Study On Rough Sets Theory Based Distributed Fault Diagnosis System

With the scale of the modern power system being continuously ampliative and fortified, when the fault occurs, a large mount of alarm message will pour into the control center. Moreover, the misjudgement may be caused by the distorted and losed information in the processs of alarm generation and transfer. Therefore, recurring to the strong ability of handling uncertainties possessed by rough set theory, this paper presents an improved attribute reduction algorithm based on discernibility matrix and information quantity, and uses it to handle the reduction progress of the decision table including all kinds of fault cases which is established by considering the signals of protection relays and circuit breakers. Besides, on the basis of the idea "divide and conquer", the paper develops the distributed fault dianosis system using the multi-agent system. The computer simulation results demonstrate that the proposed system is reliable and effective, and thus can be applied to the grid for on-line fault diagnosis.