Power System Under The Condition Of Oscillation Fault Recognition Based On Prony Algorithm

Coupled with a steady growth makes the reliability of the power grid to reducesystem size. Mutual connection between power grid, and the application of a largenumber of nonlinear components make the grid system is now gradually evolved intoa complex, large scale nonlinear system. But general analysis methods need to build adetailed mathematical model of power grid. The analytical process need to solve ahost of matrix eigenvalues. But, because the actual power system constitute a complexand lack of accurate data, this leads to matrix eigenvalue calculation analysis systemshows more and more limitations. In the real application system analysis, calculationmethod of the eigenvalues is usually dimension disaster may occur.In order to ensure the reliable operation of power grid and increase transmissioncapacity of transmission line, need the protection device can be more stable, rapidremoval of the fault. Low frequency oscillation is the stability study of grid system isan important constituent elements. Areas of interconnected power system, the largecapacity generator group leader distance conveying electricity formed the systemweak damping or underdamping, enables the low frequency oscillation often occurs.When power grid or within the low-frequency oscillation in a short period of timedidn’t get effective control and identification, grid may caused by system oscillationgenerator out of sync, short-circuit fault protection device can not effectively removedand the protection misoperation caused widespread power outages.This paper choose IEEE standard test systems simulation system low frequencyoscillation and the oscillation occurred case line three-phase short circuit fault. Firstof all, according to the basic theory of generator model in power system lowfrequency oscillation when the state equation is analyzed. Next to the failure ofdynamic electric arc theoretical analysis, established the fault arc dynamic model ofthe primary and secondary, and discusses how it can be connected to the IEEE standard test systems, improve the precision of the short circuit fault simulation.Using matrix original Prony algorithm improved beam improve computationalefficiency and accuracy, proposed improved algorithm combined with multiple signalanalysis method applied in the low frequency oscillation mode identification, themethod based on normalized resolve multiple signals appeared in the process ofsuperposition of annihilation. The improved algorithm can identify the systemoscillation is not easy to identify three phase short circuit fault, in the process ofoscillation for the relay protection device closure algorithm lays a theoreticalfoundation.The research achievements of this paper is expected to be in the future onlinemonitoring and identification of power system low frequency oscillation mode andanalysis of system oscillation occurred when the three-phase short circuit faultidentification to provide theoretical support and method support.