A catalytic sensor has different sensitivity to different gases or under different temperature.According to this characterictics

through controling a catalytic sensor worked under different temperature

different output singal can be obtained when detecting the inflammable gases.A mathematic model of analyzing the diversiform inflammable gases is erected by adopting radial basis function（RBF） neural network and dynamic learning algorithm.By the experiment of mixed gases such as firedamp

carbon monoxide and hydrogen

the results show the possibility to analyze mixed inflammable gases by one catalytic sensor.