Process Analysis And Optimization Of Crude Glycerol Autothermal Reforming Using Response Surface Methodology And Artificial Neural Network

Download or read book Process Analysis and Optimization of Crude Glycerol Autothermal Reforming Using Response Surface Methodology and Artificial Neural Network written by Christian Ekejiuba Nwosu and published by . This book was released on 2020 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: There is an urgent need to curtail the effects of global warming through the development of efficient, cheap and accessible alternative source of energy since the combustion of carbon containing compounds has become a problem. The combustion of hydrogen gas for example for either heat generation or use as fuel cells in automobiles produces only water vapour as by-product. Though, water vapour is also greenhouse gas, it does not last long in the atmosphere when released since hydrologic cycle ensures the excess water vapour is sent down via precipitation. Based on this and in addition to its high heat capacity, the use of hydrogen as an alternative to fossil fuel has become highly beneficial. Hydrogen gas can be cheaply produced from glycerol; a major waste product in bio-diesel gas production which contains very high impurities and requires a lot of money for purifications. This research work therefore focused on the in-situ production of hydrogen gas as well as optimization from crude glycerol using a very stable catalyst without having to undergo some purification steps thereby saving cost. The methodology deployed in this research work in their respective order involves; (1) the preparation of 5%Ni/CeZrCa catalyst with great performance and activity (2) characterization of the prepared catalyst via: TGA, N2 physisorption, XRD, TPR and XRF, (3) design of experiment using central composite design (CCD), (4) preparation of crude glycerol which mirrors waste glycerol from a typical bio-diesel production station, (5) laboratory experiment for hydrogen gas production through autothermal reforming of crude glycerol method which involves two important parts; partial oxidation and steam reforming processes. The experiments were carried out following the design of experiment with CCD which involves catalytic and non-catalytic runs to evaluate the performance of the catalyst. (6) optimization of hydrogen gas using response surface methodology (RSM) and artificial neural network (ANN). The design of experiment was executed with the following range of input variables; temperature: 500 - 650oC, crude glycerol flow rate: 0.0019 - 0.0033 mols C/min, Steam to carbon ratio: 1.6 - 3.6, Oxygen to carbon ratio: 0.05 - 0.2, Catalyst weight: 0.05 - 0.25g. The highest feed conversion and yield were 92 mol% and 0.995 mols respectively at temp: 575oC, feed flow rate: 0.0026 mols C/min, S/C center point of: 0.78, O/C: 0.13 and catalyst weight of 0.15g. The fitted quadratic model using RSM gave a coefficient of determination R2 of 0.951, an RMSE of 1.51% and AAD of 3.14%. and also established best operating conditions for hydrogen productions as temp: 650oC, feed flow rate: 0.0033 mols C/min, S/C: 2.34, O/C: 0.052 and catalyst weight of 0.15g with catalyst size constantly maintained at 0.05mm. Furthermore, the results of the variable ranking of the ANN using the connection weight approach showed that temperature has the highest influence with a ranking weight of 53% while O/C has the least influence with a ranking weight of 3%. The regression analysis of the ANN using the Levenberg-Marquardt Model and TANSIG as transfer function gave better results with an R2 of 0.999, MSE: 10-8, AAD: 0.34% and RMSE: 3.5x10-4%. A standard quadratic polynomial equation has been established for hydrogen gas production and optimization in the autothermal reforming process. Also, optimum conditions required for hydrogen gas production has been established. From the results of RSM and ANN predictions and those of R2, AAD, RMSE and MSE obtained, it is evidently clear that the ANN model was superior to RSM in predicting hydrogen gas.