| Summary: | This thesis presents the energy management of battery, ultracapacitor (UC) and photovoltaic (PV) power system for Electric Vehicles (EV). The proposed energy sharing control, in which each energy source is connected in parallel to the direct current (DC) bus via a power electronic converter. All energy sources are studied to investigate the different supply characteristics in order to avoid detrimental effects on the energy sources. A total of four control loops are employed in the supervisory system in order to regulate the DC bus voltage. The Proportional- Integral (PI) compensator is used in each control loop to simplify the overall system design. In this work, the Generalized Predictive Controller (GPC) is proposed to control the multiple energy system. The simulation results are then compared with the conventional PI control technique. The energy management strategy is designed according to typical vehicle operation modes based on the state-of-charge of energy storage devices and the total output power. The performance of the EV's energy management using both the PI controller and the GPC is simulated using resistive load and a DC motor drive system through MATLAB/Simulink simulation package. Then, the feasibility of the control system is validated through laboratory scale experimental tests. In the experiment, the dSPACE DS1104 is used to implement PI controller into hardware. The responses of the DC bus are analyzed based on different vehicle operation modes. Results show the proposed parallel energy-sharing control system either in simulation or hardware experiment is able to provide a dynamic response, avoid battery being overstressed by current, the UC charged according to vehicle speed, and the PV tracked the maximum power. However, between the discrete PI and GPC control, the GPC is slightly better than PI control.
|
|---|
