Artificial Neural Network-based Prediction of Bed Expansion Ratio in Gas-solid Fluidized Beds with Disk and Blade Promoters

Abstract

Artificial Neural Network (ANN) models have been developed to predict bed expansion in gas-solid fluidized bed promoted with blade and disk promoters. To model the above, two systems (eight variables problem in case of bed with disk promoter and six variables problem in case of bed with blade promoters) have been undertaken. For the training of the input-output data, the experimental values of bed expansion ratio collected under different varying conditions of the system parameters have been used. The system variables include seven numbers of disk promoters of varying disk thickness and dia and one blade promoter in addition to five numbers of distributors of varying orifice sizes, four type of bed materials, five sizes of bed material and four initial static bed heights. The values of bed expansion ratio predicted with the help of developed ANN models for respective beds have been found to be closer to the corresponding experimental ones and those obtained using developed correlations1 on dimensional analysis approach.