Design and Implementation of a Self-Validating Pressure Sensor

• Published in: IEEE sensors journal 2009-03, Vol.9 (3), p.207-218
• Main Authors: Zhigang Feng, Qi Wang, Shida, K.
• Format: Article
• Language: English
• Subjects: Capacitive sensors; Consistency checking; Diagnosis; Digital signal processing; Digital signal processors; double DSPs system; Fault detection; fault detection and diagnosis; Fault diagnosis; Faults; Mathematical analysis; Pressure measurement; Pressure sensors; Prototypes; self-validating pressure sensor; Semiconductors; Sensor phenomena and characterization; Sensor systems; Sensors; Strain gauges; support vector machine (SVM) multiclassification; Temperature sensors; Transducers; wavelet packet decomposition
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2008.2011949

A new prototype of self-validating pressure sensor is described, including structure design of the transducer, fault detection and diagnosis method and signal transmitter based on double digital signal processors (DSPs) system. The elastic body of the transducer is based on circular-flat diaphragm structure. Eight groups of strain gauges are distributed on the diaphragm. One group of metal strain gauge with little temperature drift is used as primary pressure sensor, the other seven groups of semiconductor strain gauges with high sensitivity are used as calibration references. Consistency checking of the measured data is used to detect sensor fault. The outputs of self-validating pressure sensor including validated measurement value (VMV) of pressure, validated uncertainty (VU) of pressure, and measurement value status (MVS) of pressure are calculated using the high consistency data. A new fault diagnosis method based on wavelet packet decomposition feature extraction and support vector machine multiclassification is used to identify six sensor fault status. A double DSPs system is implemented using TMS 320 F 2812 and TMS 320 C 6713 to acquire the output signal and complete fault detection, diagnosis, and self-validating parameters calculating. The experiment results show that the designed pressure sensor prototype implements the self-validating function. It can detect fault in real-time and identify sensor status with high accuracy.