基于粗糙集-RBF神经网络的采煤机液压泵故障诊断

Fault Diagnosis of Hydraulic Pump of Shearer Based on Rough Set and RBF Neural Network

针对采煤机液压泵故障征兆多、故障诊断模糊性强的特点,提出了一种基于粗糙集-径向基函数(RBF)神经网络的故障诊断方法,对液压泵内泄漏故障进行诊断。运用粗糙集理论对液压泵原始故障数据集进行属性约简,去除输入冗余信息,得到最小条件属性集。根据最小条件属性集确定RBF神经网络初始拓扑结构,通过网络训练建立故障征兆和故障类别的映射关系,使用Python编程语言实现了故障诊断。试验对比表明,该方法网络结构更加简单,网络学习效率及诊断准确性更高,在采煤机液压泵中有很好的实际应用效果。

In view of the characteristics that the fault symptoms of shearer hydraulic pump are many and the fault diagnosis is fuzzy,a fault diagnosis method based on rough set and radial basis function(RBF)neural network was proposed to diagnose the leakage fault of hydraulic pump.The rough set theory was used to reduce the attributes of the original fault data set of hydraulic pump,eliminated the input redundancy information,and got the minimum conditional attribute set.The initial topology of RBF neural network was determined based on the minimum conditional attribute set,and the mapping relationship between fault symptom and fault category was established through network training,the fault diagnosis was realized by using Python programming language.The experimental comparison shows that this method has a simpler network structure,higher metwork learning efficiency and diagnostic accuracy,and has a good practical application effect in shearer hydraulic pump.