A Probabilistic Prediction Method of Wind Power Power Based on Hierarchical Integration

Abstract

The invention discloses a method for probabilistic prediction of wind power based on hierarchical integration. The method constructs a subspace set through resampling and partial least squares, uses GMM clustering on each subspace to obtain multiple local regions, establishes a corresponding local GPR model, and uses Bayesian inference strategy and limited mixing mechanism fusion. The local model builds the first-level ensemble model. The genetic algorithm is used to select the suitable first-layer ensemble model for selective adaptive ensemble, and the selective hierarchical ensemble Gaussian process regression probability prediction model is obtained. In order to solve the problem of performance deterioration caused by the change of wind power data characteristics, an adaptive update strategy is introduced to enable the prediction model to have the ability to adaptively update. The invention uses a selective hierarchical integrated learning framework for ultra-short-term wind power prediction. Compared with the traditional integrated learning prediction method, the invention has higher prediction accuracy and stability, and the generated prediction interval can provide power for power dispatching. valid reference.

Description

technical field [0001] The invention relates to the technical field of wind power prediction, in particular to a method for probabilistic prediction of wind power based on hierarchical integration. Background technique [0002] Wind energy is a kind of non-polluting, widely distributed renewable energy, and wind power generation technology has developed rapidly in recent years. However, due to the randomness and volatility of wind energy, unstable wind power grid access will have an impact on the security and stability of the power system, which will affect the stable operation of power grid equipment. Therefore, accurate and efficient wind power prediction can effectively promote reasonable power dispatching, provide a reliable reference for the grid to arrange power generation plans, shutdown maintenance, and help ensure the safe, reliable and economical operation of the system. Wind power prediction plays a vital role in the development of the power generation industry i...

AI Technical Summary

Problems solved by technology

However, the problem of adaptation of ensemble models has been little explored in recent studies

[0004] Finally, due to the strong randomness and high uncertainty of wind energy, the traditional single-point prediction cannot make a good estimate of the wind uncertainty. For the stability of the power system, the integration of wind power needs to There is a relatively accurate estimate of the fluctuation range of the market, and a single-point forecast is not enough

Images