Nonlinear Robust Control of a MIMO Converter for Modular DC Nanogrids Based on Adaptive Disturbance Compensation

This paper presents a model-based, nonlinear, robust control strategy tailored for modular DC nanogrid applications featuring multiple input sources and output loads interconnected via a DC bus. The chosen case study is a Multi-Input Multi-Output (MIMO) DC/DC converter with a fuel cell and a battery pack as power sources, which are interfaced with the common DC bus via bidirectional Half Bridge Modules (HBMs). Such modules share a bus capacitor, and the overall load on the DC bus is modeled as an ideal current generator. The control objective is to regulate the bus voltage and the fuel cell current while ensuring system robustness under varying load and input conditions. The strategy is based on an indirect control approach supported by sliding mode compensation and a linear extended state observer (LESO), which enables disturbance rejection and robustness against parameter uncertainties. An additional external loop dynamically adjusts the equilibrium point to minimize steady-state errors. Experimental results, obtained in a real-time embedded setup using a TMS320F28379D microcontroller and high-performance HBMs, confirm the effectiveness and reliability of the proposed method under different transient and steady-state scenarios.