System theoretic model reduction techniques for nonlinear control systems

The design, analysis, and control of engineering systems or processes such as high-tech equipment, intelligent transportation systems, or smart energy systems, heavily relies on mathematical models of their dynamics. However, due to the increased complexity of such systems and the need for ever-tighter performance requirements, these mathematical models are becoming immensely complex to the extent that simulation, analysis, and controller synthesis become infeasible. As a result, there is a strong need for the development of mathematical tools for approximating complex models by models of lower order. As the latter will be used as a substitute in the (control) design process, these low-order models should not only accurately approximate the original complex model, but should also preserve fundamental properties (such as stability/dissipativity properties and/or model structure). In addition, to successfully use the low-order models in a verification procedure, strong guarantees on the accuracy of these models should be available in terms of error bounds.

This project will therefore focus on the development of model reduction procedures for nonlinear dynamical systems with provable guarantees on properties and accuracy of the low-order model.

Supervisors: Prof. Jacquelien Scherpen and Bart Besselink

PhD: Melvyn Wildeboer