Time-Varying Hybrid Observer for Load and State Estimation in DC Microgrids

This paper addresses load and state estimation in DC microgrids, with a focus on a Multi-Input Multi-Output (MIMO) converter configuration. In such systems, only the output voltage is usually easily and directly measurable. In contrast, the load cannot be measured and is considered an unknown and uncontrollable disturbance input. For this reason, a specific hybrid observer is proposed in this work to estimate the aggregated load modeled as an ideal current source connected to the DC bus, as well as input currents injected into the DC grid by the source converter. The information required for the estimation process is obtained from the voltage measurement and the knowledge of the control inputs provided to regulate the microgrid. The proposed estimation technique is contextualized within the framework of hybrid systems, providing an elegant setting for the proposed solution. In particular, some theoretical tools are provided in terms of linear matrix inequalities (LMIs), ensuring asymptotic stability. Finally, the estimation technique is validated experimentally, demonstrating its ability to accurately reconstruct the unknown load. This work contributes to the advancement of state estimation techniques in DC microgrids, offering a robust solution for systems with limited sensor availability.