Mutlivariate Statistical Process Performance Monitoring
时间: 2015/03/27 14:06:43 点击: 次
演讲人：张杰副教授 (Jie ZHANG)
School of Chemical Engineering and Advanced Materials
Newcastle University, Newcastle upon Tyne, UK
报告题目：Mutlivariate Statistical Process Performance Monitoring
In today’s chemical and process industries, plants are becoming larger, more complex and heavily instrumented. The requirements to manufacture products with minimal variations around desired quality targets and to operate safely according to health, safety and environmental protection regulations, have become essential due to market and public demand. The key to successful operation is efficient on-line process monitoring, which enables the early warning of process disturbances, process malfunctions or faults. This talk presents some multivariate statistical process performance monitoring techniques that capitalise on the huge amount of historical process operational data. Many industrial processes are characterised as “data rich and information poor”. Discovering useful information through analysing the huge historical process operational data is the key in successful process monitoring. The bases of multivariate statistical process performance monitoring techniques are multivariate projection techniques, such as principal component analysis (PCA), partial least squares (PLS), multidimensional scaling (MDS), and canonical variate analysis (CVA). The philosophy behind these approaches is to reduce the dimensionality of the problem by forming a new set of latent variables to obtain an enhanced understanding of the process behaviour. The basic techniques of multivariate statistical process monitoring will be presented. Techniques for coping with nonlinear processes, batch processes, and processes with multiple operating modes will also be discussed.
Jie ZHANG received his PhD in Control Engineering from City University, London, in 1991. He has been with the School of Chemical Engineering and Advanced Materials, Newcastle University, UK, since 1991 and is currently a Senior Lecturer. His research interests are in the general areas of process system engineering including process modelling, batch process control, process monitoring, and computational intelligence. He has published over 250 papers in international journals, books, and conference proceedings (H-index of 26 based on Web of Science). He is on the Editorial Boards of a number of journals including Neurocomputing published by Elsevier and Control Engineering of China.