ACTIVE GAS ODOR SENSING SYSTEM USING AUTOMATICALLY CONTROLLED GAS BLENDER AND NUMERICAL OPTIMIZATION TECHNIQUE

As measurement of a vapor mixture composition is a difficult technique , no method using a sensing system has yet been established in spite o f great effort by many researchers. In this paper, the authors propose a new gas/odor sensing system using a gas blender and a nonlinear num erical optimization algorithm by which the concentration of each compo nent in an unknown vapor can be quantified. The component vapors are i nternally blended and the mixture ratio is modified by the system so t hat the sensor array output pattern of the blended vapor can be made e qual to that of the unknown one. After several iterations, convergence is obtained and the vapor concentration of each component is determin ed from the mixture composition of the blended vapor. Although the con ventional system is passive, this system is considered as an active on e as it performs exploratory behavior prior to recognition. Here, gaso line vapor concentration is measured under the condition that one or t wo interference vapors exist together. Gasoline vapor has been adopted as an example of odors in the passenger compartment of a car, since i t sometimes smells unpleasant. The measurement is essential for design ing a car in order to keep it comfortable for passengers. The sensors used here are three semiconductor gas sensors and two electrochemical sensors, which are chosen in order to obtain high sensitivity to gasol ine. The nonlinear numerical optimization techniques used are the simp lex method and the gradient descent method and these two methods are c ompared here. It is found that the quantification error is within ten ppm for two- or three-component vapors.