| Autor: |
Jennifer C. Oraegbunam, Niyi B. Ishola, Babajide A. Sotunde, Lekan M. Latinwo, Eriola Betiku |
| Jazyk: |
angličtina |
| Rok vydání: |
2023 |
| Předmět: |
|
| Zdroj: |
Green Technologies and Sustainability, Vol 1, Iss 1, Pp 100007- (2023) |
| Druh dokumentu: |
article |
| ISSN: |
2949-7361 |
| DOI: |
10.1016/j.grets.2022.100007 |
| Popis: |
This current study focused on biobased diesel fuel production from sandbox (Hura crepitans) oil (HCO) through a single-step transesterification process with KOH as the base catalyst and methanol as the alcohol. The process was modeled using artificial neural networks (ANN), and the essential parameters were optimized with the genetic algorithm (GA) method. The results obtained were compared with that obtained with the Taguchi method. The input process parameters and the ranges investigated are methanol–HCO molar ratio (6:1–15:1), temperature (35–55 °C), time (25–75 min), and catalyst dosage (0.5–1.5 wt%). The developed model was statistically significant based on the coefficient of determination R2(0.8563) and mean relative percent deviation, MRPD (1.9145). The optimum values of the parameters were the methanol–HCO molar ratio of 6:1, temperature of 35 °C, time of 25 min, and catalyst dosage of 0.955 wt% with actual HCO methyl esters (HCOME) yield of 99.03 wt%. The Taguchi method was superior in modeling the transesterification process with R2of 0.9953 and MRPD of 0.3626 but lower in optimization efficiency with an HCOME yield of 97.1 wt%. Thus, ANN-GA performed satisfactorily in optimizing the HCOME production from HCO. Also, the HCOME produced met the American and European biodiesel standards and could serve as a transport fuel. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
