自动化学术论坛[2021第52-53期]：北京师范大学-香港浸会大学联合国际学院王庆国教授、波兰绿山大学Wojciech Paszke副教授学术报告会

报告时间：9月28日（星期二）15:00

报告地点：信息楼自动化学院310报告厅、腾讯会议（ID: 102 941 342）

（一）

报 告 人：王庆国，北京师范大学-香港浸会大学联合国际学院

     （United International College）教授    

报告题目：Control Design with Guaranteed Transient Performance: An Approach with

                      Polyhedral Target Tubes

内容简介：In this seminar, a novel approach is presented for control design with guaranteed transient performance for multipleinput multiple-output discrete-time linear polytope difference inclusions. We establish a theorem that gives necessary and sufficient conditions for the state to evolve from one polyhedral subset of the state-space to another. Then we present an algorithm which, given a time-varying polyhedral set called the target tube, which may be specified from typical transient performance specifications, constructs an output feedback law that guarantees that the state evolves in this tube. Our formulation is very general and includes reference tracking with any desired transient behavior in the face of disturbances, as specified, for example, by the most popular step response specifications. The approach is demonstrated by an example involving the control of water levels in two coupled tanks. Index Terms—transient performance; control design; time domain specifications; state constraints; discrete-time systems; set-theoretic methods; linear parameter-varying systems.

报告人简历：王庆国教授是南非科学院院士。他目前是北京师范大学-香港浸会大学联合国际学院的讲座教授，以及北京师范大学（珠海）BNU-UIC人工智能与未来网络研究所的教授。王教授的研究领域是自动化/人工智能领域，重点是建模、估计、预测、控制和优化。他在国际期刊上发表了 360 多篇技术论文和七部研究专着。他获得了近20000次引用，h-index为76。他在2006-2010年获得了“Automatica”杂志引用最多的文章奖，并在汤森路透2013年工程领域的高被引研究人员名单中。 2014年获得《控制理论与应用》杂志30年最具影响力论文奖，入选斯坦福大学2020年世界顶尖2%科学家名单（包括职业和年度）。他目前是 ISA Transactions（美国）的副主编。

（二）

报 告 人：Wojciech Paszke，波兰绿山大学（University of Zielona Gora）副教授

报告题目：Attenuation on Non-Repetitive Disturbances in Robust Iterative Learning Control

                     Schemes Designed Over Repetitive Setting

内容简介：This presentation will address the design of iterative learning control (ILC) scheme for a class of linear systems with uncertainties and non-repetitive disturbances. The proposed design method modifies two-dimensional/repetitive setting to include the requirement for Hinf disturbance attenuation. Also, it is shown that the conversion of the control problem to one of stability along the trial for a linear repetitive process leads to design based on linear matrix inequality (LMI) computations. Sufficient conditions for the existence of a robust ILC updating law are derived together with the design algorithms for the associated controller matrices. Obtained results show that the proposed control law is able to fulfil the imposed requirements, i.e., they are suitable for the systems with uncertainties as well as non-repetitive disturbances. An illustrative example will be given to highlight the principles, effectiveness and possible applicability of the proposed method.

报告人简历：Wojciech Paszke received his M.Sc. and Ph.D. degrees, both in electrical engineering, from the Technical University of Zielona Góra, in 2000 and 2005, respectively. Between 2008 and 2010 he was affiliated to Eindhoven University of Technology, The Netherlands, where he has been a control systems expert on high precision positioning of electron microscope. Currently, he is affiliated to the Institute of Automation, Electronic and Electrical Engineering at the University of Zielona Góra, Poland.

Wojciech Paszke has made over 120 scholarly contributions, including nearly 50 peer-reviewed research journal papers (h-index = 23; i10-index = 42). He has been funded by National Science Centre (NCN) grants since obtaining associate professorship (2014-present). Examples of the varied research where he has been Principal Investigator (PI) include: NCN project (2015-2018), which was focused on Improving control performance through learning over repetitions. Another research project (2018-2021) was funded to analyse the learning-based methods for high-performance robust control. The current project is focused on data-driven control and performance assessment for industrial batch processes.

Wojciech Paszke is also the Editor of Multidimensional Systems and Signal Processing and Associate Editor of the Journal of The Franklin Institute and the IET Control Theory & Applications. His current research emphasizes the use of repetitive and multidimensional system theory to: 1) design iterative learning control schemes 2) analyse a class of spatially interconnected systems 3) develop LMI based stability tests for multidimensional systems.