Development of MEMS composite sensor with temperature compensation for tire pressure monitoring system. (16th November 2021)

• Record Type: Journal Article
• Authors: Zhang, Jiahong; Wang, Chao; Xie, Xiaolu; Li, Min; Li, Ling; Mao, Xiaoli
• Is Part Of: Journal of micromechanics and microengineering. Volume 31:Number 12(2021)
• Publication Date: 2021-11-16
• Subjects: TPMS -- aluminum–silicon hybrid structure -- MEMS -- temperature compensation -- neural network; Microelectromechanical systems -- Periodicals; Micromechanics -- Periodicals; 621.38105
• DOI: 10.1088/1361-6439/ac349d
• Languages: English
• ISSNs: 0960-1317

Abstract

Abstract: The pressure and temperature inside the tire is mainly monitored by the tire pressure monitoring system (TPMS). In order to improve the integration of the TPMS system, moreover enhance the sensitivity and temperature-insensitivity of pressure measurement, this paper proposes a microelectromechanical (MEMS) chip-level sensor based on stress-sensitive aluminum–silicon hybrid structures with amplified piezoresistive effect and temperature-dependent aluminum–silicon hybrid structures for hardware and software temperature compensations. Two types of aluminum–silicon hybrid structures are located inside and outside the strained membrane to simultaneously realize the measurement of pressure and temperature. The model of this composite sensor chip is firstly designed and verified for its effectiveness by using finite element numerical simulation, and then it is fabricated based on the standard MEMS process. The experiments indicate that the pressure sensitivity of the sensor is between 0.126 mV/(V·kPa) and 0.151 mV/(V·kPa) during the ambient temperature ranges from −20 °C to 100 °C, while the measurement error, sensitivity and temperature coefficient of temperature-dependent hybrid structures are individually ±0.91 °C, −1.225 mV/(V °C) and −0.150% °C −1 . The thermal coefficient of offset (TCO) of pressure measurement can be reduced from −3.553%FS °C −1 to −0.375%FS °C −1 based on the differential output of the proposed sensor. In order to obtain the better performance of temperature compensation, Elman neural network based on ant colony algorithm is applied in the data fusion of differential output to further eliminate the temperature drift error. Based on which, the overall measured error is within 3.45 kPa, which is less than ±1.15%FS. The TCO is −0.017%FS °C −1, and the thermal coefficient of span is −0.020%FS °C −1 . The research results may provide a useful reference for the development of the high-performance MEMS composite sensor for the TPMS system.