Gaussian mixture model-based neural network for short-term wind power forecast
Gary W. Chang, Heng-Jiu Lu, Ping-Kui Wang, Yung-Ruei Chang and Yee-Der Lee
Version of Record online: 23 JAN 2017 | DOI: 10.1002/etep.2320
This paper proposes a Gaussian mixture model-based neural network model forecast the short-term wind power generation. The contribution of the paper lies in presenting an enhanced radial basis function neural network model with better convergence property for forecasting application. The Gaussian mixture model is adopted to substitute the traditional single Gaussian basis function in each neuron in the hidden layer and is with better initial settings of parameter values. Actual measured wind power output data are adopted to implement the proposed model. Test cases of wind power obtained by autoregressive integrated moving average, radial basis function neural network, support vector regression, and the proposed method are then under comparisons. Results show that the proposed method is superior to the compared methods in short-term forecasting accuracy while the computational efficiency is maintained.