Forecasting the wind direction by using time series models with long-term memory (case study: Nayer region)

Abstract

Purpose

Today, because of the increasing need for the energy resources and the reduction of fossil fuels, renewable energy, especially wind energy, has attracted special attention. The precise forecasting of such energy will be the main factor in designing and investing in this field. On the other hand, the wind energy forecast provides the possibility of optimal use of available resources. In addition, the produce maximum energy would be possible by identifying wind direction and putting wind turbines in the best position.

Design/methodology/approach

Time series forecasting methods with long-term memory in this research have been used.

Findings

Eventually, the autoregressive fractionally integrated moving average (3,0,0)-FIGARCH (1,0,1) long-term memory model has more acceptable performance. The obtained error is based on the RMSE (0.2889) and the TIC (0.2605) values.

Practical implications

In this paper, the forecast wind direction belongs to Ardebil province and Nayer city in Iran.

Originality/value

The speed and direction of wind are variables that constantly change; hence, it will be difficult to predict the exact wind energy. In recent years, some studies have been conducted on wind speed forecasting, whereas wind direction forecasting has been done in a fewer number of studies. Most studies are related to low-lying areas. As the height of the wind turbine is directly related to the energy generation, 78 m height has been considered in this study.

Citation

Soleimani, P., Emami, B., Rafei, M. and Shahrasbi, H. (2021), "Forecasting the wind direction by using time series models with long-term memory (case study: Nayer region)", International Journal of Energy Sector Management, Vol. 15 No. 2, pp. 385-396. https://doi.org/10.1108/IJESM-04-2019-0002