A Probabilistic Forecasting Method of Ultra-short-term Wind Power

Abstract

The invention discloses an ultra-short-term wind power probability prediction method, which collects historical data of wind farms to obtain a training sample set; generates input variables according to the historical data of influencing factors to obtain a sample set; uses wavelet analysis to perform wavelet decomposition and analysis of the sample set The wavelet coefficients are reconstructed to obtain wavelet sample sets; the extreme learning machine is used to train model parameters for each wavelet sample set to obtain the wavelet extreme learning machine prediction model, and the test set is brought into the network to obtain the wavelet ultra-short-term point prediction value. The extreme learning machine model training error and point prediction value of each wavelet are stored, added and averaged to obtain the real error and point prediction value of the model after wavelet decomposition, and the Gaussian distribution parameter estimation is performed on the real error of the model to obtain the wavelet model The Gaussian distribution function of the training error can be used to obtain the ultra-short-term probability prediction interval of the wavelet according to the confidence requirement and the predicted value of the point.

Description

technical field [0001] The invention relates to an ultra-short-term wind power probability prediction method based on wavelet analysis and extreme learning machine, which performs probability interval prediction on wind power and belongs to the technical field of new energy consumption. Background technique [0002] At present, the global economy is developing rapidly, and the energy structure is developing in the direction of low-carbon and clean new energy. As an important part of new energy, wind energy has always been valued by the world. With the continuous development of wind power technology in recent years, the installed capacity of wind power in my country has continued to increase, and wind power has gradually become the third largest source of electricity after hydropower and thermal power. [0003] Many problems have also been highlighted in the process of wind energy utilization. For example, the wind has strong randomness and instability, which makes the electr...

AI Technical Summary

Problems solved by technology

However, in specific operations, if the given data is relatively active, making it difficult for the prediction model to adapt, it may cause the prediction accuracy of the model to decrease

Images