КОМПЕНСАЦИЯ ВЛИЯНИЯ ТЕМПЕРАТУРЫ В АВИАЦИОННЫХ ПЬЕЗОРЕЗИСТИВНЫХ ДАТЧИКАХ ДАВЛЕНИЯ

Abstract

Currently, piezoresistive pressure sensors (PDS) find expanded applications in various microelectronicdevices used in aviation technology. The behavior of the electrical signal of suchPDS mainly depends on the ambient temperature. It is known that the temperature drift of the PDSoutput signal is influenced by various factors: the temperature effect, the dependence of the resistanceof the sensitive element on the concentration of impurities, the dependence of the Young'smodulus of the sensor membrane and substrate materials on temperature, etc. It was found that thepreviously developed analytical calibration model of the sensor output signal, which takes intoaccount the models describing individual temperature effects, does not allow the pressure to bemeasured with the required accuracy in the temperature range characteristic of aviation equipmentfrom –60 C to 140 C. Therefore, conventional polynomial mathematical models are used todescribe the dependence of the PDS output signal on the measured pressure and temperature.The work uses a traditional approach, when the dependence of the output voltage on pressure is represented using a polynomial of relatively low order, and the dependences of the coefficients ofthis polynomial on temperature are also specified by the corresponding polynomials. Unfortunately,the temperature dependences of the coefficients are adequately described only by high-orderpolynomials (at least 7), which complicates the model identification procedure and leads to computationerrors. Therefore, the authors proposed to look for the dependence of the coefficients ontemperature in the form of cubic splines. The paper describes in detail the identification techniqueof the polynomial model under consideration and obtains expressions for correcting the PDS readingswhen measuring pressure in wide temperature ranges. In order to experimentally confirm theefficiency of the proposed method, an intelligent industrial automated system for the calibration oftraffic rules, described in the work, was used. It is shown how it can be used to take experimentaldata to calibrate the sensor readings over a wide temperature range, and the procedure for identifyingthe mathematical model of the pressure sensor required to minimize the cost of its certificationis described. The results of experimental studies of specific pressure sensors used in aviationtechnology are presented.