Parameter-Estimation for an inverter-controlled induction machine

Kurzfassung

The presented work focuses on an application of the usage of advanced control-techniques such as Model Predictive Control, Moving Horizon Estimation and Koopman-Estimation in the context of electrified aviation. For this, we propose the usage of the Koopman-framework as a parameter estimation technique in a three phase, variable-speed induction motor drive connected to a three level inverter controlled by Model Predictive Control. The Koopman estimation technique proposed is constructed over a large set of data obtained through simulations using moving horizon estimation. An extended library is used to upgrade the dimensionality. Moreover, a dynamic mode decomposition is used to solve the optimisation problem by means of a singular value decomposition