基于权值优化的BP神经网络的煤矿配电网故障诊断

Mine Distribution System Fault Diagnosis Research Based on Weight Optimization BP Neural Networks

井下配电网发生故障后,当故障信息不完全或不一致,导致故障诊断难以得到正确结论,而传统BP神经网络存在收敛速度慢、网络的泛化能力较差等缺点。针对上诉问题,提出了一种基于优化权值的BP神经网络的配电网故障诊断方法。通过配电网实例证明了该方法具有较快的收敛速度和较高的诊断准确性,具有良好的应用前景。

If the fault information is incomplete or inconsistent when mine distribution network goes wrong, correct conclusion could not be given by fault diagnosis; traditional BP neural network have disadvantages of learning low efficiency , slow convergence and poor generalization ability. To settle this problem, a new fault diagnosis in distribution network in which weight optimization BP neural networks is designed., through examples of distribution network show that the proposed method has a fast convergence rate, high and good prospects.