Maximum Power Point Tracking for Photovoltaic System Using Model Predictive Control - Intelligent Computing for Sustainable Energy and Environment - page 515

Maximum Power Point Tracking
for Photovoltaic System
Using Model Predictive Control
Chao Ma, Ning Li, and Shaoyuan Li
Department of Automation, Shanghai Jiao Tong University
Key Laboratory of System Control and Information Processing
Ministry of Education of China, Shanghai 200240
ning [e-mail]
Abstract. In this paper, T-G-P model is built to find maximum power
point according to light intensity and temperature, making it easier and
more clearly for photovoltaic system to track the MPP. A predictive con-
troller considering constraints for safe operation is designed. The simu-
lation results show that the system can track MPP quickly, accurately
and effectively.
Keywords: PV system, MPPT, T-G-P model, MPC.
1 Introduction
In the recent decades, Photovoltaic (PV) power generation system has been paid
more and more attention. To make output power of photovoltaic system most,
energy converter should track Maximum Power Point (MPP) which is very im-
portant in order to ensure the ecient operation of the solar panel. Because
of solar panels' non-linear current-voltage characteristics, only one particular
operating point is the MPP (Fig.1). In addition, the MPP changes with temper-
ature and light intensity, making the tracking control of the MPP a complicated
problem.
To overcome these problems, many tracking control strategies have been pro-
posed. One of the most widely used methods is perturbation and observation
[1]. This method uses an algorithm known as ”hill climbing” and can be imple-
mented by applying a disturbance to the reference voltage signal of the solar
panel. P&O method is simple and suitable for changeable conditions, neverthe-
less a larger disturbance will lead to a higher value of oscillation amplitude [2].
Another conventional approach, PI controller, requires sucient time for the
system to reach steady state operation increasing the time interval between two
This work was supported by the National Science Foundation of China (60825302,
60934007,61074061), the High Technology Research and Development Program
of China (2011AA040901), the Key Project of Shanghai Science and Technol-
ogy Commission (10JC1403400), and Key Project of China's Ministry of Railways
(J2011J004).