A Non-parametric Probability Forecasting Method of Short-term Wind Power

Abstract

The embodiment of the present invention discloses a short-term wind power non-parametric probabilistic forecasting method, which includes constructing an SVM forecasting model and an SBC forecasting model for each forward-looking period; inputting the data required for wind power forecasting into the SVM forecasting model to obtain each forward-looking period The predicted value of wind power; input the data required for error distribution prediction into the SBC forecasting model, and obtain the conditional probability of forecast error in each forward-looking period; use D‑S evidence theory to integrate the conditional probability of forecast error, in which the distribution range of wind power is designed Constraints, the overall probability distribution of the forecast error in each forward-looking period is obtained; the wind power forecast value and the forecast error probability distribution are superimposed to obtain the wind power probability distribution in each forward-looking period. The invention is constructed on the basis of sparse Bayesian architecture, has high sparsity, ensures the generalization ability and calculation speed of the model, and systematically takes into account the boundary constraints of the output power of the wind farm, so that the prediction results are more realistic.

Description

technical field [0001] The invention relates to the technical field of wind power prediction in the process of new energy power generation, in particular to a short-term non-parametric probability prediction method of wind power. Background technique [0002] Wind energy is a renewable and clean energy. The development of wind energy has been highly valued by various countries. Wind power has become one of the fastest growing and most mature technologies in renewable energy. However, the volatility and uncontrollability of wind energy And other characteristics, resulting in the fluctuation and intermittency of the output power of the wind farm. In turn, the access of wind power has brought impacts to the power grid, increasing the uncertainty of the power grid and increasing the difficulty of power dispatching. Therefore, accurate forecasting of wind power is conducive to reducing the impact of wind farms on the grid, reducing adverse effects, and improving the ability of w...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a short-term wind power non-parametric probability prediction method to solve the problem of large differences between the prediction results of the wind power probability prediction method in the prior art and the actual situation

Images