Wind Power Interval Prediction Method Based on Kernel Extreme Learning Machine Quantile Regression

Abstract

The invention discloses a wind power interval prediction method based on nuclear extreme learning machine quantile regression, comprising: collecting wind farm output power and wind speed data; performing simple data processing and deleting unreasonable data; building a nuclear extreme learning machine Quantile regression model; use particle swarm optimization to optimize the quantile regression parameters of the kernel extreme learning machine and determine the regression model; bring in the test data to obtain the wind power prediction interval. The present invention effectively combines the quantile regression principle with the kernel extreme learning machine model, obtains optimal model parameters through particle swarm algorithm search and optimization, can effectively grasp uncertain information in wind power, and then obtain better predictions As a result, it can provide a basis for the safe and stable operation of wind power integration.

Description

technical field [0001] The invention belongs to the technical field of wind power prediction, and in particular relates to a wind power interval prediction method based on quantile regression of nuclear extreme learning machine. Background technique [0002] As a non-polluting and renewable new energy, wind energy has been widely used. However, due to the strong randomness and volatility of wind, as the proportion of wind power in the grid continues to increase, the randomness and fluctuation of wind power itself The stability will lead to fluctuations in the power grid, which is not conducive to the safe and stable operation of the power grid. For the safe and stable operation of the power grid, we need to make accurate predictions of wind power. Accurate predictions of wind power generation can reduce the dispatching pressure of the power grid and allow the power grid to accept more wind power. The traditional point forecasting method is easily affected by factors such as...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the disadvantages of low reliability of existing prediction methods, complex calculations, and difficult determination of parameters, the present invention proposes a wind power interval prediction method based on quantile regression of nuclear extreme learning machine, which includes the following steps:

Images