关键词: 非线性智能模型预测控制, 最大风能捕获, 双馈感应发电机（DFIG）, 转速控制, 粒子群优化, 支持向量机（SVM）

Abstract: Based on the principle of maximum wind energy capture，the maximum wind energy tracing below the rated wind speed can be realized by controlling the speed of Double-Fed Induction Generator（DFIG） to track the optimal speed.The variable speed constant frequency double-fed wind power system is taken as the research object and the control problem of maximum wind energy tracing below the rated wind speed is studied.Firstly，according to the characteristics of strong coupling，strong non-linearity and the complexity of mechanism model for DFIG，the intelligent predictive model is built by using Support Vector Machine（SVM）.Secondly，the predictive output is revised by feedback correction and the closed control loop is structured.Finally，due to the advantages that Particle Swarm Optimization（PSO） algorithm has the fewer regulated parameters，the small evolution groups and the quick calculation speed，the optimal control sequences are obtained easily and the “bottleneck problem” in the rolling optimization calculation is solved better.The simulation results validate that the adopted predictive model has better anti-disturbance and generalization abilities，and the predictive control algorithm can realize the control objects.

Key words: non-linear intelligent model predictive control, maximal wind energy capture, Double-Fed Induction Generator（DFIG）, speed control, particle swarm optimization, Support Vector Machine（SVM）