Maximum Power Point Tracking of Photovoltaic Generation Based on Forecasting Model

摘要

In order to make full utilization of photovoltaic(PV) array output power, which depends on solarirradiation and ambient temperature, maximum powerpoint tracking (MPPT) techniques are employed. Among allthe MPPT strategies, the Perturb and Observe (P&O)algorithm is more attractive due to its simple controlstructure. Nevertheless, steady-state oscillations alwaysappear due to the perturbation. In this paper, forecastingmodel of maximum power point (MPP) based on SupportVector Machine (SVM) is established, and a new MPPTalgorithm composed of the forecasting model and small stepP&O is presented. Experimental results show that SVMmodel could predict the MPP exactly, and the effectivenessof the proposed MPPT algorithm is validated usinghardware platform based on single-chip microcomputer.