Development of a High-Performance Piezoelectric Single-Crystal Accelerometer by Using Gradient Optimization Method

A piezoelectric single-crystal accelerometer with a novel tribeam structure [tribeam ring accelerometer (TBRA)] has been developed, and a dual-objective optimization algorithm based on the gradient method (DOA/G) is proposed to improve the performance of TBRA. Both sensitivity and resonant frequency can be enhanced simultaneously from a global perspective by using DOA/G, and the optimized sensitivity and resonant frequency are increased by 50.7% and 31.9%, respectively, compared with the inertial parameters. The TBRA prototype has been fabricated and characterized. The obtained voltage sensitivity is 6.12 mV/ $\text{m}\cdot \text{s}^{\text {-2}}$ , and the resonant frequency is 3890 Hz, which is consistent with the simulations. Measurements of the device’s performance demonstrate a noise level of 27 ng $/\sqrt {\text {Hz}}$ at 200 Hz, and a dynamic range exceeding 140 dB. In addition, the proposed optimization approach provides a new idea for improving the overall performance of the other similar sensors, such as hydrophones, energy harvesters, and vibration sensors.