Converter State-Space Model Estimation Using Dynamic Mode Decomposition

Abstract

Power electronic reliability is an important topic for modern industrial electronics. Each component of a power converter has its lifespan, which is affected by many factors like ambient temperature, humidity, load, and thermal cycles of the system. This study focuses on modeling and identifying a state-space model of a power converter under the degradation of passive components, namely the capacitor and inductor. A state-space model of a buck converter is estimated within the framework by dynamic mode decomposition. The algorithm requires 10 to 16 long historical samples of the state variables and the control signal. The numerical results prove that the suggested algorithm can track the system change in time.