基于优化NRS和复杂网络的柴油发电机组故障诊断

为解决柴油发电机组故障诊断中复合故障难以识别的问题,设计了一种变分模态分解、优化邻域粗糙集和社团层次聚类相结合的故障诊断方法。采用变分模态分解方法对采集的声信号进行分解,并形成初始特征集。考虑到冗余特征的影响,使用优化邻域粗糙集进行特征筛选,达到属性简约的目的。利用复杂网络中的社团结构建立故障诊断网络,通过设计社团区分准则函数找出社团结构,同时实现了故障诊断分类。试验表明,所提方法的故障诊断率达到了99.17%,其有效性及优越性得到了充分证实。

Fault diagnosis of diesel generator set based on optimized NRS and complex network

In order to solve the problem of complex fault identification in diesel generator set fault diagnosis, a fault diagnosis method combining variational modal decomposition, optimal neighborhood rough set and community hierarchical clustering was designed.The variational mode decomposition method was used to decompose the collected acoustic signals and form the initial feature set.Considering the influence of redundant features, an optimized neighborhood rough set was used to filter features to achieve the goal of attribute reduction.The fault diagnosis network was established by the community structure in the complex network, and the community structure was found by designing the association criterion function, and the fault diagnosis classification was realized at the same time.The experimental results show that the fault diagnosis rate of the proposed method reaches 99.17%, and its validity and superiority is fully confirmed.