An Integrated Model for Wind Power Forecasting Based On Maximum Entropy Principle

Abstract

Wind speed forecasting is important to both wind plant and electrical system’s operation. Improving the prediction accuracy can not only effectively eliminate the disadvantageous impact to the power system but also enhance the wind’s permeability and competitiveness against the other kinds of power in electricity market. The random and intermittent wind directly causes the same characteristics of the wind power. The maximum entropy principle information principle was introduced to construct an integrated forecasting model. In this model, all individual forecasting models’ results and historical forecasting error were regarded as constraints. The forecasting values were educed by the integrated model based on the maximum entropy principle. Application result in a real wind plant showed that the integrated model had a higher forecasting accuracy compared with individual models and combined models.