摘要
为提高装备模拟电路的软故障诊断能力,在构建多分辨率变换样本基础上,提出一种应用改进UKF算法训练小波RBF神经网络(WRNN)软故障诊断方法。它引入基于方差膨胀原理的自适应因子,改善UKF算法性能;并利用改进UKF算法优化估计WRNN参数,建立多分辨率变换样本集的故障诊断模型;再由所建模型对各种故障模式进行诊断判定。Sallen-Key带通滤波器的仿真测试表明,该方法收敛速度快,诊断准确率高,进而验证了其可行性和有效性。
To improve the capacity of soft fault diagnosis in analog circuits for military equipment,a Wavelet RBF Neural Network（WRNN）soft fault diagnosis method based on modified UKF algorithm is proposed on the basis of the samples generated by multi-resolution transform. Firstly,an adaptive factor based on variance inflation principle is introduced to improve the capacity of UKF Algorithm.Then,the modified UKF algorithm is used to optimize the parameters of WNN,establishing the diagnosis model based on sample set obtained by multi-resolution transform. Finally,each fault mode is diagnosed by built model. The simulation on Sallen-Key bandpass filter shows that,the proposed method has a good convergence rate and diagnosis correct rate. This validates its feasibility and effectiveness.
作者
郭峰
王春兰
李胜厚
杨国洲
GUO Feng;WANG Chun-lan;LI Sheng-hou;YANG Guo-zhou(XiJing College,Xi’ an 710123,China;Air Traffic Control and Navigation College,Air Force Engineering University,Xi’ an 710051,China)
出处
《火力与指挥控制》
CSCD
北大核心
2018年第7期175-180,共6页
Fire Control & Command Control
