A NMPC Scheme Based on Stirling's Interpolation Formula Approximation Method

Abstract

For a kind of nonlinear system whose input-output function is not differentiable, we proposed a model predictive control scheme based on linearization approximation method. We linearized the object nonlinear system using Stirling's interpolation formula method, and reformulated the control performance index to a quadratic optimization problem, and then, we obtained the optimization control sequences by solving the quadratic optimization problem. In order to reduce the complexity of computation, we ignored the high-order terms associated with the linearization. The simulations show that the presented NMPC scheme can achieve a satisfactory control result, also can decrease the dissipation of control energy and control time.