Predicting Marine Fuels with Unusual Wax Appearance Temperatures Using One-Class Support Vector Machines

Abstract

Accurate and robust detection of wax appearance (a medium- to high-molecular-weight component of crude oil) is crucial for the efficient operation of hydrocarbon transportation. The wax appearance temperature (WAT) is the lowest temperature at which the wax begins to form. When crude oil cools to its WAT, wax crystals precipitate, forming deposits on pipelines as the solubility limit is reached. Therefore, WAT is a crucial quality assurance parameter, especially when dealing with modern fuel oil blends. In this study, we use machine learning via MATLAB’s Bioinformatics Toolbox to predict the WAT of marine fuel samples by correlating near-infrared spectral data with laboratory-measured values. The dataset provided by Intertek PLC—a total quality assurance provider of inspection, testing, and certification services—includes industrial data that is imbalanced, with a higher proportion of high-WAT samples compared to low-WAT samples. The objective is to predict marine fuel oil blends with unusually high WAT values (>35°C) without relying on time-consuming and irregular laboratory-based measurements. The results demonstrate that the developed model, based on the one-class support vector machine (OCSVM) algorithm, achieved a Recall of 96, accurately predicting 96% of fuel samples with WAT >35°C. For standard binary classification, the Recall was 85.7. The trained OCSVM model is expected to facilitate rapid and well-informed decision-making for logistics and storage when choosing fuel oils.

Keywords

Marine fuel / One-class support vector machines / Wax appearance temperature / Wax / Machine learning

References

[1] Ahmed T. 6.2.11.1 Phase Behavior of Waxes. Equations of State and PVT Analysis-Applications for Improved Reservoir Modeling, 2007, Houston. Gulf Publishing Company498

[2] Al Shakhs MH, Libby C, Chau KJ, Molla S, Sieben VJ. Wax appearance temperature in crude oils measured by surface plasmon resonance. Petroleum Science and Technology, 2023, 43(4): 408-425.

[3] Bangert P. 3.3.3 Support Vector Machines. Machine Learning and Data Science in the Oil and Gas Industry-Best Practices, Tools, and Case Studies, 2021, Alpharetta. Elsevier48-49

[4] Benamara C, Gharbi K, Nait Amar M, Hamada B. Prediction of wax appearance temperature using artificial intelligent techniques. Arabian Journal for Science and Engineering, 2020, 45(2): 1319-1330.

[5] Bekkar M, Djemaa HK, Alitouche TA. Evaluation measures for models assessment over imbalanced data sets. Journal of Information Engineering and Applications, 2013, 3(10): 28-31

[6] Bian XQ, Huang JH, Wang Y, Liu YB, Kaushika Kasthuriarachchi DT, Huang LJ. Prediction of wax disappearance temperature by intelligent models. Energy & Fuels, 2019, 33(4): 2934-2949.

[7] Blanco M, Villarroya I. NIR spectroscopy: a rapid-response analytical tool. TrAC Trends in Analytical Chemistry, 2002, 21(4): 240-250.

[8] Broadhurst DI, Kell DB. Statistical strategies for avoiding false discoveries in metabolomics and related experiments. Metabolomics, 2006, 2(4): 171-196.

[9] Chen Y, Zhou XS, Huang TS. One-class SVM for learning in image retrieval. Proceedings 2001 international conference on image processing (Cat. No. 01CH37205), 2001, Piscataway. IEEE34-37.

[10] Ciaburro G, Joshi P (2019) 1.6 Normalization. Python Machine Learning Cookbook (2nd Edition) 19

[11] Cohen G, Hilario M, Sax H, Hugonnet S, Pellegrini C, Geissbuhler A. An application of one-class support vector machines to nosocomial infection detection. Studies in Health Technology and Informatics, 2004, 107(Pt1): 716-720

[12] Coutinho JA, Daridon JL. The limitations of the cloud point measurement techniques and the influence of the oil composition on its detection. Petroleum Science and Technology, 2005, 23(9-10): 1113-1128.

[13] Cortes C, Vapnik V. Support-vector networks. Machine Learning, 1995, 20(3): 273-297.

