双馈发电机空载并网的神经网络积分滑模控制  

NEURAL NETWORK BASED INTEGRAL SLIDING MODE CONTROL OF DOUBLY-FED INDUCTION GENERATOR FOR NO-LOAD GRID CONNECTION

作　　者：刁亮[1] 王丹[1] 郭磊[1] 彭周华[1] 

机构地区：[1]大连海事大学轮机工程学院,大连116026

出　　处：《太阳能学报》2017年第12期3385-3391,共7页Acta Energiae Solaris Sinica

基　　金：国家自然科学基金(61273137;51209026;51579023);辽宁省教育厅科学研究基金(L2013202);中央高校基本科研业务费专项基金(3132015021;3132014321);中国博士后科学基金(2015M570247)

摘　　要：通过分析双馈风力发电空载并网的运行机理,提出一种基于神经网络的积分滑模控制策略。利用神经网络对系统中的不确定项进行估计并加以补偿,进而降低滑模控制器对切换项增益的需求,减小了控制量的抖振。进一步采用自适应方法调整切换项增益,避免对神经网络逼近误差上界的先验要求。仿真和实验结果表明,所提控制方法对转速波动具有较强的鲁棒性,并网过渡过程短暂,定子电流对电网冲击小,且并网后无需更改控制器参数即可实现双馈发电机的功率解耦控制。Through analyzing the operational mechanism of doubly-fed wind power generator for no-load grid connection, an integral sliding mode control strategy was proposed based on neural network. The neural network was used to estimate and compensate the system uncertainty, which reduce the gain requirement of the switching function in the sliding mode controller. The chattering phenomenon could be minimized significantly. Moreover, the adaptive method was used to adjust the switching gain, and the priori knowledge on the upper bound of the approximation error of the neural network was avoided. The simulation and experiment results show that the proposed control scheme behaves strong robustness against the rotor speed fluctuation, the transition process of grid connection is short, and the stator currents make only a minor strike to the grid. After the grid connection, the power decoupling control of the doubly-fed induction generator can be realized without changing the controller parameters.

关 键 词：双馈风力发电机 空载并网 神经网络 积分滑模控制 

分 类 号：TM315[电气工程—电机]