A wind speed prediction method based on deep extreme learning machine and its system and unit

Abstract

The invention discloses a wind speed prediction method based on a deep extreme learning machine, a system and a unit thereof, belonging to the field of wind turbines. The wind speed value is predicted by the time series prediction method of the deep extreme learning machine, including: obtaining a set of historical observation data sequences at time t and before the wind turbine power generation process; extracting the training data through the time series prediction method of the deep extreme learning machine Sample data set; derivation of prediction model: extract the nearest neighbor to the prediction sequence Q from the training sample data set as a recombination sample, and the recombination sample corresponds to single-step prediction and multi-step prediction. There are single-step features and multi-step features; respectively; Perform deep extreme learning machine training on the multi-hidden-layer DELM models of single-step features and multi-step features, and integrate them into a prediction model through local selection; predict the wind speed value X through the prediction model t+s . The method of the invention has high accuracy and generalization performance, and can improve the prediction performance and real-time update ability.

Description

technical field [0001] The invention relates to the field of wind turbines, in particular to a wind speed prediction method based on a deep extreme learning machine, a system and a wind turbine. Background technique [0002] As an important part of the national sustainable development strategy, wind power is the conversion of air kinetic energy into electrical energy. The random fluctuation and intermittency of wind determine the fluctuation and intermittency of wind power. With the continuous expansion of wind power generation, the impact of wind farm grid connection on the power grid system will become more and more obvious. Large wind speed disturbance will cause great changes in the voltage and frequency of the system. At this time, wind farm grid connection and grid connection The stability and security issues in the future have become new issues that need to be solved urgently. Therefore, accurate prediction of wind speed is beneficial to the operation of wind turbine...

AI Technical Summary

Problems solved by technology

It can effectively improve the shortcomings of traditional neural networks such as low learning efficiency, easy to fall into local optimum, and the number of network layers restricts the training effect.

Images