基于人工神经网络的旋转机械多故障同时性诊断策略

Multiple Fault Simultaneous Diagnosis Based on Artificial Neural Networks for Rotating Machine

针对大型旋转机械多故障同时性诊断问题，基于人工神经网络，构造了一种由多个子网络组成的分级诊断网络（ＨＤＡＮＮ）．该网络旨在将一个大的分类模式空间划分为几个小的子空间，以便对各子网络进行有效的训练，提高各子网络的分类能力，从而使整个网络具有高精度的多故障同时性诊断能力．测试结果表明：ＨＤＡＮＮ网络不仅能准确地对单故障进行诊断，而且多故障同时存在的情况下，也能有效地识别出各种故障，该网络具有较高的诊断精度，可用于旋转机械工况实时监测和诊断场合．

Based on artificial neural networks, a hierarchical diagnosis network (HDANN) is proposed with respect to multiple faults simultaneous diagnosis for the rotating machine. HDANN consists of several subnetworks, and aims at dividing a large pattern space into several smaller subspaces, so that the subnetwork can be trained on the subspace, respectively, and the whole network is capable of multiple faults simultaneous diagnosing. The research results show that HDANN can not only achieve single fault diagnosing, but also recognize the existing faults in situation of multiple faults, and is available for real time condition monitoring and diagnosis of rotating machine.