Inverse fuzzy modeling for the cancellation of nonlinearity in unknown Hammerstein model

摘要

For the ease of controlling of an unknown Hammerstein model, it is usually desired to cancel out the effects of static nonlinearities. This requires the proper cancellation of all the nonlinearities or otherwise it may either create control problems or can result in the optimization complexity. The paper discusses a novel approach towards the designing of fuzzy inverse model controller that can effectively cancel out the effect of static nonlinearities in unknown Hammerstein models. The controller is used in conjunction with Generalized Predictive controller for controlling an unknown Hammerstein model. Simple single layer convex optimization is found sufficient to generate a converged optimized solution. Simulation results are presented to show the effective performance of the proposed controller.