Study Of Fault Diagnosis In Power System Based On Rough Set Theory

Based on the fact Rough Set theory can effectively handle the imprecise problems without any ancestor information except the date set itself, a decision table including all kinds of fault cases is established by considering the signals of protection relays and circuit breakers. And then, the diagnostic rules are extracted by reducing decision table. Searching for reducing decision table, the method can tell indispensable signals from the dispensable ones and discover the inherent redundancy of signal set. It shows that the method is quite efficient.Power net fault namely analyses the characters of power quantum which include alarm signal coming from protector device,state tuansformation of switch and the value of voltage and current.Based on the logic of protecting act ane the esperience of running workers,the power net can conclude maybe the fault locaton and type.The basic content of power fault commonly consist the fault measure,fault analysis and fault disposal.The means of power net fault diagnosis is disparted of the means of logic disposal and the artificial intelligence embodied the means of espert systemand and the optimizing technique.The paper analysed at present reduction anithmetic with rough sets in common use,and the measure of division in electric power network. The reduction of decision-making table can be disparted two parts in rough sets: attribute reduction,value reduction. The attribute reduction can predigest the complexity grade,and then ,the process of value reduction can delete the attribute value of redundancy in all non-infection rele tables.The main work of the paper is two facts.In the first place,the paper applys basic arithmetic to practicality,using to compile arithmetic program by MATLAB.And the paper gets the rules of power net and then compile fault diagnose by VC ++;On the other hand,the paper combines and analyses the sixth factory DA QING with the thought of butterfly division of cosmically power net.At last,the result of distribution power net is satisfactory.