Abstract— Ensuring stable grid synchronization in Modular Multilevel Converters that replicate the behaviour of Virtual Synchronous Generators is essential for maintaining reliable power system performance and achieving optimal dynamic response. This paper proposes a novel optimization strategy for VSG control within MMCs, utilizing a super-twisting algorithm-based sliding mode to systematically adjust key parameters such as synthetic inertia, damping coefficients, and power-frequency droop settings. Through continuous adaptation of these control parameters, the system enhances its ability to regulate frequency and voltage with greater accuracy, particularly under mild disturbances and variable grid conditions. The proposed method improves the converter’s responsiveness and stability, enabling it to better emulate synchronous machine characteristics in real-time. Extensive simulation studies conducted across diverse operating scenarios highlight the effectiveness of the approach. Notable improvements include faster synchronization with grid frequency, accelerated stabilization following disturbances, improved voltage recovery, and stronger attenuation of oscillations. The algorithm’s ability to fine-tune control dynamics in response to changing conditions contributes to a more resilient and adaptive power conversion system. Moreover, the framework demonstrates high flexibility and robustness, making it well-suited for integration into future smart grid environments characterized by inverter-based generation and distributed energy resources. By enhancing the dynamic behaviour and control precision of VSG-enabled MMCs, this technique offers a compelling solution for modern power systems that demand both reliability and adaptability. The STA‑SM control framework delivers marked enhancements compared to the conventional VSG approach. Numerical evaluation indicates a substantial decrease in total harmonic distortion (THD) by 42.9%, alongside a 75% improvement in settling time and a 66.7% reduction in frequency deviation, all of which contribute to stronger dynamic stability. Reactive power output is lowered by 83.75%, demonstrating superior regulation capability.