Based On Rough Set Theory Network Fault Diagnosis Expert System

As the enlarging of power system scale and increasing of power grid automatization level, more and more comprehensive alarm information is transmitted to control centers of all grades. Power system response is being complication while there is fault in the grid, especially complex fault,behavioral abnormality of automechanism,circuit breakers or protective relays and uncertainties such as losing of information because of channel disturbance, which will make trouble for power system fault diagnosis. That will lead up to serious results such as fault expanding if the fault was misjudged. Hence it is quite necessary to research and develop a fault diagnosis system with high robust which could accommodate all sorts of disadvantages, in order that dispatchers could identify the fault rapidly and that the power system could operate safe and stably. Rough Sets theory (RS) is a newly developed mathematic tool to describe information with vagueness and uncertainties. RS has been received much attention of researchers around the world due to its application potentials in data mining and dealing. Power system fault diagnosis could be described as a mode-classifying problem, and the decision table method of rough sets theory is quite fit to it. First of all, this paper introduces the study and development of power system fault diagnosis at home and abroad, then it puts emphasis on RS in power system fault diagnosis. Considering the fault information being both localized and connected, a papilionaceous partitioning method based on RS is proposed to partition large-scale power grid. And it presents an efficient way for RS in fault diagnosis of large-scale power grid. At last, the proposed method is introduced into fault diagnosis system, and has been tested in the IEEE 9-bus systems. Computer simulation results show that the proposed partitioning method is effective for fault diagnosis in large-scale power grid.