TiO2 Thick film Gas sensor for Detection H2S Gas Using ANN and Machine Learning Technique

Undoped CuO doped thick film gas sensor have been developed for H2S toxic gas detection to review the sensitivity and sensor response using ANN technique at 150°C . TiO2 based thick film sensor was untrue on a 1" x 1" alumina substrate. It incorporate of a gas sensitive layer TiO2 based thick film sensor with doped of undoped CuO, a couple of electrodes in radical to gas sensing layer serving as a channel pad for sensor. The sensitivity of sensor has been investigated at undoped CuO-doped concentration at constant temperature of 150°C upon liability of H2S toxic gas .An advanced approach is made to measure the sensitivity of undoped CuO-doped TiO2 based thick film sensor by using ANN algorithm.The training algorithm of feed –forward algorithm namely with learning heuristic was used. The performance of ANN models with specific algorithm is evaluated on reasonable sensitivity of sensor with different network transfer function. Empirically, we found that ANN model with training algorithm is more advisable for simulation of sensor and predict the sensitivity. Simulation results demonstrated in the paper shown ANN as an effective tool in the area of TiO2 based thick film sensor design.