Bayesian-Optimized Gaussian Process Regression for Robust Prediction of Biodiesel–Alcohol Blend Density under Thermophysical Group-Based Validation
DOI:
https://doi.org/10.9744/jti.28.2.%25pKeywords:
Machine Learning, Bayesian optimization, biodiesel–alcohol blends, Gaussian process regressionAbstract
Accurate density prediction is important for biodiesel–alcohol blend development; however, random train-test splitting may overestimate the performance in structured thermophysical datasets. This study evaluated whether Bayesian-optimized Gaussian Process Regression (BO-GPR) can be generalized to unseen thermophysical groups. A dataset of 1,856 observations was assessed using random row-wise split, group temperature, group pressure, and group alcohol composition validation. Seven regression models were compared using 10 repeated outer splits, validation-specific inner cross-validation, and Bayesian hyperparameter optimization. Model differences were examined using paired Wilcoxon signed-rank tests with Holm correction, and GPR uncertainty was assessed using 95% prediction-interval coverage. The BO-GPR achieved median R² values of 0.99997, 0.99864, 0.99999, and 0.99582 across the four validation schemes, with corresponding median RMSE values of 0.00024, 0.00143, 0.00012, and 0.00246 g/cm³, respectively. The GPR showedPR showed a significantly lower RMSE than all the benchmark models under group-based validation, whereas the difference from CatBoost was not significant under random splitting. The prediction interval coverage remained below the nominal 95% level, indicating that uncertainty calibration requires further improvement. These findings support the use of BO-GPR as a predictive surrogate within the evaluated domain, although independent experimental validation is required for broader applications.
Downloads
References
[1] H. Chen, Y. Zhang, X. Luan, M. Zhang, and G. Tao, “Experimental analysis and modeling of ethanol-biodiesel-diesel blends injection behavior,” Fluid Dynamics and Materials Processing, vol. 21, no. 7, pp. 1753–1770, 2025, doi: https://doi.org/10.32604/fdmp.2025.066494.
[2] V. Shanthan, J. Suryawanshi, R. Tarodiya, A. Loyte, Y. Devarajan, and N. Kaliappan, “Numerical analysis of spray characterization of blends of hydrous ethanol with diesel and biodiesel,” Scientific Reports, vol. 14, no. 1, 2024, doi: https://doi.org/10.1038/s41598-024-56444-0.
[3] F. Zheng and H. M. Cho, “Exploring the effects of synergistic combustion of alcohols and biodiesel on combustion performance and emissions of diesel engines: A review,” Energies, vol. 17, no. 24, p. 6274, Dec. 2024, doi: https://doi.org/10.3390/en17246274.
[4] R. Aitbelale, I. Abala, F. E. M’hamdi Alaoui, A. S. Eddine, N. M. Rujas, and F. Aguilar, “Characterization and determination of thermodynamic properties of waste cooking oil biodiesel: Experimental, correlation and modeling density over a wide temperature range up to 393.15 and pressure up to 140 MPa,” Fluid Phase Equilibria, vol. 497, pp. 87–96, 2019, doi: https://doi.org/10.1016/j.fluid.2019.06.003.
[5] C. Bouchot et al., “High pressure density for dodecane + biodiesel produced from residuals from cooking oil and beef tallow: Experiments and correlation with a proposed global (p, T, x) Tammann-Tait base model,” Journal of Chemical Thermodynamics, vol. 212, 2026, doi: https://doi.org/10.1016/j.jct.2025.107561.
[6] A. A. A. Alves et al., “Investigating the thermophysical properties of the 1-Butanol + biodiesel system: The impact of pressure on volumetric characteristics,” Fuel, vol. 371, 2024, doi: https://doi.org/10.1016/j.fuel.2024.132076.
[7] A. G. M. Ferreira, J. B. Santos, J. Baptista, S. Khalighi, R. M. M. Brito, and P. F. Cruz, “Effect of Isobutanol addition on the biodiesel density,” Journal of Chemical and Engineering Data, vol. 66, no. 12, pp. 4542–4562, 2021, doi: https://doi.org/10.1021/acs.jced.1c00588.
[8] F. Yang, X. Wang, H. Tan, S. He, and Z. Liu, “Experimental investigations on the thermophysical properties of methyl myristate in alcoholic solutions,” Fuel, vol. 215, pp. 187–195, 2018, doi: https://doi.org/10.1016/j.fuel.2017.10.122.
