Abstract—Ethanol is the major consumable biofuel which is highly inflammable. So the detection of ethanol at its different concentration is certainly required. In this paper a cost effective thick-film gas sensor based on ethanol detection has been described. The sensor exhibits excellent ethanol sensing characteristics at the temperatures between 175 °C to 300 °C and the characteristics of the sensor for different concentration of the ethanol gas has been successfully studied by performing a nonlinear form of principal component analysis (PCA) to cluster the different concentration of ethanol. The present method is to classify the 230 samples of each concentration levels of gas using kernel principal component analysis (KPCA). Ethanol gas is a volatile organic compound (VOC) which is little difficult to distinguish the different concentration level in their nominal method.In this regard, classifiers having the machine learning ability can be of great benefit by automatically including the newly presented patterns in the training dataset without affecting class integrity of the previously trained system. In the presented paper, the kernel principal component analysis (KPCA) is used in the clustering algorithm for ethanol concentration discrimination with the help of thick film gas sensor.
Index Terms—Cluster computing, kernel, principal component analysis, feature space, nonlinear mapping, pattern recognition.
Laxmi Shaw, Sangeeta Bagha, Ambarish G. Mohapatra and Narayan Nayak are with the Silicon Institute of Technology, Bhubaneswar, Odisha, India (e-mail: laxmishaw1983@ gmail.com; sangeetabagha@ gmail.com; firstname.lastname@example.org; email@example.com).
Cite: Laxmi Shaw, Sangeeta Bagha, Ambarish G. Mohapatra, and Narayan Nayak, "Kernel Approach on Detection of Ethanol Concentration Using ZnO Gas Sensor," International Journal of Machine Learning and Computing vol. 2, no. 1, pp. 71-75, 2012.