Design of artificial neural networks for on-line static security assessment problems

The design of artificial neural networks for static security assessment problems is systematically summarised in this paper. The static security assessment (SSA) problem is formulated as a pattern classification or pattern recognition problem which can be solved by artificial neural networks. Kohonen, backpropagation and counterpropagation networks are used for on-line SSA problems in our research. A feasible feature selection technique is also described to reduce the dimension of the input pattern. The tests results indicated that ANN-based SSA approaches have good performance.