Journals Proceedings

Abstract

This paper aims at designing an intelligent temperature measuring technique by Resistance Temperature Detector (RTD) using optimal Artificial Neural Network (ANN). The objectives of the present work is (i) to extend the linearity range of RTD and (ii) to provide intelligence in the measuring technique so as to measure the temperature under variation (within a pre-specified range) of temperature co-efficient parameters Ro, a, b and c, but without any change in calibration circuit. (iii) To achieve the objective (i) and (ii) by an optimal ANN. An optimal ANN is considered based on minimum mean square error (MSE) and Regression by comparing various scheme and algorithm. The proposed technique provides linear relationship of the overall system over a wider range and makes it independent of temperature co-efficient. Since, the proposed intelligent temperature measuring scheme produces output independent of physical properties of RTD, it avoids the requirement of repeated calibration every time the RTD is replaced.