Method for predicting wind power plant short-term power

Abstract

The invention provides a method for predicting wind power plant short-term power. The method includes the steps of firstly, setting up a wind speed statistical downscaling model of a single draught fan; secondly, generating a predicted wind speed at the height of a hub in the position where the single draught fan is located according to predicting factors of mesoscale numerical weather prediction for a wind power plant area in 48 hours and the statistical downscaling model of the single draught fan in the first step; thirdly, setting up a wind speed, wind direction-power model of each draught fan, and obtaining power prediction of each draught fan according to a wind direction predicted through the mesoscale numerical weather prediction and a wind speed predicted in the second step. According to the method, the uncertainty caused by lack of mesoscale resolution is effectively reduced, and the prediction accuracy of the wind power plant short-term power is remarkably improved.

Description

technical field [0001] The invention relates to a short-term power prediction method of a wind farm. Background technique [0002] In recent years, with the adjustment of my country's energy policy, the installed capacity of grid-connected wind power has grown rapidly, and the centralized grid-connection of large-scale wind power has brought impact on the safe operation of the power grid. Improving the predictability of wind farm output can effectively reduce the impact of wind power on the power grid and reduce the pressure of power grid dispatching and peak regulation. This is of positive significance for making full use of wind energy resources and further increasing the proportion of grid-connected wind power installed capacity. According to the technical specifications of wind farm power forecasting that have been promulgated and implemented in China, wind farms must upload the output forecast curves for the next 24 hours to the power dispatching agency according to the...

AI Technical Summary

Problems solved by technology

This method has a small computational cost, but the physical model construction and implementation process is more complicated, and the accuracy improvement is limited

[0004] 2. Use dynamic downscaling methods, such as Computational Fluid Dynamics (CFD) to simulate the evolution process of the internal flow field of wind farms. This method can obtain more accurate wind speed distribution, but it is necessary to use CFD methods when establishing a predicted wind speed query database or directly predicting wind speed Solving the Navier-Stokes equation, the engineering implementation is complex and the calculation cost is huge, and the hardware requirements are extremely high

Images