|
The first part of paper simply introduce wind farm wind speed forecasting mode, and introduce time series, neural network, support vector machine, least squares support vector machines into wind speed forecasting model. In order to improve the reliability of SVM and LSSVM parameter choice, the paper use genetic algorithm and Ant Colony Optimization to choice the model parameter. Also, compare the effect of different forecasting model, this methods provide theory support for the more long-term wind speed forecasting, wind power forecasting and wind farm site select, it has great practical value. The latter part introduces wind turbine dynamic analysis, and use ADAMS software to analyze wind turbine, analyze wind turbine dynamic response on the condition of without pitch control, normal operation and pitch control failure.
|