Multicontingency steady state security evaluation using fuzzy clustering techniques

This paper provides a description of a new approach for steady state securi ty evaluation, using fuzzy nearest prototype classifiers, The basic method has an off-line training phase, used to design the fast classifiers for on- line purposes, allowing more than the two traditional security classes. A b attery of these fuzzy classifiers, valid for a specific configuration of th e network, is adopted to produce a global evaluation for all relevant singl e Contingencies. An important feature of this approach is that it selects a utomatically the most appropriate number of security clusters for each sele cted contingency. Natural language-labeling is also used to produce standar dized sentences about the security level of the system, improving in this w ay the communication process between the system and the operator. The paper is completed by an example on a realistic model of the Hellenic interconne cted power system, where seven contingencies were simulated.