Short-term wind power interval probability prediction method

Abstract

The invention discloses a short-term wind power interval probability prediction method, which includes: obtaining a number of historical wind power of a wind farm as a sample set; constructing an optimization criterion by combining the coverage probability of the prediction interval, the root mean square of the bandwidth of the prediction interval, and the average offset; The short-term wind power interval prediction model of the worker bee colony neural network uses the artificial bee colony algorithm to optimize the optimization criteria and update the weight threshold of the neural network; establishes the neural network according to the optimal weight threshold, and performs interval prediction of the wind power to be predicted; The state of wind power is divided, the Markov chain prediction model is established, and the state transition probability of each state is calculated; the wind power power interval is predicted through the Markov chain state transition probability combined with interval prediction, and the probability is calculated for the numerical points in the prediction interval. The present invention considers the probability distribution of numerical points in the interval while predicting the short-term wind power interval, and provides a basis for optimizing the system.

Description

technical field [0001] The invention belongs to the technical field of electric power systems, and in particular relates to a short-term wind power interval probability prediction method. Background technique [0002] Wind energy is an ideal clean energy. As the installed capacity of wind power continues to increase, the gaps and uncertainties of wind power bring new challenges to the stable operation of the power system and grid dispatch. Realizing the accurate prediction of wind power is of great significance to the power balance, economic dispatch and equipment safety of the power system. It can not only optimize the reserve capacity and reduce the operating cost of the power grid, but also reduce the impact of wind power on the power grid and improve the reliability of the power grid operation. Wind power prediction methods can generally be divided into two categories: one is to use numerical weather forecasting to establish a prediction model to convert weather data i...

AI Technical Summary

Problems solved by technology

[0003] At present, wind power forecasting is generally a single point forecast, and wind power output is affected by many uncertain factors such as turbulence, wind turbine status and background noise, and a single point forecast cannot accurately reflect the uncertain factors.

Images