Long Term Optimal Scheduling of Large Scale Renewable Energy Integration System Considering Multiple Uncertainties

Abstract

The multiple uncertainties of wind, photovoltaic and hydropower are very challenge to the long-term optimal scheduling of large-scale renewable energy grid-connected systems. In order to address this problem, a long-term optimal operation theory method based on quantile nonlinear regression theory and fuzzy chance theory is proposed. Aiming at the unpredictability of wind and photovoltaic output, a nonlinear quantile regression model between the long term forecast value and the actual value of wind and output is established, and the risk caused by directly applying the deterministic results to the power system dispatching is avoided. In addition, in view of the problem of errors caused by insufficient historical data for natural inflows, the reservoir storage capacity at the beginning of the dispatch cycle is regarded as a trapezoidal fuzzy quantity to describe the uncertainty in hydropower operation. And through equivalent conversion, the fuzzy chance constraint is transformed into a deterministic model for solving. Finally, the proposed approach is tested in a large-scale renewable energy integration case with 3 cascade hydropower stations, 1 wind farm and 1 photovoltaic power station and a few thermal power units. The numerical results demonstrate that the proposed method can cope with multiple uncertainties of various renewable energy sources in long term scheduling.