Adaptive Neural Control of a Two Phase Hydrocarbon Separator

I.S. Baruch, J. Martin Godoy A., J. Roberto Valle R., and J. Martin Flores A. (Mexico)


Recurrent Neural Networks, Systems Identification, Backpropagation Learning, State Estimation, Adaptive Control, Two Phase Hydrocarbon Separator.


A parametric Recurrent Neural Network (RNN) model and an improved dynamic Backpropagation (BP) method of its learning are applied for a two phase hydrocarbon separator plant model identification and state estimation. The obtained parameters of the RNN model are used for adaptive control system design. The paper suggests to apply two main types of state-space control schemes with RNN states estimation: an indirect and a direct adaptive reference trajectory tracking control schemes, where the RNN is also a controller. The applicability of the proposed adaptive neural control schemes is confirmed by simulation results.

Important Links:

Go Back