在自适应分相重合闸中引入人工神经网络（ANN），以便能快速、准确地进行故障类型判别及重合闸相序排定，从而解决因传统重合闸盲合于永久性故障，造成对系统的冲击而引发的一系列问题。结合数字仿真及网络训练，结果证明，这是完全可行的。 The artificial neural network (ANN) is introduced in adaptive fractional reclosing so that the type of the fault can be determined quickly and accurately and the phase sequence of the reclosing can be scheduled so as to solve the problem that the traditional reclosing blindly blind to the permanent fault causes the system The impact of a series of problems caused. Combined with digital simulation and network training, the results show that this is entirely feasible.