[9] K. Yadav, S. Almansour, L. M. Alkwai, A. Yadav, and F. Ranjbar, “Reliable models for density in waste cooking oil biodiesel-alcohol mixtures using advanced learning techniques,” Industrial Crops and Products, vol. 234, p. 121602, Oct. 2025, doi: https://doi.org/10.1016/j.indcrop.2025.121602.
[10] S. Malik, P. J. Darolia, S. K. Garg, and V. K. Sharma, “Densities and excess molar volumes of mixtures containing diesel, biodiesel and alkanols at temperatures from 288.15 to 313.15 K,” Chinese Journal of Chemical Engineering, vol. 34, pp. 198–207, Jun. 2021, doi: https://doi.org/10.1016/j.cjche.2020.09.065.
[11] P. M. Florido, P. C. F. Visioli, C. N. Pinto, and C. B. Gonçalves, “Study of FAME model systems: Database and evaluation of predicting models for biodiesel physical properties,” Renewable Energy, vol. 151, pp. 837–845, 2020, doi: https://doi.org/10.1016/j.renene.2019.11.083.
[12] M. Gülüm and A. Bilgin, “Two-dimensional surface models to predict the density of biodiesel-diesel-alcohol ternary blends,” Energy Sources, Part A: Recovery, Utilization and Environmental Effects, vol. 43, no. 5, pp. 517–587, 2021, doi: https://doi.org/10.1080/15567036.2019.1630034.
[13] J. Wang, Q. An, D. Zhou, K. Gao, and S. Bi, “Prediction methods for high pressure density of biofuels: A hybrid data-driven ML-PC-SAFT equation of state,” Fuel, vol. 418, 2026, doi: https://doi.org/10.1016/j.fuel.2026.138721.
[14] K. R. Bukkarapu and A. Krishnasamy, “Investigations on the applicability of machine learning algorithms to optimize biodiesel composition for improved engine fuel properties,” International Journal of Engine Research, vol. 25, no. 7, pp. 1299–1314, 2024, doi: https://doi.org/10.1177/14680874241227540.
[15] S. O. Giwa and R. T. Taziwa, “Predicting fuel properties of nano-doped biodiesel using machine learning algorithms,” International Journal of Ambient Energy, vol. 46, no. 1, 2025, doi: https://doi.org/10.1080/01430750.2025.2505064.
[16] M. Suvarna et al., “Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning,” Renewable Energy, vol. 189, pp. 245–258, 2022, doi: https://doi.org/10.1016/j.renene.2022.02.124.
[17] A. N. Ibrahim, D. Q. Nada, R. Nurdiansyah, and A. Andoko, “Optimization of XGBoost hyperparameters using three-dimensional learning AVOA for retail demand prediction,” J. Tek. Ind. J. Keilmuan dan Apl. Tek. Ind., vol. 28, no. 1, pp. 1–12, Mar. 2026, doi: https://doi.org/10.9744/jti.28.1.1-12.
[18] S. I. Mohammad et al., “Thermophysical properties of used frying oil biodiesels blended with alcohols: Robust machine learning frameworks for density prediction,” Industrial Crops and Products, vol. 236, 2025, doi: https://doi.org/10.1016/j.indcrop.2025.121916.
[19] A. P. Rifai, S. Briliananda, and H. Aoyama, “Tool condition monitoring with convolutional neural network for milling tools and turning inserts,” J. Tek. Ind. J. Keilmuan dan Apl. Tek. Ind., vol. 25, no. 1, pp. 1–16, Apr. 2023, doi: https://doi.org/10.9744/jti.25.1.1-16.
[20] H. Hasse, S. Schmitt, and F. Jirasek, “Artificial intelligence in thermodynamics: Hybrid modeling of thermophysical properties of fluids,” Current Opinion in Chemical Engineering, vol. 51, 2026, doi: https://doi.org/10.1016/j.coche.2026.101236.
[21] F. Jirasek and H. Hasse, “Combining machine learning with physical knowledge in thermodynamic modeling of fluid mixtures,” Annual Review of Chemical and Biomolecular Engineering, vol. 14, pp. 31–51, 2023, doi: https://doi.org/10.1146/annurev-chembioeng-092220-025342.
[22] K. R. Bukkarapu and A. Krishnasamy, “A critical review on available models to predict engine fuel properties of biodiesel,” Renewable and Sustainable Energy Reviews, vol. 155, 2022, doi: https://doi.org/10.1016/j.rser.2021.111925.
[23] M. Alruqi, P. Sharma, B. Deepanraj, and F. Shaik, “Renewable energy approach towards powering the CI engine with ternary blends of algal biodiesel-diesel-diethyl ether: Bayesian optimized Gaussian process regression for modeling-optimization,” Fuel, vol. 334, 2023, doi: https://doi.org/10.1016/j.fuel.2022.126827.
