Automatic classification and permittivity estimation of glycerin solutions using a dielectric resonator sensor and machine learning techniques

Abstract

This paper presents the application of a dielectric resonator sensor to characterize glycerin solutions. Air and nine different concentrations were measured within a relative permittivity range from 1 to 78.3. Principal Component Analysis (PCA) and Support Vector Machine (SVM) were used to perform automatic classification with an 100 accuracy and the regression of both concentration and permittivity with a RMSE of 0.34 and 0.287 respectively.

This paper presents the application of a dielectric resonator sensor to characterize glycerin solutions. Air and nine different concentrations were measured within a relative permittivity range from 1 to 78.3. Principal Component Analysis (PCA) and Support Vector Machine (SVM) were used to perform automatic classification with an 100 accuracy and the regression of both concentration and permittivity with a RMSE of 0.34 and 0.287 respectively.