Imperial College London > Talks@ee.imperial > Control and Power Seminars > Approximation and Learning in Monitoring and Control: Concepts, Results and Perspectives

Approximation and Learning in Monitoring and Control: Concepts, Results and Perspectives

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Alessandro Astolfi.

The objective of this seminar is to give a tutorial overview of some learning approaches to the approximate solution of functional optimization as well as fault diagnosis problems in the context of nonlinear uncertain systems.

In the first part of the lecture, a general class of functional optimization problems not having an analytical solution will be studied. An approximate method to solve these problems will be descrive based on two steps. 1) The decision functions are constrained to take fixed functional structures, in which a certain number of ``free’’ parameters have to be optimized. This allows the functional optimization problem to be reduced to a nonlinear programming one. 2) Since the resulting nonlinear programming problems are characterized by highly complex cost functions that typically include averaging operations, stochastic approximation algorithms are considered to minimize such functions in the context of a suitable learning problem. A few considerations on the approximation capabilities of the chosen functional structures will be made as well as a quick overview about successful applications of the methodology under concern.

Subsequently, a more detailed presentation of a learning-based approximate fault diagnosis methodology will be presented. This approach is based on the design of a monitoring module that provides the information on the detection of a fault and the information on the specific fault that occurred in a class of a priori-specified fault structures. This module is made of a bank of nonlinear adaptive estimators. One of the nonlinear adaptive estimators is the fault detection and approximation estimator (FDAE) used for detecting and approximating faults. An on-line approximation model, typically based on neural approximators, is used in the FDAE . The remaining ones are fault isolation estimators (FIEs) used only after a fault is detected for isolation purposes. Each FIE corresponds to a particular type of fault in a pre-specified class. Several aspects of the fault diagnosis problem will be covered.

Finally, perspectives on the extension of the monitoring methodology to the case of distributed large-scale systems, with an emphasis on challenging applications, will be given.

This talk is part of the Control and Power Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


Changes to Talks@imperial | Privacy and Publicity