Development of Piezoresistive Micropressure Sensor Based on Grooved Diaphragm with Back Peninsulas and Trenches

To validate the effectiveness of the grooved diaphragm with back peninsulas and trenches (GDPT) and the radial basis function neural network (RBFNN)-based dimension generator in the development of a 1 kPa piezoresistive micropressure sensor, this paper presents a comprehensive investigation into the design, fabrication, and characterization of the anticipative sensor prototype. The GDPT achieves favorable sensing stress and reduced deflection, enabling a favorable trade-off between sensitivity and nonlinearity; the RBFNN-based generator efficiently determines practicable dimensions for the complex structure, offering a significant improvement over the conventional trial-and-error process. Characterization results demonstrate that the fabricated sensor achieves sensitivity of 13.01 mV/(V·kPa) and nonlinearity of 0.26% FS within the pressure range of 0–1 kPa, in good agreement with the preset design target. This work verifies the validity of the proposed approach and offers a holistic methodology for developing high-performance piezoresistive sensors.