当前位置: 首页 > 文章 > 基于DS证据理论的精馏塔故障诊断方法 长江大学学报(自科版)农学卷 2013,10 (28) 5+66-68
Position: Home > Articles > Distillation Column Fault Diagnosis Method Based on DS Evidence Theory Journal of Yangtze University(Natural Science Edition)Agricultural Science Volumn 2013,10 (28) 5+66-68

基于DS证据理论的精馏塔故障诊断方法

作  者:
杨帆;江星;陈茂林;吴迅;张岗
单  位:
武汉工程大学电气信息学院
关键词:
小波分析;BP神经网络;DS证据理论;故障诊断;信息融合
摘  要:
研究了一种基于多传感器信息融合算法对精馏塔出现的故障进行诊断的方法。首先利用小波分析对精馏塔传感器信号进行有效的滤波预处理,预处理后的数据送入BP神经网络进行初级融合,然后将初级融合后的数据当作DS证据理论的证据对精馏塔故障进行诊断,由最终概率赋值结果可知最大可信度,则可以判断故障的发生。仿真结果显示该算法比单信息故障诊断能取得更准确、更可靠的诊断结果。
译  名:
Distillation Column Fault Diagnosis Method Based on DS Evidence Theory
作  者:
YANG Fan;JIANG Xing;WU Xun;CHEN Mao-lin;ZHANG Gang;Wuhan Institute of Technology,Hubei Intelligent Robot Key Laboratory;Wuhan Institute of Technology;
关键词:
wavelet analysis;;BP neural network;;DS evidence theory;;fault diagnosis;;information fusion
摘  要:
The research of distillation column fault diagnosis is based on the multi-sensor information fusion algorithm. Wavelet analysis is used to filter the signals of distillation column sensor.The pretreatment data are sent into the BP neural network for primary fusion.Then the data after the primary fusion act as the DS evidence for the distillation column to be diagnosed.The final probability assignment results achieve the maximum credibility.The simulation result shows that this algorithm can achieve more accurate and reliable diagnostic results than that of single information fault diagnosis.

相似文章

计量
文章访问数: 13
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