Risk prediction evaluation method for wind power grid integration

Abstract

The invention discloses a risk prediction evaluation method for wind power grid integration, comprising the following steps of: collecting original data of a wind power plant to obtain an initial wind power time sequence; performing phase space reconstruction on the initial wind power time sequence to obtain a phase point vector; performing K mean clustering computation on the phase point vector; introducing a weight vector to improve criterion for distance and trend of proximal points, and thereby obtaining a proximal point set; using the proximal point set as a training set of volterra self-adaptive filter to obtain wind power plant power prediction data. The risk prediction evaluation method for wind power grid integration provided by the invention eliminates the technical defect in the prior art which ignores different influence of the time sequence of own different coordinate components of the phase point on the prediction point and is prone to introducing false proximal points, effectively avoids the false proximal points, chooses the proximal point with distance and evolutionary trend both similar to the prediction point, and is thereby improved in computation accuracy and speed.

Description

technical field [0001] The invention relates to the field of electric power, in particular to a method for predicting and evaluating risks of wind power connected to a power grid. Background technique [0002] With the continuous development of wind power technology and the continuous increase of installed capacity of wind farms, the intermittency and volatility of wind power make it a disturbance source when connected to the grid, which has a great impact on the safe operation level of the power grid. Therefore, it is of great significance and value to accurately predict short-term wind power power and use the prediction results to evaluate the dynamic security of the grid after wind power access to improve the safe operation level of the grid after wind power access. [0003] In recent years, in the short-term prediction of wind power, the Volterra adaptive filter has attracted the attention of many scholars due to its fast training speed and small sample size, such as "Vo...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing technology, the Volterra adaptive filter adjacent point calculation used in short-term wind power forecasting ignores that the time order of different coordinate components of the phase point itself has different effects on the predicted point, and it is easy to introduce "false adjacent points". ”, leading to poor calculation accuracy and calculation speed stability of the algorithm

Images