LAUSR

This work presents the development and implementation of a hybrid intelligent system (neuro-fuzzy) based controller applied to a non-linear flow control process in oilfield equipped with electrical submersible pumping as artificial lifting method. The efficient control of industrial plants requires that the user apply unconventional techniques, particularly when using different set-points or when it is required to follow certain system reference trajectories. The need to obtain expert knowledge in inference fuzzy logic membership functions and parameters are set automatically, using for training their own application data. The neuro-fuzzy model obtained is used to generate control actions in series the process presenting a suitable dynamic response.