Back to Search
Start Over
Internal combustion engine fuel synthesis, suitability, physical property evaluation using mixing models and backpropagation ANN algorithm.
- Source :
-
Engineering Applications of Artificial Intelligence . Jun2024, Vol. 132, pN.PAG-N.PAG. 1p. - Publication Year :
- 2024
-
Abstract
- An experimental analysis comprising pongamia, soyabean, and corn biodiesel was selected and mixed with diesel and di-butyl ether at different proportions for its viability and phase separation. The test samples were experimentally confirmed for their density, sound speed, and refractive index characteristics by varying the temperatures between 298 K and 343 K. The mixing models, namely the Lorentz-Lorenz (L-L) and Gladstone-Dale (G-D) methods, were tested for the uncertainty of experimental results. The use of artificial neural networks (ANN) was made as an attempt to characterize the backpropagation algorithm's gradient to pursue the minimum error function of experimental and mixing models. The L-L and G-D with and without ANN equations were used to compare the different proportions and temperatures for the fuel blends of D50PD50, D50PD45DBE05, and D50PD25DBE25, and so on. Furthermore, each sample has been computed for the excess molar volume, and interactions between the liquids have been verified. The exergy analysis for PD, SB, and CB biodiesel under ambient conditions and the experimental and predicted results of energy destruction can be narrowed. The use of ternary fuel in place of diesel shows significant improvements in terms of energy annihilation and homogeneity, which have a lesser impact on emissions and the environment. [Display omitted] • Qualitative proportions endorse fuel parameters due to temperature influences. • Experimental and mixing model error analysis using the ANN backpropagation algorithm. • Hypothetical and attainable outcomes of ternary blends in energy and exergy analysis. • Exergy analysis provides substantial prospects for resolving environmental problems. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 09521976
- Volume :
- 132
- Database :
- Academic Search Index
- Journal :
- Engineering Applications of Artificial Intelligence
- Publication Type :
- Academic Journal
- Accession number :
- 177088702
- Full Text :
- https://doi.org/10.1016/j.engappai.2024.107970