基于MSET和模糊识别的凝汽器故障诊断

针对凝汽器的故障诊断问题,基于多元状态估计原理,采用模糊C均值聚类提取凝汽设备的运行状态矩阵,建立了设备的状态监测模型,以相似度指标衡量设备的运行状态,并以此作为参数模糊化的评判依据,获得凝汽器典型故障的专家知识库。在此基础上,根据模糊规则对设备的故障进行识别,提出了基于多元状态估计和模糊识别的故障诊断方法,并给出了该诊断方法的整体流程。以国内某600 MW机组凝汽器为例,采集实际运行数据进行建模与分析,结果表明:所提出的方法对设备参数的估计较为准确,具有一定的监测跟踪能力,并且能够及时准确地诊断设备故障,整体流程清晰直观,适合系统化应用。

Condenser Fault Diagnosis based on MSET and Fuzzy Recognition

For the condenser fault diagnosis problem,condenser equipment operation state matrix is extracted by using fuzzy C-means clustering based on MSET principle,and then the equipment state monitoring model is established.The similarity obtained by MSET model is used to measure the equipment operation state and as the basis for evaluating parameter fuzzification.Furthermore the equipment fault is identified according to fuzzy rules on the basis of condenser typical faults knowledge database,then fault diagnosis method based on MSET and fuzzy identification is proposed and the whole procedure of this method is presented graphically.With the condenser of a 600 MW thermal power unit as an example,the real operating data is obtained for modelling and analysis.Results show that the presented method has excellent estimation for equipment parameters and can be used for monitoring and tracking equipment state.It has been also demonstrated that the equipment fault can be used as early warning and diagnosed,showing the presented method can be used for real project application.