Skip to main navigation Skip to search Skip to main content

Portable vehicular electronic nose system for detection of automobile exhaust

  • Qi Wang*
  • , Kai Song
  • , Tiandong Guo
  • *Corresponding author for this work
  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we developed a powerful vehicular electronic nose system for detection of automobile exhaust gases such as monoxide and hydrocarbon (CO/HC). Commercially available metal oxide semiconductor (MOS) multi-sensor system and artificial neural network based gas pattern recognition method were used to improve selectivity of gas sensors and accurately discriminate gas components. A singlechip was used for sampling and processing sensor response data as well as gave the automobile exhaust detection result. The classification for emitted gases was based on a momentum and adaptive learning rate backpropagation (BP) artificial neural network whose weights and biases were trained in advance and programmed in the microcontroller unit (MCU). Experimental results demonstrate that the system not only could effectively detect the individual components from their mixtures but also could monitor the risk grade of each gas with sufficient accuracy.

Original languageEnglish
Title of host publication2010 IEEE Vehicle Power and Propulsion Conference, VPPC 2010
DOIs
StatePublished - 2010
Event2010 IEEE Vehicle Power and Propulsion Conference, VPPC 2010 - Lille, France
Duration: 1 Sep 20103 Sep 2010

Publication series

Name2010 IEEE Vehicle Power and Propulsion Conference, VPPC 2010

Conference

Conference2010 IEEE Vehicle Power and Propulsion Conference, VPPC 2010
Country/TerritoryFrance
CityLille
Period1/09/103/09/10

Keywords

  • Artificial neural network
  • Automobile exhaust
  • Emission detection
  • Gas sensor
  • Vehicular electronic nose

Fingerprint

Dive into the research topics of 'Portable vehicular electronic nose system for detection of automobile exhaust'. Together they form a unique fingerprint.

Cite this