¹Department of Automation, North China Electric Power University, Baoding, China
²State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, Beijing, China

---

Abstract—In order to make full utilization of photovoltaic (PV) array output power, which depends on solar irradiation and ambient temperature, maximum power point tracking (MPPT) techniques are employed. Among all the MPPT strategies, the Perturb and Observe (P&O) algorithm is more attractive due to its simple control structure. Nevertheless, steady-state oscillations always appear due to the perturbation. In this paper, forecasting model of maximum power point (MPP) based on Support Vector Machine (SVM) is established, and a new MPPT algorithm composed of the forecasting model and small step P&O is presented. Experimental results show that SVM model could predict the MPP exactly, and the effectiveness of the proposed MPPT algorithm is validated using hardware platform based on single-chip microcomputer.

Index Terms—Maximum power point tracking, Perturb and Observe, support vector machine, forecasting model.

[PDF]

Cite: Weiliang Liu, Changliang Liu, Liangyu Ma, Yongjun Lin, Jin Ma, "Maximum Power Point Tracking of Photovoltaic Generation Based on Forecasting Model," Journal of Software vol. 8, no. 10, pp. 2569-2574, 2013.