Error Analysis on Renewable Generations and Decision Making Method for Energy Storage Allocation Based on Vasicek Model

The probabilistic analysis on the output prediction error of renewable generations is of great significance to reserve energy decision making of the power system. Regular probabilistic analysis of renewable generations is usually based on specific analytical forms. It is not only inflexible but also prone to errors. To compensate for the influence of randomness, centralized renewable generations are typically equipped with sufficient energy storage, which may have impacts on their economic benefits. In view of this, from the perspective of stochastic process, according to a unique stochastic differential equation, that is, Vasicek model, we put forward a method for estimating the parameters of the renewable generation output model. This method can avoid analytical expression of the probability distributions. Later, an improved configuration method for the capacity of energy storage system is put forward. Case studies indicate that our method has good performance in terms of short term prediction, and the energy storage allocation decision made based on our method is more precise and economic.