Model-Free Based Anti-Disturbance Predictive Control for Nonlinear Systems

Aiming at the issues of model mismatch and poor performance of model-based control methods under the influence of measurement perturbations, a novel model-free adaptive predictive control method (MFAPC) with anti-disturbance scheme is proposed in this paper. First, the compact form dynamic linearization (CFDL) method is used to equate the complex nonlinear model to an incremental dynamic linear time-varying model. Then an improved unscented Kalman filter based on the I/O data of the system is derived based on the dynamic linearized data model of the controlled system and the statistical properties of the measurement noise. On this basis, the generalized model in model-free control is applied to the predictive control of nonlinear systems, and the prediction equation is derived. Finally, the optimal control law is solved according to the optimization strategy of predictive control rolling. The analysis of the experimental results shows that the proposed method has strong anti-interference and robustness, and has a fast dynamic response.