A Novel Machine Learning for Ethanol and Methanol Classification with Capacitive Soil Moisture (CSM) Sensors

Aug 4, 2026

Upload by Rowan

Although Gas Chromatography (GC) provides high accuracy for alcohol detection, its high cost motivates the need for an affordable alternative method. Ethanol and methanol differ in evaporation behavior and dielectric constants, enabling classification from dielectric changes during evaporation. This study develops a novel machine learning workflow using Capacitive Soil Moisture (CSM) sensors by measuring time-varying evaporative dielectric properties from evaporated samples on CSM plates. The dataset is processed in Python, evaluated with cross-validation, and trained using multiple classifiers. Results demonstrate strong discrimination between pure ethanol and methanol, with reported accuracies up to 96.67%, supporting a simpler, lower-cost initial step toward alcohol content estimation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Background: Physical properties and dielectric behavior

Conventional identification using Gas Chromatography (GC)

ComTech: Computer, Mathematics and Engineering Applications, 15(2), December 2024, 109−118 DOI: 10. 21512/comtech.v15i2 .12051 P-ISSN: 2087-1244 E-ISSN: 2476-907X A Novel Machine Learning for Ethanol and Methanol Classification with Capacitive Soil Moisture (CSM) Sensors Devina Intan Sari1, Suryasatriya Trihandaru2, and Hanna Arini Parhusip3* 1-3Magister Sains Data, Fakultas Sains & Matematika, Universitas Kristen Satya Wacana Salatiga, Indonesia 50711 [1](1632023001@student.uksw.edu)[632023001@student.uksw.edu](1632023001@student.uksw.edu); [2](2suryasatriya@uksw.edu)[suryasatriya@uksw.edu](2suryasatriya@uksw.edu); [3](3hanna.parhusip@uksw.edu)[hanna.parhusip@uksw.edu](3hanna.parhusip@uksw.edu) [Received](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[: 19](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[th](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[ August 2024/](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[ Revised](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[: 5](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[th](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[ November 2024/](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[ Accepted](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[: 6](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[th](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024)[ November 2024](Received: 19th August 2024/ Revised: 5th November 2024/ Accepted: 6th November 2024) How to Cite: Sari, D. I., Trihandaru, S., & Parhusip, H. A. (2024). A Novel Machine Learning for Ethanol and Methanol Classification with Capacitive Soil Moisture (CSM) Sensors. ComTech: Computer, Mathematics and Engineering Applications, 15(2), 109−118. [https://doi.org/10.21512/comtech.v15i2.12051](https://doi.org/10.21512/comtech.v15i2.12051) Abstract-Although Gas Chromatography (GC) is highly accurate, it is costly, highlighting the need for a more affordable method for alcohol detection. Ethanol and methanol have different evaporation ratesand dielectric constants, suggesting the potential for classification as an alternative initial step to GC based on differences in dielectric due to evaporation using Capacitive Soil Moisture (CSM) sensors, although it has not been previously attempted. The research aimed to present a novel machine learning for ethanol and methanol classification with CSM sensors. The method involved placing evaporated samples on CSM plates and measuring the change in evaporative dielectric properties over time. The data were then processed using Python, preprocessing data, splitting data, and training various classifiers with key differentiators based on standard deviation, mean, difference, and cumulative summary. Then, model accuracy was evaluated. The research results show that the approach can distinguish between pure ethanol and methanol based on the dielectric differences in each substance's evaporation rate using machine learning training methods with classifiers such as Random Forest, Extra Trees, Gaussian Naive Bayes, AdaBoost, and Logistic Regression with seven folds in cross-validation, L2 regularization, and Newton-Cholesky solver, with accuracies of 96.67%, 96.67%, 96.67%, 93.33%, and 93.33%, respectively. Although the research is limited to the classification of two types of alcohol, the novel approach can classify methanol and ethanol, leading toa potential initial step in determining alcohol content in the future. It can be an alternative to GC with a simpler and more affordable setup using CSM sensors. Keywords: machine learning, ethanol classification, methanol

FAQs

Why is a machine learning approach with CSM sensors proposed instead of Gas Chromatography (GC)?

GC is highly accurate but costly, so the study seeks a simpler and more affordable alternative based on measurable dielectric differences during evaporation.

How do CSM sensors support ethanol and methanol classification in the proposed method?

Evaporated samples are placed on CSM plates and dielectric properties are measured over time; the resulting dielectric-change patterns are used for classification.

Which classifiers and performance levels are reported for distinguishing ethanol from methanol?

The study trains several models including Random Forest, Extra Trees, Gaussian Naive Bayes, AdaBoost, and Logistic Regression, reporting accuracies up to 96.67% using cross-validation.

Aug 4, 2026

Upload by Rowan