سال انتشار: ۱۳۸۱

محل انتشار: هفدهمین کنفرانس بین المللی برق

تعداد صفحات: ۹

نویسنده(ها):

Mahdi Banejad – Signal Processing Research Centre Queens land University of Queens land University of Technology
Gerard Ledwich – Queens land University of Queens land University of Technology

چکیده:

Due to operating power systems near their stability limits, there is more frequent occurrence of problem of low frequency oscillations. Some changes in power systems like line outage, customer load variation can initiate low frequency oscillations. Local oscillations are oscillations associated with a single generator. The oscillations are well recognised and have frequencies in the range of 0.7 to 2Hz. Inter-area oscillations are associated with groups of generators and have the frequencies in the range of 0.1 to 0.8Hz [1-3].
Load dynamics can have a strong effect on the interconnected power system and the appropriate modelling of the load is important for extending transmission limits. Methods are being developed to model electric loads [4-8].
Therefore, identification of the load dynamics is very important in mode shape determination of power system. Following the identification of the load, the quantification of load contribution in damping is essential in low frequency oscillation studies. This paper is aim at specifically identifying the load contribution in damping.
This paper firstly develops a theoretical background of identification* of modal frequencies. The method used in this part is based on cross-c or relation (autocorrelation). It is found that due to averaging property of cross-correlation (autocorrelation) and being low noise sensitive, the correlation based method is more effective than the other conventional method such as Prony’s method [9]. After that the load of the models is to be found on the basis of correlation. The load modelling identification method is not dependent on the nature of the load. In the final part of the proposed method the participation of the load at each resonant frequency is determined. The developed algorithm is simulated on a test system to find out what is the contribution of a synthetic load of the damping. Then for a substation of Australian network the quantification of load contribution in damping is determined.