Wind power plant cluster active power prediction method and system

Abstract

The invention provides a wind power plant cluster active power prediction method and system. The prediction method comprises: acquiring historical data of active power of a wind power plant cluster; based on an error feed-forward prediction method and a continuity method, establishing an error feed-forward prediction model reflecting the wind power change trend of the wind power plant cluster; training the error feed-forward prediction model based on historical data, determining an optimal error limit value in the error feed-forward prediction model, and obtaining a trained error feed-forward prediction model; and inputting the trained error feed-forward prediction model based on historical data of the first n moments of a to-be-predicted moment to obtain an active power prediction value of the to-be-predicted moment. The influence of recent and historical wind power plant and wind power cluster power data and the regularity shown by cluster data are fully considered, the prediction precision is improved by combining the feedforward prediction method and the continuation method, the prediction algorithm is simplified, the calculation amount is reduced, and ultra-short-term prediction of the wind power plant cluster is achieved.

Description

technical field [0001] The invention relates to the field of wind power cluster prediction analysis, in particular to a wind farm cluster active power prediction method and system. Background technique [0002] With the rapid development of the economy, people are facing enormous pressure on energy and environment at the same time. As a renewable energy, wind energy has the characteristics of safety, cleanness and wide distribution, and has attracted more and more attention. Large-scale centralized wind power grid-connected power generation system is one of the development trends of wind power generation. Most of them are distributed in remote areas with relatively rich and stable wind energy resources, far away from the load center, and connected to the system through high-voltage, long-distance transmission lines. [0003] Wind energy is different from other renewable energy sources such as solar energy and hydropower, and has its unique natural characteristics. The powe...

AI Technical Summary

Problems solved by technology

For the Northeast region with large-scale wind power connected to the grid, if the prediction accuracy is not enough, it will lead to the failure to send out the electricity that could have been generated, that is, the phenomenon of wind abandonment, or the phenomenon that the grid requires wind farms to output power but cannot generate the corresponding power; if the prediction The model is relatively complex, and there will be a problem of long calculation time. For the ultra-short-term forecast information with a time scale of 5 minutes or 15 minutes, it cannot be uploaded in time, which leads to the inability to connect the forecast information with dispatching instructions, resulting in wind abandonment or insufficient output.

[0005] For large-scale wind power clusters, the investment in wind farm construction is large, but the conversion efficiency of wind energy is not high

Images