Combined Wavelet Transform and ANN for power system security analysis

Power system security is one of the vital concerns in competitive electricity markets due to delineation of the system controller and the generation owner. This paper presents a new approaches based on artificial neural networks combined with wavelet transform is developed for power system security analysis. The off-line AC load flow calculations are used to construct three kinds of violation indices namely line over load index (LOI), power margin index (P_IP) and voltage stability index. Wavelet transform is used to decompose the data into several coefficients which are fed to artificial neural network for estimating the violation indices. The effectiveness of the proposed approach has been demonstrated on 6-bus and IEEE 30- bus systems for contingency screening ranking at different loading conditions and comparisons are made with conventional method. Good calculation accuracy, faster analysis times are obtained by using wavelet transform based neural network.