The use of wavelet theory and ARMA model in wind speed prediction

Wavelet theory is first used to pick up the low frequency parts through the decomposition of the whole wind speed, then ARMA model is used to forecast the wind speed on the gentled data to improve the prediction precision.

Abstract

In order to reduce the influence of wind power to power grid, and to reduce the rotating spare capacity and operation cost of power supply system, it is necessary to predict the wind speed. Because the wind speed has very good succession and randomness, it is quite appropriate to use Auto Regressive Moving Average (ARMA) model of times series to predict the wind speed. In order to improve the prediction precision further, this paper first use wavelet theory to pick up the low frequency parts through the decomposition of the whole wind speed, then use ARMA model to forecast the wind speed on the gentled data. This paper take the wind speed directly measured from a certain wind farm as an example. Practical example shows that: This combination model can effectively improve the wind speed prediction accuracy. It has certain practical value.