Wind power prediction method and system and computer readable medium

Abstract

The invention relates to a wind power prediction method and system and a computer readable medium. The prediction method specifically comprises the following steps: S1, collecting wind power sample data, and dividing the wind power sample data into a training set and a test set; S2, calculating a clustering center of a training set through fuzzy C-means clustering, calculating Euclidean distances from sample points in a test set to the clustering center, and matching corresponding categories; S3, performing decomposition and denoising on the clustered wind power sample data through a variational mode decomposition algorithm, and inputting the reconstructed wind power sample data into a plurality of extreme learning machine prediction models for training; and S4, selecting and optimizing kernel functions in the extreme learning machine prediction models through a particle swarm algorithm to obtain an optimized extreme learning machine prediction model, obtaining wind power real-time data, inputting the wind power real-time data into the optimized extreme learning machine prediction model, and outputting wind power prediction power. Compared with the prior art, the method has the advantages that the accuracy of the prediction result of the output of the wind power plant is improved, and controllable and schedulable wind power generation is realized.

Description

technical field [0001] The invention relates to the technical field of wind power measurement, in particular to a wind power prediction method, system and computer-readable medium based on data mining and particle swarm algorithm optimization support vector machine. Background technique [0002] Wind power prediction methods can be divided into physical methods and statistical learning methods. The physical method is based on surface information and meteorological data, combined with aerodynamic and thermodynamic equations to solve the wind speed and direction at the hub of the fan, and then through the wind speed of the wind turbine. - Power curves to get predicted results. The statistical learning method is to analyze and learn the historical power and meteorological data of the wind farm, obtain the corresponding relationship between the power output and the input of the meteorological conditions, establish an input-output model, and predict the wind power for a period of...

AI Technical Summary

Problems solved by technology

[0003] However, the current statistical learning prediction method relies on a large amount of historical output data. The current method does not consider analyzing the original data information, fully mining the characteristics of the data, and establishing an accurate prediction model.

Moreover, the parameters in the support vector machine, especially the selection of penalty parameters and kernel function parameters, have a great impact on the prediction performance of the model. The selection of these parameters lacks theoretical guidance. The traditional parameter optimization method takes a long time and the effect is not ideal.

Images