Optimized Algorithm for Open-Circuit Fault Detection in Switches and Capacitor Voltage Balancing Control in Modular Multilevel Converters

The modular multilevel converter (MMC) is a favored topology in the industry, but its reliability is at risk with an increase in the number of sub-modules (SMs) due to a rise in switching components. The essential need for maintaining capacitor voltage balance in each arm leads to increased complexity and cost, as numerous voltage sensors are required. This study introduces an innovative approach to minimize the number of voltage sensors by employing an enhanced algorithm for open-circuit fault detection in switches. The proposed scheme organizes each arm into groups, each containing two SMs and one voltage sensor, aiming to reduce the overall sensor count. A novel fault detection mechanism is presented, identifying open-circuit faults by comparing group output voltages in healthy and defective conditions. The capacitor voltage estimation algorithm in the sensor reduction scheme is noted for its simplicity compared to other methods. The effectiveness of these methods is validated through simulations and experimental implementations across diverse scenarios, affirming their reliability.