Identification and Quantification of Common Adulterants in Extra Virgin Olive Oil Using Microwave Dielectric Spectroscopy Aided by Feedforward Neural Networks
Júlio C. P. Alarcon,Mateus I. O. Souza,Vinícius M. Pepino,Ben-Hur V. Borges
DOI: https://doi.org/10.1109/jsen.2024.3448221
IF: 4.3
2024-10-04
IEEE Sensors Journal
Abstract:In this work, we combine microwave dielectric spectroscopy and machine learning techniques to assess the authenticity of extra virgin olive oil (EVOO). We investigate four common adulterants of EVOO, namely, soybean oil, corn oil, sunflower oil, and canola oil. We use a low-cost spiral-shaped microwave resonator-based sensor operating at 546.8 MHz to detect changes in the complex permittivity of oil samples. A vector network analyzer (VNA) is used to extract complex scattering parameters and that serve as inputs for two different feedforward artificial neural network (ANN) classifiers. The first model, using only the real and imaginary parts of , achieves an overall accuracy of 91.4% in predicting the applied adulterant in test samples. In contrast, the second model, incorporating the real and imaginary parts of both and , attains a 99.2% accuracy for test samples. In addition, we investigate the relationship between the measured (in dB) and the adulteration level, expressed as the percentile value of the volume of adulterant per volume of the sample (mL/mL). For each adulterant, a calibration equation is developed using partial least-squares regression (PLSR) to predict adulteration levels up to 50%. The maximum root mean square error (RMSE) is 2.1% for canola oil adulteration prediction. PLSR yields RMSE values of 0.9% for soybean oil, 1.1% for corn oil, and 1.0% for sunflower oil adulteration. This methodology offers both qualitative and quantitative analyses of EVOO, capable of identifying adulterations as low as 5% with a simple, portable, and practical system.
engineering, electrical & electronic,instruments & instrumentation,physics, applied