A Comprehensive Evaluation Model for Optimizing the Sensor Array of Electronic Nose

Abstract

As the acquisition unit of gas information, the sensor array directly determines the overall performance of the electronic nose system (E-nose). This paper proposed a new method for optimizing the sensor array. Firstly, four evaluation indicators (sensitivity, selectivity, correlation, and repeatability) were selected to evaluate the sensor array. Subsequently, different evaluation indicators were assigned different weight values according to their contributions to the overall performance of the E-nose. Finally, a comprehensive evaluation model was established based on the EWM-TOPSIS algorithm to optimize the sensor array. In order to verify the effectiveness of the as-proposed model, it was applied to the optimization of the E-nose sensor array composed of 10 gas sensors, and the influence of the sensor array optimization on the gas recognition ability of the E-nose was investigated. The experimental results showed that the optimized sensor array can identify the CO-CH4 gas mixtures with an accuracy of 96.5%, which a significant improvement compared with the accuracy of 78.3% before the sensor array optimization.