CFD-based deep neural networks (DNN) model for predicting the hydrodynamics of fluidized beds
Mahesh Nadda, Suresh Kumar Shah, Sangram Roy, Ashutosh Yadav
Abstract
Fluidized beds are central to numerous applications such as drying, combustion, gasification, pyrolysis, CO2 utilization, mixing, and separation. The design and development of fluidized beds still need to be improved owing to the complex hydrodynamics. Various experimental investigations and CFD simulations are carried out to understand the hydrodynamics of these beds. Although the experimental approaches are very costly and limited to small scale, on the other hand, CFD modeling requires a lot of computational resources and time. In this contribution, we propose a hybrid CFD-based ML model for estimating the hydrodynamics of fluidized beds. The CFD simulations of Taghipour et al., 2005 were performed and validated with the experimental results for the wide range of inlet gas velocities covering various flow regimes. A time-averaged simulation data of the CFD model was used for the Deep Neural Network (DNN) model. The hydrodynamic parameters, such as solid velocity field, volume fraction, and bed pressure drop, are predicted using the CFD-based DNN model. The results demonstrate that DNN has superior spatial learning capabilities and that, when used with CFD, it can reduce the computational power required without sacrificing accuracy. To evaluate the versatility of the CFD-based DNN model with different operating conditions and hydrodynamic parameters, data from (Cloete et al., 2013) and (Tingwen et al., 2013) were used, and the results showed good agreement.