Calculation of the Nusselt Module in a Spouted Bed Reactor from Artificial Neural Networks

The Nusselt modulus calculation is an important factor for the determination of the heat transfer of the particles. This calculation can be done by evaluating the fluid dynamics of the particles by means of the minimum fluidization velocity. On the other hand, artificial neural networks have been used since the last decade as a satisfactory alternative for the prediction of the fluid dynamic behavior of particles. Therefore, the objective of this work has been to develop a model based on artificial neural networks (ANN) suitable to quantify the influence of multiple factors on the rate of heat transfer in a conical bed reactor. The Nusselt module has been taken as the starting point and nine input factors have been evaluated, among which are the height of the bed, the diameter of the contactor, the angle of the cone and the minimum discharge velocity, among others. It has been found that the model fits appropriately to the equations proposed in the literature and can be used as a suitable model to predict the behavior of heat transfer in biomass-operated cone-bed reactors.