Abstract
This chapter deals with the comparison of performances of different intelligent techniques for detecting proportion of different component gases present in manhole gas mixture. Toxic gases found in manhole gas mixture are Hydrogen Sulfide (H2S), Ammonia (NH3), Methane (CH4), Carbon Dioxide (CO2), Nitrogen Oxide (NOx), Carbon Monoxide (CO), etcetera. Detection of these toxic gases is essential since these gases influence human health even due to very short exposure. This study is centered on design issues of an intelligent sensory system for detecting proportion of different components in manhole gas mixture and comparison of different intelligent techniques applied for this. The investigation encompasses linear regression based statistical technique, backpropagation algorithm, neuro genetic techniques (using genetic algorithm to train neural network), and neuro swarm techniques (using particle swarm optimization algorithm to train neural network).
TopBackground
We referred few papers reported in the literature on gas recognition and gas detection system. The referred papers are illustrating applications related either to the development of gas recognition system or to the development of an E-Nose (Electronic Nose) mechanism. We discuss these papers briefly to present a picture that how much and what development and research has been taken place in the past decades. We also provide a list of these papers in brief.
Key Terms in this Chapter
Backpropagation: In Backpropagation algorithm error computed at output layer is back propagate to earlier layers to adjust synaptic weights such that sum squared error can be minimized to train neural network.
Manhole Gas: In gases found in sewer pipelines are known as manhole gas (Typically in India).
Cross Sensitivity: The cross-sensitivity is the sensitivity of a gas sensor to any non target gas.
Neuro Genetic: Using Genetic Algorithm to searches out optimized value of synaptic weights for a given neural network.
Sensor Array: Array of semiconductor based gas sensor.
Neuro Swarm Optimization: Using particle Swarm Optimization Algorithm to searches out optimized value of synaptic weights for a given neural network.
Safety Limit: Maximum allowed concentration level of any gas in any human working environment.