模糊神经网络技术在故障诊断中的应用

Application of fuzzy-neural networks in fault diagnosis

总结了模糊神经网络在故障诊断中的应用,分析和归纳了不同结构的模糊系统与神经网络的合成形式。对直接模糊神经网络、B样条模糊神经网络、RBF模糊神经网络以及几种扩展的模糊神经网络在故障中的应用进行了讨论。针对故障诊断通用模型,给出了子模块的改进和选取方法,从而使故障诊断的模糊神经网络方法更为系统和有效。

Fault diagnosis become more and more important to keep the whole plant safe and reliable. The model-based fault detection and diagnosis approach can not satisfy the increasing complexity and nonlinear. The causalities between the faults and the symptoms are quite difficult to be modeled analytically, so a free model approach is more suitable to acquire represent and process the diagnostic knowledge base. The combinatorial forms of several fuzzy-neural networks with different structures are analyzed. The direct fuzzy-neuron network (FNN), B-spline function network RBFNN and some extending fuzzy-neural networks applications in FDD are studied. And a general fault diagnosis model and method is proposed.