Modelling and simulation of reactive gas-solid flows in circulating fluidized bed reactors
MetadataVis full innførsel
The thesis presents a study on mathematical modelling and numerical simulation of reactive gas-solid flows operating in circulating fluidized bed (CFB) reactors, in particular environmentally-friendly methane reforming processes for hydrogen production. The total mass, momentum, heat and species mass balances, simplified to their one-dimensional forms, were discretized over the problem domain using the Finite Volume Method and implemented numerically with Matlab. The transfer of solids between the reactor units constituting a CFB reactor was modelled via the addition of source and sink terms in the governing equations. A Proportional-Integral-Differential (PID) controller model was used in each reactor unit to regulate the temperature of the solid phase and keep it close to optimal values. A PID controller was also used to regulate the solid circulation rates between the CFB’s reactor units through control of the inlet gas velocities. The model developed represents a step further in complexity from the conventional Kunii-Levenspiel type of models for CFB reactors. The Kunii-Levenspiel models assume a stagnant or steady state condition for the solids in the bed, which is not adequate because the solid circulation rates must be calculated and because the properties of the solids change over time. Yet the one-dimensional model developed represents a trade-off between accuracy and computational cost. The chemical processes studied were the Steam Methane Reforming (SMR) process operated in a bubbling bed fluidized bed reactor as well as the following processes operated in a CFB reactor: the Sorption-Enhanced Steam Methane Reforming (SE-SMR) process, the Chemical Looping Reforming (CLR) process and the CO2 post-combustion adsorption process. In order to validate the fluid dynamic model and the kinetic models of the different chemical processes, simulation results were compared with experimental data from the literature. The performance of each process was investigated, and the solid circulation rates as well as the heat transfer between the reactor units of the CFB due to solid circulation were found to be of importance. In the CO2 post-combustion adsorption process it was found that there is a maximum calciner diameter that can be employed due to the high heat exchange requirements imposed by CO2 desorption. Simulations of the SE-SMR process usually indicated a dry hydrogen mole fraction above 95%. In one publication, the optimal range of operation for the SE-SMR process was found to be between 830 K and 930 K. For the CLR process, a model was developed and validated in which Ni-based solid particles act both as catalyst for the SMR reactions and as oxygen carriers. A study on the energy requirements for hydrogen production via three different chemical processes was also conducted. In a first stage, the fuel was reformed via either SMR, CLR or SE-SMR. In a second stage, the product stream information from the first stage was used as input in a Pressure Swing Adsorption (PSA) unit model for hydrogen purification. The results indicated that the SMR process required the least energy for hydrogen production, followed by the CLR and the SE-SMR processes.