Optimal design for multiple mode charging station with PI controller using particle swarm optimization

Abstract

This project proposes a suitable design for charging station (CS) that can produce different charging categories (slow, medium, and high). The CS mode is based on power size, rated voltage levels, number of phases and rated current as well as charging time. The higher charging mode will require less time to complete the charging process as compared to lower charging mode. However, the integration of CS gives negative impact to the distribution system such as power loss and voltage profile. This is due to harmonic that produced from non-linear load of CS. Thus, this research focusing on the analysis of the effect of harmonic on different charging categories (slow, medium and high). The different mode of CS can be obtained by varying the value of pulse width in pulse generator at the buck converter. Next, the suitable universal design for a passive filter is proposed to reduce harmonic distortion based on design CS. A closed loop system is introduced by implementing PI controller in order to reduce the error between the output of CS with output demand. In term of analysis, the values of proportional constant (Kp) and integral constant (Ki) in this project had been obtained by using Particle Swarm Optimization (PSO). All the circuits are designed, simulate and analyses by using MATLAB/Simulink. From the result, it is proved that the installation of single tuned passive filter at 5th order and 7th order can reduce both THDv (2.94%-5.26%) and THDi (4.83%-9.11%). Besides that, closed loop system shows better performance as compared to open loop system in term of THD level and fulfill output demand. The results also fulfill the recommendation of IEEE 519 power harmonic standard.