A Scenario Optimal Reduction Method for Wind Power Time Series

Generating representative scenarios of wind power time series to reflect its stochastic characteristic is important for power system planning and operation. However, there is no comprehensive model to optimize the space and moment distances between original and generated scenarios simultaneously, which deteriorates the quality of generated scenarios. This letter proposes an optimal model to minimize the space and moment distances simultaneously and proposes a heuristic search method (HS) to solve it. Simulation results show that scenarios generated by the proposed model have better ability to approximate the original scenarios set both in moment distance and space distance than the traditional optimization model. Meanwhile, the HS method is much faster than the traditional backward/forward reduction methods.