神经网络的同伦算法与电力系统状态估计

摘要

本文将同伦算法引入到神经网络的学习训练中，提高了神经网络的学习效率；该方法具有稳定性强，收敛性好的优点；用这种神经网络进行了电力系统的状态估计，算例表明了该方法的有效性.%Besed on the homotopy method of neural net work, thus is canlearn faster jn lower calculation cost and can imprive generalization ability and learning efficiency for network. The method is robust and has very good convergence property. Application this neural network on power system state estimation,examples demonstrale the effectiveness of the homotopy method.