动态系统集合元素参数与状态估计报告

讲座名称：动态系统集合元素参数与状态估计报告

讲座人：Tarek Ra??ssi 教授

讲座时间：4月15日12:00-4月18日18:00

地点：亚博足彩会议中心303—1

讲座人介绍：

Tarek Ra??ssi received the Engineering degree from National Engineers School of Tunis in 2000, the Master in Automatic Control from Central School of Lille in 2001, the Ph.D. degree from the University of Paris XII in 2004 and the Accreditation to Supervise/Conduct Research (HDR) from the University of Bordeaux in 2012. From 2005 to 2011 he was an Associate Professor at the University of Bordeaux. Currently, he is a Full Professor at the Conservatoire National des Arts et Metiers, Paris. He is a member of the IFAC Technical Committee “Modelling, Identification and Signal Processing” and a Senior member of IEEE. His research interests include fault detection and isolation, nonlinear systems estimation and robust control.

讲座内容：

讲座内容介绍：The goal of these four lectures is to present basic to advanced materials for set-membership parameter and state estimation for dynamical systems subject to measurement noise and uncertainties. The case of switched and non-switched systems are considered. The applications of these technique for systems prognosis is highlighted. These lectures target both beginners and experienced researchers.

时间：2024年4月15日15:00-16:00

报告题目：Set-membership parameter estimation for nonlinear systems

报告摘要：Parameter estimation is key tool in several automatic control applications. During this lecture, the case of nonlinear systems (nonlinearities with respect to the parameter vector to estimated is considered. In such a case, it is shown that the estimation problem can be formulated under a set-inversion problem which can be solved using interval analysis tools. The extension of classical arithmetic operations are then extended to intervals and inclusion functions are introduced. The computational complexity is also discussed during this lecture.

时间：2024年4月15日16:00-17:00

报告题目：Implementation of set-membership parameter estimation using C++ programming

报告摘要：During this lecture, interval arithmetic operations and inclusion functions are implemented using C++. Then, several set-inversion techniques are evaluated and tested for nonlinear parameter estimation. It is shown that these techniques allow to characterize the set of all feasible values which are consistent with the model, the data and the measurements noise bounds. Several simulations are performed.

时间：2024年4月18日15:00-16:00

报告题目：Robust prognosis analysis for uncertain systems

报告摘要：This lecture is dedicated to the problem of model-based prognosis using interval observers and constraint satisfaction techniques for uncertain systems. Under the assumption that the measurement noise and the disturbances are bounded, lower and upper bounds of the unmeasured state and the unknown inputs are jointly computed. Then, the damage state estimation is formulated as an inverse problem. Several models can be considered for the degradation dynamics. During this lecture, a simple polynomial model is considered. The latter is used for the Remaning Useful Life (RUL) prediction based on an interval evaluation of such a degradation model over a time horizon. During this lecture, unknown input interval observers, set-membership parameter estimation and interval prediction are used.

时间：2024年4月18日16:00-17:00

报告题目：Optimal interval estimation for switched systems

报告摘要：The main goal of this lecture is to present a recent methodology to design interval observers for discrete-time linear switched systems affected by bounded, but unknown disturbances. Two design techniques are presented during this lecture. The first one requires that the observation error dynamics are nonnegative while the second one relaxes this restrictive

requirement by a change of coordinates. It will be also shown that the estimation can be optimized based for example on the H_\infty formalism. Some other frameworks, such us L_\infty, can also be considered.

主办单位：机电工程学院