| 基于多传感器数据融合技术的电子设备故障诊断 |
 点此下载全文 |
| 引用本文：吴苏， 吴文全，王薇.基于多传感器数据融合技术的电子设备故障诊断[J].计算技术与自动化,2016,(1):27-30 |
| 摘要点击次数: 1077 |
| 全文下载次数: 22 |
|
|
| 中文摘要:针对电子装备故障诊断中单一类型故障特征量和诊断方法无法完成诊断任务，导致故障诊断率不高的问题，将多传感器数据融合技术应用于多注速调管发射机装备故障诊断。构建故障诊断模型，提出把故障诊断过程分为两个层次。首先，借助不同的神经网络实现多输入信号的函数变换的功能，获得各种故障基本概率分配值；然后，在决策层利用D-S证据理论的合成法则将各神经网络诊断结果融合起来统一判决，得到最终综合诊断结果，通过实例仿真，并与初步诊断结果进行比较，结果表明早期故障识别率大大提高。 |
| 中文关键词:故障诊断 数据融合 发射机 神经网络 D-S证据理论 |
| |
| Fault Diagnosis of Electronic Equipment Based on Multi-sensors Data Fusion Technology |
|
|
| Abstract:Because the solution of insufficient test data and single information can not represent all the fault states in the analog circuit fault diagnosis, a fault diagnosis model and a fusion algorithm based on data fusion technology were proposed. The fault diagnosis information was fused with two levels: For the character level, the voltage and current of testing nodes were processed by different neural network in order to acquire BPA of various faults. For the decision-making level, the ultimate result was acquired by means of D-S evidence theory. The simulations of diagnosis example indicate that taking two levels for data fusion is more accurate than using neural network as single fusion level. |
| keywords:fault diagnosis data fusion transmitter neural network D-S evidence theory |
| 查看全文 查看/发表评论 下载pdf阅读器 |
