GRID-CONNECTED FAULT-TOLERANT MODULAR MULTILEVEL INVERTERS FOR SMART MICROGRID STABILITY ENHANCEMENT

Abstract

The rapid expansion of smart microgrids necessitates advanced power electronic solutions to ensure stable, efficient, and reliable operation. This study focuses on the development of a grid-connected fault-tolerant modular multilevel inverter (MMI) designed to enhance the stability and fault resilience of smart microgrids. The proposed system aims to mitigate power disruptions, reduce harmonic distortions, and ensure uninterrupted power supply by incorporating fault-detection mechanisms, real-time compensation strategies, and AI-driven control algorithms. The proposed MMI topology is structured with redundant submodules and intelligent reconfiguration strategies to handle switching failures, open-circuit faults, and short-circuit conditions. The system employs model predictive control (MPC), sliding mode control (SMC), and deep reinforcement learning (DRL) algorithms to dynamically optimize switching sequences and ensure seamless power conversion. Key hardware components include SiC MOSFETs, real-time monitoring sensors, and an FPGA-based digital signal processor for high-speed control implementation. The system was tested using MATLAB/Simulink simulations and validated on a hardware prototype under various grid conditions, including voltage sags, phase imbalances, and load fluctuations. The experimental findings indicate that the fault-tolerant MMI system achieved: 99.2% fault compensation efficiency, ensuring minimal power disruptions. Improved total harmonic distortion (THD) of less than 1.8%, significantly lower than conventional inverter designs. Adaptive load-balancing capabilities, maintaining system stability even under 50% module failures. Reduction in switching losses by 18%, improving overall system efficiency and reliability. The results demonstrate that the proposed fault-tolerant MMI significantly enhances smart microgrid stability by ensuring uninterrupted operation under fault conditions. The AI-based control strategies enable real-time fault compensation, ensuring seamless power delivery. This research highlights the practical applicability of AI-integrated modular inverters for next-generation smart grids and renewable energy applications. Future work will focus on optimizing control strategies for large-scale grid integration and real-world deployment.