Improvements in signal evaluation techniques for semiconductor gas sensors

The applied chemical sensor research focuses on sensor arrays and signal evaluation methods, to improve reliability, selectivity and other features of the single sensor. State of the art is the use of self adapting systems like Artificial Neural Networks (ANN), mostly used for classification purposes. Seldomly system for the prediction of gas concentrations were investigated, because one of the main problems for those signal processing systems is the enourmous amount of training data and the time dependency of the sensor signal. This work uses an array of semiconductor sensors (R.G.T.0 method and commercial sensors) and a modified ANN method for signal processing. After a drift correction based on an empirical modell, a feed forward network predicts gas concentrations more precisely. A new method (DTPID method = dynamic test point distribution) is invented, which achieves a significant reduction of the calibration time together with a high accuracy in calculating the gas concentration