摘要
大量新能源发电和电力电子设备引起的次同步振荡(SSO)严重影响了电力系统稳定性。基于广域测量系统(WAMS)和相量测量单元(PMU)提供的基波同步相量辨识次/超同步振荡参数具有广域同步和高刷新率的核心优势,可实现次/超同步振荡的动态在线监测。该文从基波同步相量在次/超同步振荡下的复数域频谱特性出发,综述并分析了目前已有次/超同步振荡参数辨识方法的参数辨识特性及关键问题。首先,根据同步相量频谱分析了频率辨识范围、各分量叠加耦合特性及相位翻转特性;其次,分析了参数辨识的关键问题在于振荡分量耦合关系、基波分量耦合关系的解耦问题及频谱泄漏问题,并围绕复数域频谱分析方法分析了在解决以上问题时的三个关键难点:针对次/超同步分量的耦合关系有效地辨识超同步分量的问题、不同同步相量采样频率下进行参数辨识的适应性问题,以及如何处理频谱泄漏从而缩短数据窗长的难题;对此,综述了相应解决方法并展望了未来可能的技术突破点;最后,通过仿真数据和实际PMU数据对比分析验证了该文观点。
Subsynchronous oscillations(SSO)caused by considerable renewable generation and power electronic equipment seriously affect the stability of power systems.The parameter identification of sub/supersynchronous oscillations based on fundamental synchrophasors,which are provided by the wide-area measurement system(WAMS)and phasor measurement units(PMU),has the core advantages of wide-area synchronization and high refresh rate.These advantages will enhance the dynamic monitoring of sub/supersynchronous oscillations significantly.Based on the complex spectrum characteristics of the fundamental synchrophasors under sub/supersynchronous oscillations,this paper summarizes and analyzes the parameter identification characteristics and essential issues of the existing parameter identification techniques for sub/supersynchronous oscillations.First,the identified frequency range,the coupling characteristics of each component,and the phase reversal characteristics are analyzed according to the spectrum of synchrophasors.Then,the essential issues of the parameter identification are analyzed,including the spectrum leakage problem,the decoupling problems of the coupled relationship of the oscillational component and the coupled relationship of the fundamental component,respectively.Focusing on the spectrum-based techniques,three essential difficulties in solving the above problems are concluded and analyzed:the problem of effectively identifying the supersynchronous component considering the coupled relationship between the sub/supersynchronous components,the adaptability of parameter identification under different synchrophasor refresh frequencies,and the issue of dealing with spectrum leakages to shorten the data window.Thus,this paper summarizes the corresponding solutions and looks forward to the possible technological breakthroughs in the future.The results of simulation data and actual PMU data in different cases show that in the case of a single subsynchronous oscillation component,a couple of sub/supersynchronous oscillation components,and multiple couples of sub/supersynchronous oscillation components,spectrum-based techniques can effectively identify parameters using a sampling frequency of 100 Hz or under a given applicable range using a sampling frequency of 50 Hz.The following conclusions can be drawn:(1)The sub/super-synchronous oscillation parameter identification with synchrophasors is affected by the synchrophasor sampling frequency(i.e.,the reporting rate),and the identifiable frequency range conforms to the sampling theorem with complex numbers.Specifically,using synchrophasors with a sampling frequency of 100 Hz can identify the parameters of all sub/supersynchronous components without additional information;When the sampling frequency of synchrophasors is 50 Hz,additional conditions are needed to distinguish the wrong results to identify the sub/supersynchronous oscillation parameters.(2)The most essential problem of sub/supersynchronous oscillation parameter identification based on synchrophasors is how to use a shorter synchrophasor sequence to achieve high-precision parameter identification,which may also be one of the breakthroughs in future parameter identification techniques.(3)The influence of the frequency coupled relationship between the sub/supersynchronous oscillations on the parameter identification needs to be considered.The coupled relationship between the sub/supersynchronous components is the coupling of the positive/negative frequencies of the oscillation rather than the mutual spectrum leakage,which is completely different from the influence between the fundamental component and the sub/supersynchronous component.(4)The advantages of the spectrum-based techniques are simple with low computational complexity.The spectrum results are especially suitable for the case of multiple pairs of oscillational components.However,the DFT-based methods need to approximate the spectrum leakage,which makes it difficult to further shorten the data window even if a better window function is used,and the non-DFT-based methods may make a breakthrough in this problem.(5)The problem of data loss and bad data in the synchrophasor sequence may exist in practical applications and needs to be considered.The existing methods rely on the continuity of the synchronous phasor sequence without an in-depth study of the above problems,and it will be another breakthrough point in the practical application of the sub/supersynchronous oscillation parameter identification method based on the synchronous phasor.
作者
张放
李佳欣
史静舒
Zhang Fang;Li Jiaxin;Shi Jingshu(School of Electrical Engineering,Beijing Jiaotong University,Beijing100044,China)
出处
《电工技术学报》
EI
CSCD
北大核心
2024年第19期6018-6038,6053,共22页
Transactions of China Electrotechnical Society
基金
国家自然科学基金资助项目(52077004)。
关键词
同步相量
次/超同步振荡
参数辨识
广域测量系统
频谱分析
Synchrophasor
sub/supersynchronous oscillation
parameter identification
wide-area measurement system(WAMS)
spectral analysis
