T-S fuzzy model predictive speed control of electrical vehicles

Area of Science:

• Control Systems Engineering
• Fuzzy Logic Control
• Nonlinear System Modeling

Background:

• Model Predictive Control (MPC) is crucial for complex systems.
• Takagi-Sugeno (TS) fuzzy models offer a way to represent nonlinear dynamics.
• Electric vehicle (EV) speed control faces challenges due to parameter uncertainty.

Purpose of the Study:

• To develop a novel nonlinear MPC using TS fuzzy models and LMIs.
• To guarantee stabilization and satisfy control input constraints.
• To improve EV speed tracking performance under uncertainty.

Main Methods:

• Formulation of a nonlinear MPC based on TS fuzzy models.
• Design of a non-parallel distributed compensation (non-PDC) fuzzy controller.
• Utilization of a non-quadratic Lyapunov function (NQLF) for stability analysis.
• Minimization of a quadratic cost function with infinite horizons subject to constraints.

Main Results:

• The proposed MPC, utilizing non-PDC and NQLF, guarantees system stabilization.
• The controller effectively manages control input Euclidean norm constraints.
• Simulations on a nonlinear EV model demonstrate superior speed tracking compared to conventional MPC and OFPI controllers.
• Performance validated using experimental data from the New European Driving Cycle (NEDC).

Conclusions:

• The novel nonlinear MPC approach provides robust and stable speed control for EVs.
• The method effectively handles parameter uncertainty and system constraints.
• This fuzzy-MPC framework offers a promising advancement for electric vehicle control applications.