Selective methane detection under varying moisture conditions using static and dynamic sensor signals

摘要: Abstract This paper reports the development of a methane analyser, which meets some requirements reliable domestic detector. The equipment is based on five-element array that includes four tin oxide gas sensors and humidity sensor. pattern recognition engine multilayer perceptron (MLP) neural networks trained with back-propagation algorithm. Both static dynamic signals from are fed into to identify quantify ethanol. system identifies quantifies unknown samples these contaminants, even when ambient relative varies broadly (from 25% 80%). As result, detects methane, rejects false alarms caused by ethanol complies aspects European directive for detectors.