Due to the power-voltage characteristics of photovoltaic(PV) arrays under partially shaded conditions are characterized by multiple peaks, the traditional single-peak maximum power point tracking(MPPT) schemes are unable to extract maximum power point(MPP) from the PV array. The highly non-linear and multi-peak characteristics of PV make the tracking of the real maximum power point(RMPP) a difficult task. This paper proposes a novel algorithm based on Monte Carlo to track the global power peak under partially shaded conditions. In this paper, a probabilistic model of the global maximum power point tracking is constructed to prediction the MPP. In order to accelerate the tracking speed, the formulation of the algorithm is based on principle of statistical sampling. Then, the single-peak MPPT algorithm is used to track RMPP. This research paper inspires its motivation from the fact that the algorithm is extremely high accuracy in the case of larger sample and the convergence speed is not affected by the complex of multiple peaks. At last, the experiments and simulations demonstrate that the new method could track global MPP effectively under complex irradiance conditions. ©, 2015, Chinese Machine Press. All right reserved.