El-Adawy, Mohammed;
Zayed, Mohamad E;
Shboul, Bashar;
Ashraf, Waqar Muhammad;
Nemitallah, Medhat A;
(2024)
Performance improvement of compression ignition engine fueled by second generation biodiesel fuel blends enriched with ZnO nanoparticles: Experimental study and Gaussian process regression AI modeling.
Process Safety and Environmental Protection
, 190
(Part-A)
pp. 1372-1385.
10.1016/j.psep.2024.07.069.
![[thumbnail of Ashraf_Final Manuscript _PSEP.pdf]](https://discovery.ucl.ac.uk/style/images/fileicons/text.png) |
Text
Ashraf_Final Manuscript _PSEP.pdf Access restricted to UCL open access staff until 7 August 2025. Download (1MB) |
Abstract
Artificial intelligence (AI) methods are currently being utilized to efficiently predict diesel engine performance, offering reduced complexity and shorter computation times. This study introduces a comparative experimental investigation and advanced machine learning (ML) modeling to analyze a bio-fueled diesel engine. Experiments were performed on refining biodiesel properties through nano-fuel technology, specifically examining the impact of ZnO nanoparticles in pure diesel fuel B0 (100 % diesel and 0 % biodiesel) and B20 (80 % diesel and 20 % bio-diesel) blends on the engine performance. A novel Gaussian process regression (GPR) modeling is developed to forecast the thermal performance of engine considering four different bio-fuel blends nanoparticles additives, namely B0, B0ZnO, B20, and B20ZnO. The model incorporates thorough search optimization and implements a combination of squared exponential, rational quadratic, and matern function kernels to effectively determine optimal GPR parameter values and identify the finest kernel that maximizes the modeling accuracy. Experimental results reveal that ZnO nanoparticles integration enhances engine performance by augmenting heating value, improving heat transfer, and enhancing combustion efficiency. On average, the addition of ZnO nanoparticles results in 5.50 % and 4.90 % increase in engine torque, 4.99 % and 4.98 % improvement in brake thermal efficiency, and substantial reductions in specific fuel consumption by 6.01 % and 5.37 % for B0 and B20 biodiesel fuel blends, respectively. Moreover, the predicted simulations demonstrated that the GPR model's optimal solution is formulated by integrating the matern-32 kernel function. This formulation revealed superior results with a maximal determination coefficient of 0.9999 and minimal root mean square error (RMSE) of 0.0003 for predicting the nano-biodiesel engine performance. These findings underscore the potential of nano-fuel technology and AI methods in optimizing biodiesel fuel blends for efficient diesel engine operation amidst energy challenges.
| Type: | Article |
|---|---|
| Title: | Performance improvement of compression ignition engine fueled by second generation biodiesel fuel blends enriched with ZnO nanoparticles: Experimental study and Gaussian process regression AI modeling |
| DOI: | 10.1016/j.psep.2024.07.069 |
| Publisher version: | http://dx.doi.org/10.1016/j.psep.2024.07.069 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Compression ignition engines; Second generation biofuels; Nanoparticles fuel additives, Artificial intelligence modeling, Gaussian process regression. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Chemical Engineering |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10197149 |
Archive Staff Only
 |
View Item |
