Gas identification using gas sensor array and self organizing competitive neural network

In order to identify hydrogen and carbon monoxide gases in the concentration range of 40 to 1000 ppm (10(-6)), the gas recognition system composed of a SnO2-based thick-film gas sensor array and the pattern recognition techniques of Self Organizing Competitive Network is developed. An on-line data acquisition system is built for measurement of dynamic and steady state in information across the sensor array, and three kinds of data normalization methods are used to pre-process the sensors responses for the discrimination of gas species and concentrations to get good separation among gases. The results are presented in this paper that the accurate identification of hydrogen and carbon monoxide in the finite range could be accomplished by the third normalization method.