摘要
针对传统的故障诊断方法在装备故障诊断中带来的准确率低问题,提出了一种基于小波变换(WT)和深度置信网络(DBN)的通信控制设备故障诊断方法。该方法通过采集样本故障数据、分析处理故障数据、输入神经网络进行学习、对测试数据进行识别等步骤对通信控制设备进行故障诊断,并与支持向量机(SVM)、k类临近法(KNN)和BP神经网络等传统故障诊断方法进行比较。仿真实验结果表明,提出的方案识别准确率达到93%,与传统方法相比具有更高的准确率,表现更好。
Aiming at the high time-consuming and low-accuracy rate brought by traditional fault diagnosis methods in equipment fault diagnosis,a fault diagnosis method for communication control equipment based on wavelet transform(WT)and deep confidence network(DBN)was proposed.The method performed fault diagnosis on the communication control device by collecting sample fault data,analyzing and processing the fault data,learning by inputting the neural network,and identifying the test data.It was compared with traditional fault diagnosis methods such as support vector machine(SVM),k-class proximity(KNN)and BP neural network.The simulation results showed that the proposed scheme had an accuracy of 93%,which was of higher accuracy and better performance than the traditional method.
作者
陈子兆
矫文成
孙慧贤
陈旭
CHEN Zizhao;JIAO Wencheng;SUN Huixian;CHEN Xu(The Army Engineering University of PLA,Shijiazhuang 050003,China)
出处
《探测与控制学报》
CSCD
北大核心
2020年第2期86-91,共6页
Journal of Detection & Control
基金
电子信息装备远程维护保障技术基金项目资助(41404030102)。
