基于粗糙集-BP神经网络的垃圾破碎机故障诊断

针对目前垃圾破碎机故障诊断效率低的问题,设计了一种基于粗糙集理论与BP神经网络的故障诊断系统。结合粗糙集理论和BP神经网络的优点,首先利用粗糙集对原始故障诊断样本进行处理,然后对条件属性进行约简,删除冗余的信息,减少神经网络输入端的数据,从而简化神经网络的结构。并将基于粗糙集-BP神经网络的故障诊断系统对垃圾破碎机进行故障诊断。利用粗糙集对故障知识进行约简,简化BP神经网络结构,提高故障诊断的速度及准确度。将此方法应用于某型号垃圾破碎机的故障诊断中,诊断结果表明所提诊断方法可简化神经网络结构,提高诊断效率。

The fault diagnosis of garbage crusher based on rough Set-BP neural network

To improve the fault diagnosis efficiency of garbage crusher,a fault diagnosis system for garbage crusher is designed,which is based on rough set and BP neural network.First the rough set is used to process the original fault diagnosis sample.Then conditional attribute is simplified by deleting redundant information and lessening data at input of neutral network,thus the structure of the neutral network is simplified.Meanwhile fault diagnosis for the garbage crusher is carried out by the fault diagnosis system based on rough set-BP neutral network.Afterwards fault knowledge is simplified by utilizing the rough set,and the structure of BP NN is simplified as well as the diagnosis speed and accuracy is improved.The proposed method is applied in the fault diagnosis of garbage crusher,which results show that the proposed method can simplify structure of BP NN,improves efficiency of the diagnosis.