Methods of Power Systenms Remiability Assessment Based on Rough Nerual Network

The number of components states combination and the power flow calculation are main causes producing “computation catastrophe” of reliability assessment calculation about composite generation and transmission systems. The input variables of artificial neural network are reduced, learning samples are extracted, stochastic events are roughly classified, a probable rule set about the relation between stochastic event classes and system states are draw out by means of rough set methods. A contingency pattern identification model-Rough Neural Network (RNN), is presented. Furthermore, a power system reliability evaluation algorithm based on RNN is put forward for increasing the calculation speed of reliability assessment. The numerical experiments for reliability testing systems show the correctness, feasibility and usefulness of the presented method.