Quality assessment of engine oil: An impedance spectroscopy based approach

A.C impedance spectroscopy is a very popular and decisive method to determine certain physicochemical properties of samples. In this work we use classification algorithms to identify the used or fresh class of engine oils, based on their impedance spectra. A proposed equivalent circuit, with multiple combination of resistors, capacitors and constant phase element (CPE), along with raw impedance data constituted the input feature matrix for the classifiers. Classifiers, based on Support Vector Machine (SVM) and Artificial Neural Network applied separately, were able to identify between used samples and fresh samples with an accuracy of about 98 to 100 percent. The qualitative performance of the method was defined by false positive rate, false negative rate, sensitivity rate and specificity rate.