The Research Of Mixed-gas Test System On Gas Sensor Array

This paper is about gas identification system based on gas sensor array, which named electronic nose (EN) also. An EN is an instrument, which comprises an array of cross-sensibility gas sensors with partial specificity pattern-recognition system, capable of recognizing simple or complex odors. EN finds promising applications in food industry, aviation and spaceflight, environment inspection; thus this research area is of great importance and become an international research focus.In this dissertation the principle of the EN was analyzed by comparing with those of olfaction organism. According the requirement of this experiment, we designed the experiment location namely test-cavity and confirm the way to confect mixed-gas, which completed in lab. The array of sensors is composed by series TGS and MQ, and design the part of hardware for signal acquisition and pre-process. Designed mixed-gas with the static principle of static mixed-gas, achieved require of mixed-gas detect in the experiment. The error of this method is about 2 ppm more or less, accord with require of the experiment.The whole system of EN had applied for classify and identify of H2S, CO and CH4, which mixed with different concentration. Can detect the poison gas and explosive gas with low-concentration, the periodicity only need about 20 minutes, and the dimension of detect is 10-1000ppm.The data acquisition of EN and BP network's program base on the visual C++ language, keep the program's readability and maneuverability. The result of operation is perfect. Improved BP network partly, finished the training and testing for BP network, and analyze and treatment for experiment data using it. With the result we can see that the error is keep up in a special scope.