Application of Artificial Neural Networks and Adaptive Neuro‐Fuzzy Inference Systems for Attenuating Effects of Temperature and Humidity on Sensor Accuracy

Abstract

AbstractThe novelty of this work is the correction of measurement errors in a galvanic cell sensor, which might occur due to temperature and humidity variations, using artificial neural networks (ANN) and adaptive neuro‐fuzzy inference systems (ANFIS). The training process for the ANN and ANFIS employed 540 data sets consisting of voltage responses of the galvanic cell to known oxygen concentrations, temperatures, and relative humidities. The trained ANFIS was embedded within a central processing unit (CPU) and tested on another 144 data sets with various oxygen concentrations, temperatures, and relative humidities. The deviations for the sensor with ANFIS were less than 0.5 % for 109 data points (75 %), and there was no deviation greater than 2.5 %.