Security assessment is to predict a power system's ability towithstand a set of next contingencies. An ANN-based pattern recognitionmethod is used to perform static security assessment for power systemsdue to its potential in terms of speed and accuracy for onlineapplication. With the input pattern for ANN be composed of power systempre-contingency state described in busbar power injections (P, Q), theoutput pattern of ANN is composed of the performance index (PI) valuesof power system post-contingency state to a list of next contingencies.So the output vectors of ANN will indicate not only either `secure' or`insecure' state of the current system but also the severity of securitylimit violations under contingencies. To cope with the curse ofdimensionality and improve efficiency of ANN, R-ReliefF algorithm isintroduced to extract those variables that are with more discriminatoryinformation from (P, Q) set to realise the nonlinear mapping from inputspace to output space. The proposed algorithm is tested on a 77-busbarpractical power system with promising results
展开▼