Gradient gas sensor microarrays for on-line process control - a new dynamic classification model for fast and reliable air quality assessment

A new dynamic gas classification model was developed to achieve a reliable on-line discrimination at very fast response times. The aim was to be able to follow rapid changes in gas compositions using an electronic nose in con sumer applications. The electronic nose is based on a microarray especially designed for production at very low costs. This is essential for applicati on in mass products. Common classification methods used fur signal evaluati on of electronic noses such as Linear Discriminant Analysis (LDA), Neural N etworks (NN) or Soft Independent Modelling of Class Analogy (SIMCA) fail to detect non-stationary gas mixtures. The new model, however, combines class ification of steady states with transient evaluation via time series analys is. Rapid signal transients are detected by appropriate digital filters, st eady state signals are classified by the above mentioned standard methods. The simplicity of the algorithm model allows implementation in low-cost ele ctronic units, containing micro controllers with very limited memory capaci ty. To give an example, the automatic control of thr ventilation flap of au tomobiles was investigated. Intermediate streams of bad air could be detect ed within 1-2 s. The error of pollutants detection was reduced from 25%, ap plying static classification only, to 10% for the new dynamic model. (C) 20 00 Elsevier Science S.A. All rights reserved.