Prediction of wind power generation and power ramp rate with time series analysis

Abstract

The use of fossil fuel in the world has been increasing and it generates lots of greenhouse gases. As a result, environmental pollution brought us a serious weather change. In order to reduce the environmental pollution, we should use renewable energy that does not produce any pollution such as wind data. However, wind data can change much in a short time, which is called ramp event. It can make the demand and response imbalance and also cause damages to the wind turbines. Therefore, we should predict the power generation and power ramp rate (PRR) to avoid these problems. In this paper, we predicted the wind power generation and PRR with exponential smoothing method and ARIMA. The prediction method predict wind power generation and PRR after 1 minute using data measured 1 hour ago at 10 intervals. We got forecasting error rate such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), and then we compared two results of ARIMA and exponential smoothing method. The comparison results showed that exponential smoothing method gets better prediction accuracy than ARIMA.