METHANE AND BUTANE CONCENTRATIONS IN A MIXTURE WITH AIR DETERMINED BYMICROCALORIMETRIC SENSORS AND NEURAL NETWORKS

A pair of microcalorimetric sensors (pellistors), one with a hexamethy ldisiloxane (HMDS) pretreatment, is applied for the determination of t he composition of butane/methane mixtures in air. Both gases show a st rong interaction on the sensor surface, excluding any linear superposi tion algorithms. An evaluation of the butane and methane concentration s from the pellistor signals by two neural networks, one for each comb ustible, is described here. Maximum errors below 4% LEL (lower explosi on limit) for butane and 10% LEL for methane are attained with the neu ral networks being trained with 21 measured points and applied to 64 p oints.