Wind speed prediction and determination of wind power output with multi-area weather data by deterministic chaos

Wind power generation increases rapidly and the availability of wind energy depends on the wind speed, which is a random variable. For the wind-farm operator, this poses difficult in the system scheduling and energy dispatching, as the schedule of the wind-power availability is not known in advance. In this research, we propose an intelligent technique for forecasting wind speed and power output of wind turbine from several hours up to 24 hours ahead. This technique is based on deterministic chaos. The data was offered by the Japan Meteorological Agency. These data include the “multi-story meteorological data” and the “ground meteorological observation data” of Aomori area where is located in the north of Honshu, Japan. The Jacobian Matrix Estimation Method based on deterministic chaos is then supplied with the data to establish the relationship time-series data. In this research, the authors tried to employ the weather data of multiple areas around the targeted site of introduced wind turbine. This method is believed to bring more information for predicting the wind speed and in result have better determination of wind generation output. The model based on the deterministic chaos with such multi-area weather data demonstrated a good agreement and produced the wind forecast with the accuracy of 90% and above.