Nitric oxide(NO), nitrogen dioxide (NO2), nitrous oxide (N2O), and their derivatives generally known as nitrogen oxides (NOx) are primary pollutants in the atmosphere originated from natural and anthropogenic sources. The paper presents investigation of electric performance of novel chemiresistor NOx gas sensors. A novel material was utilized for active sensing layer-conductive copolymer and zinc oxide blend. The main advantage of the presented solution is low-cost and environment-friendly production. A series of this type of sensors was manufactured and tested experimentally. During the tests, the gas flow was controlled and signals of sensor responses, temperature, and humidity were computer-acquired using LabVIEW program. Sensor behavior for different thicknesses of the active layer has been investigated and interpreted. The research revealed that the electrical resistance of the sensors has changed in predictable manner depending on the gas concentrations. A recurrent artificial neural network architecture is proposed as a mathematical model to classify sensor responses to gas concentrations variation in a time-dependent regime. In this research, an enhanced method for gas concentration prediction is proposed using non-linear autoregression model with exogenous input (NARX). The performed simulations show good agreement between simulated and experimental data useful for predictions of sensor gas response.
Chemiresistor gas sensors based on conductive copolymer and ZnO blend – prototype fabrication, experimental testing, and response prediction by artificial neural networks
Capizzi G.;Lo Sciuto G.
2022-01-01
Abstract
Nitric oxide(NO), nitrogen dioxide (NO2), nitrous oxide (N2O), and their derivatives generally known as nitrogen oxides (NOx) are primary pollutants in the atmosphere originated from natural and anthropogenic sources. The paper presents investigation of electric performance of novel chemiresistor NOx gas sensors. A novel material was utilized for active sensing layer-conductive copolymer and zinc oxide blend. The main advantage of the presented solution is low-cost and environment-friendly production. A series of this type of sensors was manufactured and tested experimentally. During the tests, the gas flow was controlled and signals of sensor responses, temperature, and humidity were computer-acquired using LabVIEW program. Sensor behavior for different thicknesses of the active layer has been investigated and interpreted. The research revealed that the electrical resistance of the sensors has changed in predictable manner depending on the gas concentrations. A recurrent artificial neural network architecture is proposed as a mathematical model to classify sensor responses to gas concentrations variation in a time-dependent regime. In this research, an enhanced method for gas concentration prediction is proposed using non-linear autoregression model with exogenous input (NARX). The performed simulations show good agreement between simulated and experimental data useful for predictions of sensor gas response.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.