抄録
In recent years, according to the modelling of excellent information processing ability of neurons, it has become possible to carry out pattern recognitions or optimizations with the neural network theory. In this paper, we show how to introduce the neural network theory to the load-flow calculation of rectangular coordinate. A problem which input-output functions have for neural network applications is pointed out, and effects of changes in load patterns and system size are discussed in terms of the optimal multiplier. Results of simulations are shown with the convergence characteristics.
