Neural network modeling and control of cold flow circulating fluidized bed

Circulating fluidized beds (CFB) are a relatively new method of forcing chemical reactions to occur in chemical and petroleum industries. Compared with conventional fluidized beds, CFB have many advantages including better interfacial contacting and reduced back mixing. The recycle nature of CFB allows for a better process, but also makes modeling and understanding it many more times difficult. The plant under consideration is a cold-flow circulating fluidized bed (CF-CFB), meaning there is no combustion component in it. In the absence of conventional means to derive a reliable model, we have devised a model of the CFB using neural networks (NN), which have the ability to characterize such complex systems. This stems from their ability to approximate arbitrary nonlinear mappings. The main objective is to train a NN model and controller to simulate and control the CFB operation. It has been shown that a NN can be used effectively for the identification and control of nonlinear dynamical processes. Results are presented.