Intelligent optimization design of 2–2 piezo-composites for ultrasonic transducer

Abstract 2–2 piezo-composite material (PCM) with high electromechanical coupling coefficient and low acoustic impedance is a kind of common PCMs for fabricating piezo-composite ultrasonic transducer (PCUT). The performance of PCUT is obviously influenced by the design parameters (DPs) of 2–2 PCM, such as its thickness, kerf, and slice width. Combining the finite element analysis simulation and artificial intelligence methods, an intelligent optimization design method is proposed to optimize the DPs of 2–2 PCM to fabricate high-performance 2–2 PCUT. In the proposed method, the neural network models are trained by the simulation data to build up the relationship between the DPs and the performance of PCUT. Using the optimization criteria established based on the center frequency (CF), bandwidth (BW), and peak-to-peak voltage (PV), the modified particle swarm optimization algorithm is employed to determine the DPs of 2–2 PCM. The desired CF, BW, and PV are 4 MHz, 40% and 3.2 V, and the optimized thickness, kerf and slice width of 2–2 PCM are 376 µm, 69.6 µm, and 34.4 µm, respectively. Finally, the optimized 2–2 PCUT is fabricated based on the optimized DPs, and then characterized systematically. The optimized 2–2 PCUT exhibits a CF of 4.57 MHz, a BW of 41.6%, a PV of 2.89 V, and an electromechanical coupling coefficient of 0.71. The simulation and experiment results are consistent with the designed targets, which verifies the availability of the proposed method.