AbstractIn process industry, there exist many Wiener systems with input magnitude constraints for which, however, most of the existing control algorithms cannot guarantee to have sufficiently large regions of asymptotic stability. In this paper, the subspace method is applied to separate the nonlinear and linear blocks in a constrained multi‐input/multi‐output (MIMO) Wiener system and a novel dual‐mode nonlinear model predictive control algorithm is developed to maximize the region of the asymptotic stability. Simulation results are presented to demonstrate the virtues of this new control algorithm. The limitation is the requirement that the state and input matrices of the Wiener system's linear block should be accurately identified. Copyright © 2009 John Wiley & Sons, Ltd.