results well agree with the designed ones. The CF and the BW of UT fabricated by the developed method are 26.77(±0.02) MHz and 65.11(±0.06) %, while the CF and the BW of UT fabricated by the traditional method are 23.42 MHz and 51.4 %.