摘要
多元统计分析可以应用于过程监控,主元分析方法与多变量统计计算可以确定热工多变量动态过程的主导因素。根据机组的运行机理,确定引起主要监控变量变化的相关过程变量,采集相关过程变量的运行数据,建立输入数据矩阵和主元模型,计算T2统计量及其控制限,计算在最大时对应的各过程变量对主元的贡献大小,确定机组运行过程中引起主要监控变量变化的主导因素。主导因素的确定为热工过程动态建模和故障诊断创造了条件。
The multivariate statistical analysis can be applied to the monitoring and control of a process.The main variant analytic method and multivariable statistical calculation can be used to determine the major factors influencing the dynamic process of thermodynamic multivariables.In the light of the operating mechanism of a machine unit,the relevant process variables that cause a change of the main monitoring and control variables were determined,the operating data of the relevant process variables collected,and data input matrix and main variant model established.Moreover,T2 statistical amount and control thresholds were calculated with the contributions of various corresponding process variables to the main variant being computed when T2 is at its maximal value.The major factors causing a change of the main monitoring and control variables during the operation of the machine unit were determined,which has created a favorable condition for the dynamic modeling of the thermodynamic process and the conduct of fault diagnosis.
出处
《热能动力工程》
EI
CAS
CSCD
北大核心
2007年第4期414-417,共4页
Journal of Engineering for Thermal Energy and Power
基金
河南省教育厅自然科学研究基金资助项目(2006470010)
华北水利水电学院青年科研基金资助项目(HSQJ2005001)
关键词
现场数据
动态过程
主元分析
多变量统计
主导因素
on-site data,dynamic process,main variant analysis,multivariable statistics,major factor
