Robust model predictive control of nonlinear and time-delay systems

• Main Author: Teng, Long
• Other Authors: Cai Wenjian
• Format: Theses and Dissertations
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
• Online Access: http://hdl.handle.net/10356/73381
• Institution: Nanyang Technological University

Industrial systems usually suffer from severe nonlinearities and large time delays, which may cause degradation of the control performance or even instability of the whole system. Control of such systems is not an easy problem especially if the two characteristics exist simultaneously. The thesis is to develop advanced control methodologies for such systems. This thesis is divided into three parts: (1) robust model predictive control of nonlinear systems modeled as T-S fuzzy systems with nonlinear local models; (2) robust fuzzy model predictive control of nonlinear systems with time delays; (3) model reference tracking control of networked systems subjected to time delays and package dropouts, and its practical application to a linear brushless dc motor. In the first part of this thesis, robust model predictive control of a recently developed T-S fuzzy system that relies on nonlinear local models is investigated. Due to the advantage of the T-S fuzzy system with nonlinear local models, the computational burden of online optimization for model predictive control is decreased when compared to an existing fuzzy model predictive control approach of T-S systems with linear local models. In the second part of this thesis, we derive robust model predictive control of nonlinear time-delay systems, with a special focus on the Lyapunov Razumikhin function rather than the Lyapunov Krasovskii functional. Both online and efficient off-line model predictive control algorithms are considered for systems with multiple delays and time-varying delays, respectively. We show that the proposed methods have better performances as well as computational advantages over several existing model predictive control approaches for time-delay systems. In the third part, taking the advantages of the Lyapunov Razumikhin approach in dealing with systems with time-varying delays which is also mentioned in the second part, model reference tracking control of a networked linear brushless dc motor is investigated, with both networked induced communication delays and data package dropouts involved. Simulation and experimental results are provided to verify the effectiveness of the proposed method.