抄録
A real time FFT based on-line identification has been presented for the real time monitoring of power system oscillation modes. The monitoring system is set up by using a personal computer with a DSP board with AD conversion interfaces. For the monitoring, the real power flow signal is utilized. The real power signal is sampled through AD conversion interfaces and stored in the memory. The monitored signal is also passed to the workspace in the Matlab/Simulink environment for the FFT analysis and also the visualization of the obtained results by using a color mapping to show the variation of oscillations in real time. Furthermore, a fuzzy reasoning based filtering has been proposed to washout individual oscillation modes included in the measured real power signals. The efficiency of the proposed method has been demonstrated through the testing using a test signal, monitoring signals on the analog power system simulator at the Research Laboratory of the Kyushu Electric Power Co. In addition, the proposed system has been tested by using real power flow signals measured on actual 500 kV trunk lines.
