A Proposed Two-Level Classification Approach for Forensic Detection of Diesel Adulteration Using NMR Spectroscopy and Machine Learning

Aug 5, 2026

Upload by Eliana

Diesel fuel adulteration remains a major concern in many developing countries, including Ghana, despite regulatory schemes. Limitations of the current solvent tracer analysis motivate alternative or supplementary detection tools. A two-level classification strategy combines 1H NMR spectroscopy data with machine learning to classify diesel as neat vs. adulterated and then identify adulterant type. Models use training adulterant levels of 20–40% w/w typical for Ghana, achieving 2.5% errors at level 1 and low errors for specific adulterants at level 2, with reduced reliability below 20% w/w.

LanguageEnglish

CategoryResearch & Report

FormatDOCX

Pages25

Reading Time1 h 3 min

AccessFree to download

Abstract

Keywords

Introduction

Background: crude oil refining and diesel use

Fuel adulteration as a criminal offence and regional trends

Adverse impacts of adulterated diesel

Factors influencing adulterant choice

Common adulterants in Ghana

Existing detection and monitoring approaches

A Proposed Two-Level Classification Approach for Forensic Detection of Diesel Adulteration Using NMR Spectroscopy and Machine Learning Dadson, J. K.1*, Asiedu N. Y.2, Iggo. J. A.3, Konstantin, L.3, Ackora-Pra, J.4, Baidoo, M.F.2 and Akoto, O5. 1Department of Biochemistry and Biotechnology, Kwame Nkrumah University of Science and Technology, Ghana 2Department of Chemical Engineering, Kwame Nkrumah University of Science and Technology, Ghana 3Department of Chemistry, University of Liverpool, UK 4Department of Mathematics, Kwame Nkrumah University of Science and Technology, Ghana 5Department of Chemistry, Kwame Nkrumah University of Science and Technology, Ghana *Corresponding Author: jaydadson1990@gmail.com ABSTRACT Adulteration of diesel fuel poses a major concern in most developing countries including Ghana despite the many regulatory schemes adopted. The solvent tracer analysis approach currently used in Ghana has over the years suffered several limitations which affects the overall implementation of the scheme. There is therefore a need for alternative or supplementary tools to help detect adulteration of automotive fuel. Herein we describe a two-level classification method that combines NMR spectroscopy and machine learning algorithms to detect adulteration in diesel fuel. The training sets used in training the machine learning algorithms contained 20-40% w/w adulterant, a level typically found in Ghana. At the first level, a classification model is built to classify diesel samples as neat or adulterated. Adulterated samples are passed on to the second stage where a second classification model identifies the type of adulterant (kerosene, naphtha or premix) present. Samples were analyzed by 1H NMR spectroscopy and the data obtained were used to build and validate Support Vector Machine (SVM) classification models at both levels. At level 1, the SVM model classified all 200 samples with only 2.5% classification errors after validation. The level 2 classification model developed had no classification errors for kerosene and premix in diesel. However, 2.5% classification error was recorded for samples adulterated with naphtha. Despite the great performance of the proposed schemes, it showed significantly erratic predictions with adulterant levels below 20% w/w as the training sets for both models contained adulterants above 20% w/w. The proposed method, nevertheless, proved to be a potential tool that can serve as an alternative to the marking system in Ghana for the fast detection of adulterants in diesel. Keywords: fuel adulteration; Proton NMR spectroscopy; machine learning; Chemometrics Introduction Crude oil remains an essential source of fossil fuel in the world [1]. The refinement of crude oil leads to the production of gasoline, diesel, naphtha, jet fuels, liquefied petroleum gas, kerosene, waxes etc. Out of the distillates, approximately 80% are used as fuel for various modes of transportation. The automobile sector, in particular, has become a major consumer of fuel over the years with 66% of crude oil being refined into gasoline and diesel. The consumption of fuel in Africa is expected to double in the next 30 years from 4 million to approximately 8 million barrels per day [2]. Due to the competition in the downstream petroleum industry, there has always been tendencies for oil marketing companies to attempt to maximize profits illegally through fuel adulteration. Fuel adulteration typically involves mixing large amounts of lower grade crude oil products into commercial fuel [3]. and is a criminal offence in most countries. Since 2017, many developing countries including Nigeria, Morocco, India and Brazil have experienced an all-time rise in fuel adulteration despite the many regulations implemented [4,5]. For example, fuel tests conducted in 2015 in Morocco detected an increase adulteration, studies conducted in Brazil in 2012 revealed that approximately 40% of fuel randomly sampled from filling stations showed vari

FAQs

What detection problem does the document address?

The document addresses diesel fuel adulteration, which is a major concern in developing countries and is linked to legal, economic, health, and environmental impacts.

How does the proposed two-level classification method work?

Level 1 classifies diesel samples as neat or adulterated using NMR-derived data and an SVM model. Level 2 then determines the adulterant type (kerosene, naphtha, or premix) for samples identified as adulterated.

What performance results are reported for the models?

At level 1, the SVM model classified 200 samples with only 2.5% classification errors after validation. At level 2, kerosene and premix showed no classification errors, while naphtha had a 2.5% classification error.

Aug 5, 2026

Upload by Eliana