摘要
根据ART-2网络学习与记忆的特点,对模型进行改进,解决了原始ART-2网络权值学习的随机偏移问题,有效地过滤了噪声。最后给出该算法在轴承运行实时监测和故障诊断中的应用,验证了该算法的有效性和可行性。
According to the learning and remembering feature of ART- 2. noise and random excursion were improved by the newmodel. At last,an example of real time monitoring and fault diagnosis in bearing＇ s detection,validation the effectiveness and feasibilityof the alogrithm.
出处
《仪器仪表用户》
2009年第6期46-47,共2页
Instrumentation
