Configuration optimization method and apparatus of measurement point of flow field of wind farm

Abstract

The embodiment of the application discloses a configuration optimization method and apparatus of a measurement point of a flow field of a wind farm. The method includes steps: establishing an initialsample space, wherein the initial sample space includes at least one sample of a wind farm flow field measurement point position deployment decision, the sample of the wind farm flow field measurementpoint position deployment decision includes the position of at least one wind farm flow field measurement point, and the wind farm flow field measurement point is configured on a fan and / or a wind measurement mast; obtaining N samples of the wind farm flow field measurement point position deployment decision from the initial sample space; ranking the N samples according to linear approximation error results of a one-dimensional flow field curve of the wind farm, selecting samples whose linear approximation error results are less than a preset value from the N samples, and forming an ideal sample space S; and performing accurate simulation evaluation with the wind farm output power prediction precision on the ideal sample space S, and obtaining a position optical solution omega of the measurement point of the flow field of the wind farm.

Description

technical field [0001] The present application relates to the field of wind power generation, in particular to a configuration optimization method and device for flow field measurement points of a wind farm. Background technique [0002] The installed capacity of wind power in China ranks first in the world. At the same time, the annual wind power curtailment has grown rapidly, and the average wind curtailment rate has reached 21%. The large uncertainty of wind farm power is an important factor affecting the grid connection of wind power and resulting in the increase of curtailed wind power. Optimizing the power prediction effect of wind farms is very important to improve the stability of wind power grid connection and reduce the curtailment of wind power. The energy industry standard stipulates that "the monthly root mean square error of the ultra-short-term forecast at the 4h forecast value should be less than 0.15". At present, the root mean square error of the predict...

AI Technical Summary

Problems solved by technology

One of the main problems of low prediction accuracy is that the location of the wind farm flow field measurement point is not reasonable enough, which will lead to the poor use of the characteristics of the sensors at the wind farm flow field measurement point

[0005] However, this solution is dedicated to reducing the types and quantities of sensors required in power prediction based on the existing sensors. That is, the configuration of flow field measurement points in the prior art is mainly realized by grouping fans and adjusting grouping rules. To further improve the prediction accuracy of wind farm power, how to optimize the position of the flow field measurement point is not involved. For example, in the wind farm or in each group, which wind turbine cabin or configuration is the flow field measurement point specifically configured in? on which wind tower

Images