Noise auxiliary signal decomposition method and Elman nerve network wind power combined prediction method

Abstract

A noise auxiliary signal decomposition method based on complex data experience modal decomposition and Elman nerve network combined prediction method can realize short period prediction of the wind power, and belongs to the wind power prediction technical field; the combined prediction method comprises the following steps: 1, doping white noises into an original signal sequence; 2-7: solving and obtaining an IMF component and balance; 8, using the obtained IMF component and the balance to built an Elman nerve network prediction model, using the model to predict, and finally carrying out total superposition so as to obtain the final prediction result. The noise auxiliary signal decomposition method and Elman nerve network wind power combined prediction method can further reduce the modal aliasing problems existing in a conventional EEMD decomposition method, thus improving the short period wind power prediction precision.

Description

technical field [0001] The present invention proposes a combined prediction method based on complex data empirical mode decomposition noise-assisted signal decomposition method and Elman neural network for short-term prediction of wind power, which belongs to the technical field of wind power prediction. Background technique [0002] Due to the randomness of wind speed, wind power has nonlinear characteristics, so the Elman neural network method that can deal with nonlinear problems has certain advantages. However, existing studies have found that the prediction accuracy of wind power using a single prediction method needs to be improved, because the prediction accuracy of a single prediction method is easily disturbed by high-frequency components in the data, resulting in a decrease in prediction accuracy. In order to solve the interference of wind power non-stationarity on the prediction results, some scholars put forward the idea of ​​combined prediction. Mainly on the b...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the deficiencies in the existing combination prediction technology, the present invention provides a wind power combination prediction method based on the noise-assisted signal decomposition method and the Elman neural network. This algorithm can further reduce the modal aliasing problem existing in the old EEMD decomposition method , to improve the accuracy of short-term wind power forecasting

Images