Cataluminescence Sensor Detecting and Processing by S-Transform Combined with Principal Component Analysis

Abstract

In our study, a cataluminescence (CTL) online sensor system based on LaCO3OH microspheres was developed for the microscale detection and processing of volatile organic compounds. Multi-dimensional CTL test was carried out to analyse the correlation characteristics of heterogeneous analytes among the catalytic activity, sensitivity, response/recover time, stability, and photoelectric signals by changing the parameters of wavelength, flow rate, temperature, and time. Next, the generalized S matrix transform and inverse transform formula were employed to analyse the frequency distribution characteristics of the CTL patterns. As a result, the weak CTL signal was filtered based on the time-frequency characteristics of the CTL pattern signal power frequency interference and background noise. The S-transform combined with the principal component analysis algorithm was adopted to analyse the CTL patterns. Each vapour gave its unique CTL patterns, which illustrated that the PCA-S algorithm showed an application perspective for the microscale detection and extraction of CTL signals.

Keywords: Cataluminescence Sensor, LaCO3OH Microspheres, Volatile Organic Compounds, pattern recognition