An Improved Gas Classification Technique Using New Features and Support Vector Machines
In this paper, we propose a gas classification technique based on extracting new features and support vector machines (SVM) in a chemical plant. First, various gases are collected using semiconductor gas seniors, and then we calculate the composition ratio of these gasses, which are defined as features. These extracted features are highly discriminative and quantify the presence of gas. Moreover, these features are used as the SVM input for classifying gas types. In addition, we apply a grid search technique in SVM for tuning hyper-parameters such as misclassification rate, C, and kernel bandwidth, σ, to improve the classification performance. To verify the proposed technique, we collect various gases composition using a cost-effective self-designed test rig. The experimental results indicate that the proposed method is highly capable of classifying various hazardous gases with good accuracy.
KeywordsFeature extraction Gas sensor array Gas classification Support vector machine Chemical plants
This research was supported by the Ministry of Science and Technology, Ministry of Information and Communication, and the Korea Information and Telecommunication Industry Promotion Agency (No. S0702-18-1045).
- 1.Lee, I.S., Shim, C.H.: Gas classification and fault diagnosis of the semiconductor type gas sensor system. J. Korea Inf. Sci. Soc. 7, 48–57 (2009)Google Scholar
- 2.Lee, J.H., Cho, J.H., Jeon, G.J.: Concentration estimation of gas mixtures using a tin oxide gas sensor and fuzzy ART. J. Korea Electron. Eng. 43, 21–29 (2006)Google Scholar
- 3.Lee, D.S., Jung, H.Y., Ban, S.W., Lee, M.H., Huh, J.S., Lee, D.D.: Fabrication of semiconductor gas sensor array and explosive gas sensing characteristics. J. Korea Electron. Eng. 37, 9–17 (2000)Google Scholar
- 4.Lee, K.C., Rye, K.R., Hur, C.W.: Fabrication and yield improvement of oxide semiconductor thin film gas sensor array. J. Korea Inf. Commun. Soc. 6, 315–322 (2002)Google Scholar
- 5.Heo, J.Y., Yang, J.Y.: SVM based stock price forecasting using financial statements. J. Korea Inf. Sci. Soc. 21, 167–172 (2015)Google Scholar
- 6.Park, J.H., Hwang, C.S., Bae, K.B.: Analysis of target classification performances of active sonar returns depending on parameter values of SVM kernel functions. J. Korea Inf. Commun. Soc. 17, 1083–1088 (2013)Google Scholar
- 7.Min, J.H., Lee, Y.C.: Support vector bankruptcy prediction model with optimal choice of RBF kernel parameter values sing grid search. J. Korea Manag. Sci. 30, 55–74 (2005)Google Scholar
- 8.Kim, Y.H., Kim, J.Y., Jeong, I.K., Kim, Y.H., Kim, J.-M.: A method of detecting boiler tube leakage using a genetic algorithm and support vector machines. Korea Comput. Inf. Soc. 26, 55–56 (2018)Google Scholar
- 9.Kim, J.Y., Kim, J.-M., Choi, B.K.: Bearing fault diagnosis using adaptive self-tuning support vector machine. Korea Comput. Inf. Soc. 24, 19–20 (2016)Google Scholar
- 12.Islam, M.M.M., Islam, M.R., Kim, J.-M.: A hybrid feature selection scheme based on local compactness and global separability for improving roller bearing diagnostic performance. In: Artificial Life and Computational Intelligence, pp. 180–192. Springer (2017)Google Scholar