[14] de Oliveira MCK, Teixeira A, Vieira LC, de Carvalho RM, de Carvalho ABM, do Couto BC. Flow assurance study for waxy crude oils. Energy & Fuels, 2012, 26(5): 2688-2695.

[15] Ding Z. Diversified ensemble classifiers for highly imbalanced data learning and its application in bioinformatics, 2011, Atlanta. Georgia State University1-18

[16] ExxonMobil. ISO 8217 Marine Fuel Characteristics Definitions, 2022

[17] Fanali S, Haddad PR, Poole CF, Riekkola M-L. 21.3.3 Normalization. Liquid Chromatography-Fundamentals and Instrumentation, Volume 1, 20172nd Edition518-519

[18] Fletcher T. Support vector machines explained. Tutorial Paper 1-19, 2009

[19] Guerbai Y, Chibani Y, Hadjadji B. The effective use of the One-Class SVM classifier for reduced training samples and its application to handwritten signature verification. 2014 international conference on multimedia computing and systems (ICMCS), 2014, IEEE. Piscataway362-366.

[20] Huang J, Romero-Torres S, Moshgbar M. Practical considerations in data pre-treatment for NIR and raman spectroscopy. American Pharmaceutical Review, 2010

[21] Japper-Jaafar A, Bhaskoro PT, Mior ZS. A new perspective on the measurements of wax appearance temperature: Comparison between DSC, thermomicroscopy and rheometry and the cooling rate effects. Journal of Petroleum Science and Engineering, 2016, 147: 672-681.

[22] Kok MV, Létoffé JM, Claudy P, Martin D, Garcin M, Volle JL. Comparison of wax appearance temperatures of crude oils by differential scanning calorimetry, thermomicroscopy and viscometry. Fuel, 1996, 75(7): 787-790.

[23] Kök MV, Varfolomeev MA, Nurgaliev DK. Wax appearance temperature (WAT) determinations of different origin crude oils by differential scanning calorimetry. Journal of Petroleum Science and Engineering, 2018, 168: 542-545.

[24] Kuligowski J, Pérez-Guaita D, Quintás G. Application of discriminant analysis and cross-validation on proteomics data. Statistical Analysis in Proteomics, 2016, 1362: 175-184.

[25] Lammoglia T, de Souza Filho CR. Spectroscopic characterization of oils yielded from Brazilian offshore basins: Potential applications of remote sensing. Remote Sensing of Environment, 2011, 115(10): 2525-2535.

[26] Lang R, Lu R, Zhao C, Qin H, Liu G. Graph-based semisupervised one class support vector machine for detecting abnormal lung sounds. Applied Mathematics and Computation, 2020, 364: 124487

[27] Li K, Wu M, Gu X, Yuen KF, Xiao Y. Determinants of ship operators' options for compliance with IMO 2020. Transportation Research Part D: Transport and Environment, 2020, 86: 102459.

[28] Li H, Chen H, Li Y, Chen Q, Fan X, Li S, Ma M. Prediction of the optical properties in photonic crystal fiber using support vector machine based on radial basis functions. Optik, 2023, 275: 170603.

[29] Liu Y. 3.1.2 Scenario 2-Determining the Optimal Hyperplane. Python Machine Learning by Example, 20203rd EditionBirmingham. Packt Publishing78-79

[30] Liu Y. 3.1.3 Scenario 3-Handling Outliers. Python Machine Learning by Example, 20203rd EditionBirmingham. Packt Publishing80-81

[31] Long J, Buyukozturk O. Automated structural damage detection using one-class machine learning. Dynamics of Civil Structures, 2014, Luembourg. Springer117-128

[32] MathWorks. Cvpartition, 2022

[33] MathWorks. Fitcsvm, 2022

[34] MathWorks. OneClassSVM, 2023

[35] MathWorks. ResubPredict, 2024

[36] Martí L, Sanchez-Pi N, Molina JM, Garcia ACB. Anomaly detection based on sensor data in petroleum industry applications. Sensors, 2015, 15(2): 2774-2797.

[37] Mansourpoor M, Azin R, Osfouri S, Izadpanah AA. Study of wax disappearance temperature using multi-solid thermodynamic model. Journal of Petroleum Exploration and Production Technology, 2019, 9(1): 437-448.

