Canonical Variate Dissimilarity Analysis for process Incipiant Fault Detection

来源: 机电工程学院 作者:李明利 添加日期:2018-06-04 08:28:28 阅读次数: 2524

  报告内容: Early detection of incipient faults in industrial processes is increasingly becoming important, as these faults can slowly develop into serious abnormal events, an emergency situation, or even failure of critical equipment. Multivariate statistical process monitoring methods are currently established for abrupt fault detection. Among these, the canonical variate analysis (CVA) was proven to be effective for dynamic process monitoring. However, the traditional CVA indices may not be sensitive enough for incipient faults. In this work, an extension of CVA, called the canonical variate dissimilarity analysis (CVDA), is proposed for process incipient fault detection in nonlinear dynamic processes under varying operating conditions. To handle the non-Gaussian distributed data, the kernel  density estimation was used for computing detection limits. A CVA dissimilarity based index has been demonstrated to outperform traditional CVA indices and other dissimilarity  based indices, namely the dissimilarity analysis, recursive dynamic transformed component statistical analysis, and generalized canonical correlation analysis, in terms of sensitivity when tested on slowly developing multiplicative and additive faults in a continuous stirred-tank reactor under closed-loop control and varying operating conditions. 报告人简介: Yi Cao is a Reader in Control Systems Engineering, Cranfield University. He Obtained PhD in Control Engineering from the University of Exeter in 1996, MSc in Industrial Automation from Zhejiang University, China in 1985. His main research interest is in developing systematic approaches to solve various operational problems involved in industrial processes using both models and data. Dr Cao is the main inventor of the Inferential Slug Control technology to mitigate slugging of multiphase flow in offshore oil and gas production systems. A successful field trial has showed that the technology was able to increase oil production by 10%. This achievement received the Innovation Award from the EastEngland Energy Group (EEEgr) in 2010. His recent research is focusing on data driven self-optimizing control methodology. By applying it to water flooding process for oil enhanced recovery, it can achieve near optimal operation in spite of the uncertainties of oil reservoirs.