基于EEMD-NExT的低频振荡主导模式工况在线辨识与预警  

Online dominant mode identification and warning based on EEMD-NExT for low-frequency oscillation in operating conditions

作　　者：汪颂军 刘涤尘[1] 廖清芬[1] 周雨田[1] 王亚俊[1] 王乙斐[1] 赵一婕[1] 

机构地区：[1]武汉大学电气工程学院,湖北武汉430072

出　　处：《电力自动化设备》2014年第12期111-116,122,共7页Electric Power Automation Equipment

基　　金：国家高技术研究发展计划(863计划)项目(2011AA-05A119);国家电网公司大电网重大专项资助项目课题(SGCC-MPLG029-2012)~~

摘　　要：结合集合经验模式分解(EEMD)和自然激励技术(NEx T),基于广域测量系统(WAMS)的动态量测信息,提出低频振荡主导模式识别方法。该方法借助EEMD处理非平稳信号,利用EEMD时空滤波器、互相关系数和信号能量权重筛选出主导模式分量;通过NEx T求互相关函数,并利用Teager能量算子识别时变幅值和频率,采用信号能量分析法辨识阻尼比并应用于预警系统。算例仿真结果表明,所提方法能够实时准确地辨识出系统的主导模式信息,且无需人工激励并剔除虚假模式,同时具有较强的抗噪性能。A method of online low-frequency oscillation dominant mode identification based on the dynamic measurements of WAMS（Wide-Area Measurement System） in operating conditions is proposed,which combines EEMD（Ensemble Empirical Mode Decomposition） with NExT（Natural Excitation Technique）. The EEMD is used to deal with the unstable signal and select the dominant mode with its spatiotemporal filter,cross- correlation coefficients and signal energy weights. The NExT is used to obtain the cross-correlation function. The time-varying amplitude and frequency are identified by the Teager energy operator and the damping ratio is identified by the signal energy analysis,which are applied to the early warning system. Simulative results of case study show that,without artificial incentive and with strong anti-noise ability,the system dominant mode is identified and the illusive mode is eliminated accurately in realtime.

关 键 词：集合经验模式分解 自然激励技术 工况模式分析 低频振荡 主导模式识别 稳定性 

分 类 号：TM712[电气工程—电力系统及自动化]