Wind pattern recognition in neural fuzzy wind turbine control system

This paper introduces a new approach utilizing a fuzzy classifier and a modular temporal neural network to predict wind speed and direction for advanced wind turbine control systems. The fuzzy classifier estimates wind patterns and then assigns weights accordingly to each module of the temporal neural network. The finite-duration impulse response multiple-layer structure of the temporal network makes it possible that a static network represents the underlying dynamics of physical phenomena. Using previous wind measurements and information given by the classifier, the modular network trained by a standard backpropagation algorithm predicts wind speed and direction effectively. Meanwhile, the feedback from the network helps auto-tuning the classifier. In general, the principle of this is applicable to nonstationary time series modeling and prediction problems