ABSTRACT
The predictive control strategies for input nonlinear systems can be classified into two categories. These are input saturation and Hammerstein nonlinearity. Two-step control applies to a class of special systems, that is, systems with input nonlinearities. Input saturation can be regarded as a kind of input nonlinearity, and also can be regarded as a constraint in the optimization. The nonlinear separation with respect to the Hammerstein model invokes the special structure of the Hammerstein model, and groups the controller designing problem into the linear control, which is much simpler than the overall category. For the Hammerstein model with input saturation, first utilize the linear sub-model and unconstrained Model predictive control algorithm to compute the desired intermediate variable, and then solve the nonlinear algebraic equation to obtain the control action, and utilize desaturation to satisfy the input saturation constraint. The input saturation constraint is usually inevitable in the real applications.
