Ultra-short-term wind power prediction method with adaptive time resolution

Abstract

The invention discloses an ultra-short-term wind power prediction method with adaptive time resolution. The ultra-short-term wind power prediction method comprises the following steps: calculating a hidden prediction error in a wind power prediction period, and establishing a corresponding error fluctuation amplitude and a fluctuation rate index to measure the fluctuation characteristic of the wind power prediction error; predicting wind power fluctuation to judge whether the time resolution needs to be adjusted or not; and then performing offline learning on historical data, performing groupdistance grouping on the adjustment coefficient and the fluctuation rate according to the learned numerical relationship between the adjustment coefficient and the fluctuation rate, and then performing adaptive ultra-short-term wind power prediction by adjusting the time resolution online. The ultra-short-term wind power prediction method can effectively reduce the wind power prediction error, especially has a good prediction effect on the minute-level violently fluctuating wind power, opens up a new idea for improving the wind power prediction precision, and facilitates the improvement of thesafety stability and wind power absorption capability of a wind power grid-connected power system.

Description

technical field [0001] The invention relates to the field of wind power, in particular to an ultra-short-term wind power prediction method with self-adaptive time resolution. Background technique [0002] With the increasing shortage of traditional fossil energy and the increasingly prominent problems of environmental pollution, clean energy represented by wind energy has been developed and utilized on a large scale, and its penetration rate in the power system has continued to increase. However, the volatility and randomness of wind power resources bring challenges to the safe and stable operation of the power system. Therefore, it is of great significance to grasp the fluctuation characteristics of wind power and improve its prediction accuracy to enhance the ability of the grid to accommodate wind power and improve the safety and stability of the power system. [0003] According to the time scale, wind power forecasting can be divided into long-term, medium-term, short-t...

AI Technical Summary

Problems solved by technology

It can be seen that during the minute-level violent fluctuations of wind power, the prediction model with constant time resolution (usually 15 minutes) cannot accurately track the minute-level wind power fluctuations within the time step, resulting in large instantaneous forecasts within the time step error

This instantaneous prediction error will bring great uncertainty to the real-time dispatch of the power system with a period of 5 minutes, seriously threaten the safety and stability of the system, and affect the wind power consumption capacity of the system.

Images