Artificial Neural Networks Applied to Gas Mixture Analysis

摘要

An array composed of sixteen gas sensors was constructed to anaylze gas mixtures quantitavely. The data of responses from the sensor array to ethane, propane and propylene were treated by three-layer ANN with BP algorithms and PLS. The analgtical results indicated tha the concentration predicted with ANN is better than that with PLS. The average prediction errors for ethane, propane and propylene were 5. 11%, 8.28%, 2.64%, respectively.