Nonlinear inverse filtering for measurement of thermal hysteresis

A nonlinear inverse filter for distortion correction in the measurement of thermal hysteresis is proposed in this paper. The temperature measurement sensor is a thermistor, which has static nonlinearities and slow dynamic response. When the rate of change of temperature is high as compared with the speed of response of the thermistor there is a degradation in the measured temperature and the hysteresis curve thus obtained is a distorted version of the true characteristic. To compensate this effect, we employ a infinite impulse response IIR filter to remove the distortion. The filter coefficients are obtained from experimentally measured characteristics, and the filtered hysteresis data is compared with the quasi-static experimental hysteresis curve.