[24] B. Wang and I. Alruyemi, “Comprehensive modeling in predicting biodiesel density using Gaussian process regression approach,” BioMed Research International, vol. 2021, 2021, doi: https://doi.org/10.1155/2021/6069010.
[25] G. Yang, Q. Xu, N. Liu, W. Wu, M. Zhou, and X. Wang, “Accurate prediction of thermodynamic properties of R1234yf refrigerant using Gaussian process regression models,” International Journal of Refrigeration, vol. 186, 2026, doi: https://doi.org/10.1016/j.ijrefrig.2026.106907.
[26] M. Zhou, C. Ni, and X. Wang, “Data-driven modeling of thermophysical properties using Gaussian process regression,” presented at the AIAA SciTech Forum and Exposition, 2024, 2024. doi: https://doi.org/10.2514/6.2024-0869.
[27] J. J. Cano-Gómez, G. A. Iglesias-Silva, P. Rivas, C. O. Díaz-Ovalle, and F. De Jesús Cerino-Córdova, “Densities and Viscosities for Binary Liquid Mixtures of Biodiesel + 1-Butanol, + Isobutyl Alcohol, or + 2-Butanol from 293.15 to 333.15 K at 0.1 MPa,” Journal of Chemical & Engineering Data, vol. 62, no. 10, pp. 3391–3400, Oct. 2017, doi: https://doi.org/10.1021/acs.jced.7b00440.
[28] R. Ait Belale, F. E. M’hamdi Alaoui, Y. Chhiti, A. Sahibeddine, N. Munoz Rujas, and F. Aguilar, “Study on the thermophysical properties of waste cooking oil biodiesel fuel blends with 1-butanol,” Fuel, vol. 287, p. 119540, Mar. 2021, doi: https://doi.org/10.1016/j.fuel.2020.119540.
[29] R. Aitbelale, M. Achak, Y. Chhiti, N. Mǔnoz Rujas, F. Aguilar, and F. E. M’hamdi Alaoui, “Effects of 2-Butanol addition on waste cooking oil biodiesel density: An updated experimental measurement and thermodynamic modeling study,” Journal of Chemical & Engineering Data, vol. 66, no. 8, pp. 3123–3141, Aug. 2021, doi: https://doi.org/10.1021/acs.jced.1c00220.
[30] A. Sumayli, “Development of advanced machine learning models for optimization of methyl ester biofuel production from papaya oil: Gaussian process regression (GPR), multilayer perceptron (MLP), and K-nearest neighbor (KNN) regression models,” Arabian Journal of Chemistry, vol. 16, no. 7, p. 104833, Jul. 2023, doi: https://doi.org/10.1016/j.arabjc.2023.104833.
[31] A. Sumayli and S. M. Alshahrani, “Modeling and prediction of biodiesel production by using different artificial intelligence methods: Multi-layer perceptron (MLP), Gradient boosting (GB), and Gaussian process regression (GPR),” Arabian Journal of Chemistry, vol. 16, no. 7, p. 104801, Jul. 2023, doi: https://doi.org/10.1016/j.arabjc.2023.104801.
[32] N. Sultana, S. M. Z. Hossain, M. Abusaad, N. Alanbar, Y. Senan, and S. A. Razzak, “Prediction of biodiesel production from microalgal oil using Bayesian optimization algorithm-based machine learning approaches,” Fuel, vol. 309, p. 122184, Feb. 2022, doi: https://doi.org/10.1016/j.fuel.2021.122184.
[33] I. Yahya, M. Gassoumi, M. Attia, and R. Ennetta, “Application of AI and machine learning techniques to improve microwave-assisted biodiesel yield prediction,” Biomass and Bioenergy, vol. 202, p. 108250, Nov. 2025, doi: https://doi.org/10.1016/j.biombioe.2025.108250.
[34] K. Raj Bukkarapu and A. Krishnasamy, “Support vector regression approach to optimize the biodiesel composition for improved engine performance and lower exhaust emissions,” Fuel, vol. 348, p. 128604, Sep. 2023, doi: https://doi.org/10.1016/j.fuel.2023.128604.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Ahmad Setyo Putra, Areta Triani Putri, Rudi Nurdiansyah, Andoko
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in the Jurnal Teknik Industri: Jurnal Keilmuan dan Aplikasi Teknik Industri will be Open-Access articles distributed under the terms and conditions of the Creative Commons Attribution License (CC BY).
This work is licensed under a Creative Commons Attribution License (CC BY).
