Contingency screening for steady-state security analysis by usingFFT and artificial neural networks

摘要

A new approach based on artificial neural networks (ANNs) combinednwith fast Fourier transform (FFT) is developed for single linencontingency screening in steady-state security analysis. The offlinenfast decoupled load flow calculations are adopted to construct two kindsnof performance indices, PIp (active power performance index)nand PIv (reactive power performance index) which reflect thenseverity degree of contingencies. The results from offline calculationsnof the load flow are used to train a multilayered artificial neuralnnetwork for estimating the performance indices. FFT is used fornpreprocessing the inputs to improve and speed up the ANN training. Theneffectiveness of the proposed method is demonstrated by contingencynranking on two IEEE test systems and comparisons are made with thentraditional method. Good calculation accuracy, high contingencyncapturing rate and faster analysis times for contingency screening arenobtained by using the ANNs