论文信息 - Contingency ranking using neural networks by Radial Basis Function method

Contingency ranking using neural networks by Radial Basis Function method

Radial Basis Function (RBF) Networks are used for contingency evaluation of bulk power system. The off-line Newton-Raphson load flow calculation are adopted to construct two kinds of performance indexes, Pip (active power performance index) and PIv (reactive power performance index) which reflect the severity degree of contingencies. The results of off-line load flow calculation are used to train a Radial Basis Function Neural Network for estimating the predefined performance indices. The effectiveness of the purposed method is demonstrated by contingency ranking on IEEE 30-Bus test system. Faster analysis times for contingency ranking are obtained by using the Neural Network.

S. Jadid | M. Khazaei | M. Khazaei | S. Jadid

[1] Tarlochan S. Sidhu,et al. Contingency screening for steady-state security analysis by using FFT and artificial neural networks , 2000 .

[2] Mohamed A. El-Sharkawi,et al. Dynamic security contingency screening and ranking using neural networks , 1997, IEEE Trans. Neural Networks.

[3] Shahram Jadid,et al. An expert system to improve power system contingency analysis , 1997 .

[4] Mohamed Mohandes,et al. Radial basis function networks for contingency analysis of bulk power systems , 1999 .

[5] K. L. Lo,et al. Fast real power contingency ranking using a counterpropagation network , 1998 .

[6] V. Brandwajn,et al. Complete Bounding Method for AC Contingency Screening , 1989, IEEE Power Engineering Review.

[7] Hiroyuki Mori,et al. A preconditioned fast decoupled power flow method for contingency screening , 1995 .