Keynote Address 3	Research on Data-driven Modeling and Its Application in Process Industry
Date/Time	Tuesday, May 31, 2016 11:00-12:00
Venue	International Leture Hall of the 2nd floor
Presenter	Prof. Qunxiong Zhu , Beijing University of Chemical Technology

Abstract

The complexity of industrial processes and data leads to the difficulty in modeling. Now the data-driven method becomes a promising alternative and one of the research hotspots. This presentation will discuss the small sample processing and the model structures design to improve the efficiency of the model for industrial applications. It is well known that even in big data era, small sample problems cannot be ignored. Virtual sample generation (VSG) is a promising technology which generates plenty of new virtual samples by the information acquired from small sample sets, improving the accuracy of the forecasting model. A particle-swarm-optimization-based VSG is proposed to iteratively generate most-feasible virtual samples over the search-space defined via the boundary constraints and a non-linear inequality constraint. As one of the modeling methods, the feedforward neural network has attracted important attention from researchers. This presentation will focus on the hierarchical structure design and the double parallel structure design for the extreme learning machine and its applications to process modeling, energy efficiency evaluation, and alarm root identification of petrochemical industrial processes. The report will further discuss the status and trends of the data-driven modeling, give research issues and the challenging directions of the data-driven modeling methods in complex industrial process.

Biography

Qunxiong Zhu Dr. Qunxiong Zhu is now a Professor and the Dean of the College of Information Science and Technology at the Beijing University of Chemical Technology, China. He is also the director of the Engineering Research Center of Intelligent Process System Engineering, Ministry of Education of China. He has been the President of the Beijing Association of Automation, an executive member of Chinese Association of Automation (CAA), the Vice-Chairman of Technical Committee on Process Control of CAA, a member of Chemical Industry and Engineering Society of China (CIESC), the Vice-Chairman of Technical Committee on IT application of CIESC. He received the first and second prizes respectively for scientific and technological progress from the China Petroleum and Chemical Industry Federation, the first prizes for scientific and technological progress from the China Petroleum and Chemical Industry Automation Application Association, Comprehensive Contributions Award to Chinese Process Control from CAA, Technical Contributions Award and Education Contributions Award to IT application from CIESC. He is also a gainer of Teaching Celebrities in Beijing Higher Education. His research interests include intelligent modeling, engineering optimization, data mining, alarm system design, fault diagnosis, and emergency rescue virtual reality.