题目: System Identification in Presence of Outliers
报告人：王庆国教授
时间：5.28上午9:00-11:00
地点：逸夫会议中心中心会议室
简介:
The outlier detection problem for dynamic systems is formulated as a matrix decomposition problem with low-rank and sparse matrices, and further recast as a semidefinite programming (SDP) problem. A fast algorithm is presented to solve the resulting problem while keeping the solution matrix structure and it can greatly reduce the computational cost over the standard interior-point method. The computational burden is further reduced by proper construction of subsets of the raw data without violating low rank property of the involved matrix. The proposed method can make exact detection of outliers in case of no or little noise in output observations. In case of significant noise, a novel approach based on under-sampling with averaging is developed to denoise while retaining the saliency of outliers, and so-filtered data enables successful outlier detection with the proposed method while the existing filtering methods fail. Use of recovered “clean”" data from the proposed method can give much better parameter estimation compared with that based on
The outlier detection problem for dynamic systems is formulated as a matrix decomposition problem with low-rank and sparse matrices, and further recast as a semidefinite programming (SDP) problem. A fast algorithm is presented to solve the resulting problem while keeping the solution matrix structure and it can greatly reduce the computational cost over the standard interior-point method. The computational burden is further reduced by proper construction of subsets of the raw data without violating low rank property of the involved matrix. The proposed method can make exact detection of outliers in case of no or little noise in output observations. In case of significant noise, a novel approach based on under-sampling with averaging is developed to denoise while retaining the saliency of outliers, and so-filtered data enables successful outlier detection with the proposed method while the existing filtering methods fail. Use of recovered “clean”" data from the proposed method can give much better parameter estimation compared with that based on
报告人简介：
Qing-Guo WANG was born in Suzhou, the People"s Republic of China (PRC), 1958. He received, respectively, the B.Eng. in Chemical Engineering in 1982, the M. Eng. in 1984 and Ph.D. in 1987 both in Industrial Automation, all from Zhejiang University (ZU) of the PRC. From 1987 to 1989 he was a PRC Postdoctoral Fellow with the Institute for Fluid Power Transmission and Control/Department of Mechanical Engineering, ZU. In 1989 he joined the Department of Chemical Engineering/the Research Institute of Industrial Process Control, ZU, as an Associate Professor. Since 1992 he has been with the Department of Electrical and Computer Engineering of the National University of Singapore where he is currently a Full Professor. He held a Alexander-von-Humboldt Research Fellowship of Germany with Me-, Steuer- und Regelungstechnik, FB 7 --- Maschinenbau, Universitt-GH-Dui**urg from 1990 to 1991 and with Regelungs- und Systemtheorie, FB 16 --- Elektrotechnik, GH Kassel Universitt from 1991 to 1992. He was a co-developer of the first-ever paper machine computer control system in PRC. He received the Young Scientist Award of the Chinese Association of Science and Technology in 1990, and a 2nd-class Technological Progress Award of the State Education Committee of PRC in 1990. He was awarded as an Outstanding China-conferred Ph.D by the State Education Committee of PRC in 1991. His present research interests are mainly in modeling, estimation, prediction, control, optimization and automation for complex systems, including but not limited to, industrial and environmental processes, new energy devices, defense systems, medical engineering, and financial markets. He has published over 300 technical papers, of which over 220 are in international journals with more than 8500 citations, h-index of 51, the Automatica most cited article in 2006-2010 and in the Thomson Reuters list of highly cited researchers 2013 (one of 250 in Engineering worldwise). He authored or co-authored six books entitled, respectively, “Finite Spectrum Assignment for Time Delay Systems”, 1999; “Advances in PID Control”, 1999; “Relay Feedback: Analysis, Identification and Control, 2003, “Decoupling Control”, 2003, and "LMI Approach to Analysis and Control of Takagi-Sugeno Fuzzy Systems with Time Delay", 2007, “PID Control for Multivariable Processes”, 2008, all published by Springer-Verlag. He co-holds five patents in USA and Singapore, two of which have been licensed to USA. He is currently the deputy Editor-in-Chief of the ISA Transactions (USA), a member of the Editorial Advisory Board of Recent Patents in Engineering, and a member of the Editorial Board, THE OPEN CHEMICAL ENGINEERING JOURNAL, Journal of Neural Computing Systems, The Open Electrical and Electronic Engineering Journal, and The Open Nanoscience Journal. He was an Associate Editor of Journal of Process Control (an IFAC journal), Jan. 2002-Dec. 2004. He was the chairman of the IEEE control chapter, Singapore, Jan. 2000-Dec. 2001; chair of the chapter for Jan.2012-present. chair of steering committee of Asian Control Professor Association, 2001-2002. General and program chair of the 4th Asian Control Conference, Singapore, September 25-27, 2002; and general chair of the fourth International Conference on Control and Automation, Montreal, Canada, June 10–12, 2003, and tenth International Conference on Control and Automation, Hangzhou, China, June 12–14, 2013. He has supervised about 20 postdoctoral research fellows and about 30 postgraduate research students.