论文部分内容阅读
为了减少基于无线传感器网络(WSN)的轴承故障诊断系统数据传输总量和网络负载同时提高故障诊断准确性,提出一种采用主元分析(PCA)与径向基(RBF)神经网络结合轴承数据的融合与故障诊断算法.首先建立基于LEACH协议的3层融合模型,然后簇首节点采用PCA对大量多传感器数据降维,最后Sink节点采用RBF对数据进行决策级融合.仿真结果表明:该算法3个成员节点各上传10个数据包,簇头节点融合后剩余4个,融合率为86.7%,每组故障识别准确率大于85%.该算法具有很好的识别率和高压缩率,能够很好应用于煤矿设备故障监测.
In order to reduce the total data transmission and network load of bearing fault diagnosis system based on Wireless Sensor Network (WSN) and improve the accuracy of fault diagnosis at the same time, this paper presents a new method based on Principal Component Analysis (PCA) and Radial Basis Function (RBF) neural network combined with bearing data The fusion algorithm and the fault diagnosis algorithm are proposed.Firstly, a 3-layer fusion model based on LEACH protocol is established, and then a large number of multisensor data is dimensionally reduced by using PCA at the head node of the cluster. Finally, RBF is used by the Sink node to fuse data at decision level. Simulation results show that this algorithm The three member nodes each upload 10 data packets, and the remaining four cluster head nodes are fused, the fusion rate is 86.7% and the accuracy of fault identification for each group is greater than 85% .The algorithm has good recognition rate and high compression rate, Well used in coal mine equipment fault monitoring.