Linearizing a thermistor characteristic in the range of zero to 100 degree C with two layer artificial neural networks

Abstract

Abstruct - Artificial neural networks appear as an efficient tool to correct input-output nonlinearities of sensors. In this paper, an artificial neural network ((ANN) with two hidden layers to linearize a static characteristic of a thermistor is discussed. The data used were taken from the input-output calibrating thermistor bridge. Both backpropagation (BP) and random optimization method (ROM) have been combined to adjust the weights of the neural network. Simulation results show effectiveness and ability of the method suggested to linearize a thermistor characteristic in the range of zero to 100 degree C.