An artificial neural-net based technique for power system dynamic stability with the Kohonen model

The authors present an artificial-neural-network (ANN)-based technique for evaluating power system dynamic stability. The method is based on estimating the dynamic stability index that corresponds to the most critical eigenvalue of the S-matrix method. The ANN of Kohonen is used to estimate the index so that computational efforts are reduced and numerical instability problems are avoided. The Kohonen model is based on the self-organization feature mapping (SOFM) technique that transforms input patterns into neurons on the two-dimensional grid. Power system conditions are assigned to the output neurons on the two-dimensional grid with the SOFM technique. Two methods are presented to calculate the index so that an input neuron calls the index corresponding to an input pattern. A comparison of the linear and nonlinear decreasing functions employed at the learning process is made. The effectiveness of the proposed method is demonstrated in a sample system.<
>