摘要
Prony算法是获取电力系统低频振荡参数的有效方法之一,但因算法系统阶数过高,既影响算法计算速度又可能导致模式误辨,进而影响振荡参数辨识的准确性。对此,将能量排序概念引入Prony算法,先介绍了Prony算法的原理及Prony算法参数的选取,而后根据能量大小对振荡模式进行排序,最后确定低频振荡主导模式,辨识低频振荡主导模式参数。算例分析表明,相较传统Prony算法,能量排序Prony算法有效降低了低频振荡模式的误辨率,提高了算法效率。
Prony algorithm is viewed as one of the most effective methods to get parameters of low frequency oscillation in power systems. The high system degree of the algorithm not only affects the computation speed, but also leads to the error of pattern recognition. Furthermore, it has impact on the accuracy of oscillation parameter identification. So, the concept of energy sorting is brought to Prony algorithm in this paper. The principle of Prony algorithm and the parameter selection of Prony algorithm are firstly introduced. According to the energy value, the oscillation mode is sorted. Finally, the dominant mode of low frequency oscillation is determined and the corresponding parameters are identified. Compared with the traditional Prony algorithm, example results show that the energy-sorted Prony algorithm can reduce the error recognition rate of low frequency oscillation and improve the algorithm＇s efficiency.
出处
《水电能源科学》
北大核心
2015年第9期192-195,共4页
Water Resources and Power
基金
天津市科技计划项目(13ZXCXGX35700)
航天技术应用产业产品孵化项目(JSKF201407290015)
海洋公益性行业科研专项经费项目(201405028)
关键词
能量排序
低频振荡
线性预测模型
PRONY算法
参数选取
energy sorting
low frequency oscillation
linear prediction model
Prony algorithm
parameters selection
