Intelligent tetrahydrothiophene gas detection based on electrochemical sensor array

With extensive use of natural gas energy, various gas accidents occur frequently. Reasonable odorization of natural gas is an effective way to detect gas leakage in time and avoid gas explosions. Therefore, it is necessary to realize the identification and concentration detection of natural gas odorants to ensure early warning without degrading the gas quality. Given the cross-interference effect of electrochemical gas sensors and the poor accuracy of conventional analysis methods, this paper proposed a method based on principal component analysis and the K-nearest neighbor algorithm to realize gas recognition. In addition, the backpropagation-AdaBoost model combined with an electrochemical sensor array was employed to estimate the concentration of tetrahydrothiophene, an odorant in natural gas. The natural gas from the Chenghua district of Chengdu was used as the gas source to verify the reliability of the method. The experimental results show that the gas recognition rate reaches 90.17%, and the tetrahydrothiophene concentration detection average relative error reduced to 3.37%. The results demonstrate that the method can effectively improve the prediction accuracy and reduce the impact of cross-interference on the detection of tetrahydrothiophene. The method can provide technical support for natural gas odorant detection with great engineering significance for solving gas safety problems.