基于GD-FNN的特高压直流输电暂态稳定控制  

Transient stability control of UHVDC based on generalized dynamic fuzzy neural network

作　　者：蒋平[1] 严栋[1] 刘盛松 胡伟 

机构地区：[1]东南大学电气工程学院,江苏南京210096 [2]江苏电力调度控制中心,江苏南京210024

出　　处：《电力系统保护与控制》2013年第10期1-6,共6页Power System Protection and Control

基　　金：江苏省电力公司2012年科技项目(GC201205010002024)

摘　　要：特高压直流输电输电容量大、控制灵活迅速,在我国"西电东送"战略中扮演了重要角色,其对受端交流系统暂态稳定性的影响也随输电容量的增大而变大。提出采用广义动态模糊神经网络(GD-FNN)控制来提高系统暂态稳定性,通过附加控制信号动态调节输送功率从而给系统提供足够阻尼。根据系统选择合适的调制信号以及控制维数,对GD-FNN系统训练后通过系统误差和模糊规则的ε-完备性作为判据来优化系统结构,同时对隶属度函数的参数进行修正,从而保证了控制器具有紧凑的结构和良好的泛化能力。仿真结果表明,所设计的暂态稳定控制器在保持系统稳定方面具有优越的性能,并且鲁棒性较好,可有效保障机组和电网的安全稳定运行。UHVDC plays an important role in Project for Delivering Electricity from the West to the East, for its flexible and quick control and large electrical capacity. Its influence on AC system is becoming larger as transferring more DC power. A method using generalized dynamic fuzzy neural network in UHVDC to increase transient stability is presented, which regulates power to supply enough damping with additional control signal. Suitable modulation signal and dimension are chosen. After GD-FNN system is trained, systematic error and ε-completeness of fuzzy rules are used as criterion for optimizing system structure. Meanwhile, degree of membership is modified to ensure that the system has a compact structure and a great generalization ability. Simulation shows that the controller designed can increase the generator＇s electrical damping remarkably in different conditions and suppress the oscillation effectively, so it has a certain robustness.

关 键 词：特高压直流输电 广义动态模糊神经网络 暂态稳定 PSCAD EMTDC 

分 类 号：TM721.1[电气工程—电力系统及自动化]