[38] Meyer D, Leisch F, Hornik K. The support vector machine under test. Neurocomputing, 2003, 55(1–2): 169-186.

[39] Nguyen TBT, Liao TL, Vu TA. Anomaly detection using oneclass SVM for logs of juniper router devices, 2023

[40] Nie J, Dong Y, Zuo R. Construction land information extraction and expansion analysis of Xiaogan City using one-class support vector machine. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 3519-3532.

[41] Sandak J, Sandak A, Meder R. Assessing trees, wood and derived products with near infrared spectroscopy: hints and tips. Journal of Near Infrared Spectroscopy, 2016, 24(6): 485-505.

[42] Schiilkop P, Burgest C, Vapnik V. Extracting support data for a given task. Proceedings of the 1st International Conference on Knowledge Discovery & Data Mining, 1995, Reston. AIAA252-257

[43] Schölkopf B, Platt JC, Shawe-Taylor J, Smola AJ, Williamson RC. Estimating the support of a high-dimensional distribution. Neural Computation, 2001, 13(7): 1443-1471.

[44] Seo K-K. An application of one-class support vector machines in content-based image retrieval. Expert Systems with Applications, 2007, 33(2): 491-498.

[45] Speight JG. .4.6 Wax Analysis and Wax Appearance Temperature. Introduction to Enhanced Recovery Methods for Heavy Oil and Tar Sands, 20162nd Edition222-224

[46] Sun DW. 4.3 Evaluation of Classification Performances. Infrared Spectroscopy for Food Quality Analysis and Control, 200992-93

[47] Taheri-Shakib J, Shekarifard A, Kazemzadeh E, Naderi H, Rajabi-Kochi M. Characterization of the wax precipitation in Iranian crude oil based on wax appearance temperature (WAT): The influence of ultrasonic waves. Journal of Molecular Structure, 2020, 1202: 127239.

[48] Thomas B. An automatic test method for wax appearance temperatures of VLSFOs, 2019

[49] Thijssen P, Hadjiloucas S. 12.3.2 Advances in Support Vector Machine Classifiers. State Estimation in Chemometrics-The Kalman Filter and Beyond, 20202nd Edition237

[50] Uba E, Ikeji K, Onyekonwu M. Measurement of wax appearance temperature of an offshore live crude oil using laboratory light transmission method. Nigeria Annual International Conference and Exhibition, 2004

[51] VPO. VPS launches automatic test method for wax appearance temperature of VLSFOs, 2019

[52] Westerhuis JA, Hoefsloot HC, Smit S, Vis DJ, Smilde AK, van Velzen EJ, van Duijnhoven JP, van Dorsten FA. Assessment of PLSDA cross validation. Metabolomics, 2008, 4(1): 81-89.

[53] Xiao Y, Wang H, Xu W. Parameter selection of Gaussian Kernel for one-class SVM. IEEE Transactions on Cybernetics, 2015, 45(5): 941-953.

[54] Yau K, Chow KP, Yiu SM. Detecting attacks on a water treatment system using oneclass support vector machines. IFIP Advances in Information and Communication Technology. 16th Annual IFIP WG 11.9 International Conference on Digital Forensics, 2020, 2020, New Delhi. Springer Science and Business Media Deutschland GmbH95-108

[55] Zhang S, Sun X, Liu C, Zhang H, Miao X, Zhao K. Characterization of wax appearance temperature of model oils using laser-induced voltage. Physics of Fluids, 2022, 34(6): 067123.

[56] Zhang T. An introduction to support vector machines and other kernel-based learning methods. AI, 2001, Magazine22(2): 103-103

[57] Zhao K, Li C, Xia X, Fang K, Yao B, Yang F. Optical techniques for determining wax appearance temperature of waxy crude oil. 2021 International Conference on Optical Instruments and Technology: Optoelectronic Measurement Technology and Systems, 2022, Bellingham. SPIE492-500

RIGHTS & PERMISSIONS

Harbin Engineering University and Springer-Verlag GmbH Germany, part of Springer Nature