Wind power prediction error modeling method based on meteorological mode recognition

Abstract

The invention relates to a power grid wind power prediction technology, and especially to a wind power prediction error modeling method based on meteorological mode recognition. The method comprises the steps of: performing k-means clustering analysis on historical meteorological data to obtain a corresponding meteorological mode; according to meteorological data characteristics under each meteorological mode, training a support vector machine classifier; dividing the historical wind power prediction error data into sub-data sets in each meteorological mode by using the historical wind power prediction error data; and performing statistical analysis on the sub-data sets to obtain corresponding probability density curves, and obtaining a wind power prediction error probability density modelin each meteorological mode through least square fitting based on a universal distribution model to complete wind power prediction error modeling. According to the method, the influence of meteorological factors on the wind power prediction precision is considered, so that the wind power prediction error modeling result is more accurate; a universal distribution model is adopted, so that the fitting effect is better, and the analytic property of an expression is better; and an accurate wind power prediction error probability density model is provided.

Description

technical field [0001] The invention belongs to the technical field of grid wind power forecasting, and in particular relates to a wind power forecasting error modeling method based on meteorological pattern recognition. Background technique [0002] At present, the problem of wind power accommodation is to judge the future of the system under the premise that the wind power and load forecast results in the next day or several hours are given, and the system uses conventional unit output (including start-up and stop) as adjustment means and meets certain operating constraints. How much wind power can be accommodated. If the wind power prediction error is not considered, the actual value of wind power may be greater or smaller than the planned value of wind power at a certain point in the future, which will lead to the phenomenon of wind curtailment and load shedding, which is not conducive to the consumption of clean energy and will also It will have adverse effects on the ...

AI Technical Summary

Problems solved by technology

This approach essentially ignores the differences in the prediction accuracy of the same forecasting method under different weather conditions. For example, the forecast results may be more accurate in the smooth and sunny weather conditions, but may be more accurate in the windy and windy weather conditions. poor

Therefore, mixing all forecast errors together in statistical modeling will increase the difficulty of coordinating "accuracy" and "analytical calculation" on the one hand, and on the other hand may make the results of stochastic economic dispatch conservative

Images