Adaptive Volt-Var Control with PV inverter and Solid State Transformer Interface Mobile Energy Storage Systems

The growing integration of distributed energy resources (DERs) such as PV systems and mobile energy storage systems (MESSs) is transforming the conventional distribution model into the active distribution networks (ADNs). However, this transition introduces a highly dynamic and non-linear nature into the system, posing significant challenges to voltage stability and reactive power management. Conventional volt-var control (VVC) approaches incorporating slow-responding devices lack the capability to respond against dynamic environmental conditions, limiting their potential to mitigate voltage violation rate (VVR) and network power losses. To address these issues, this paper proposes an intelligent VVC strategy leveraging PV inverters and solid-state transformer-enabled mobile energy storage systems (SST-MESS). The SST-MESS is capable of providing bidirectional power flow along with reactive power compensation, alongside PV inverters to improve real-time system efficiency. To optimize the proposed multi-objective framework, a novel two-critic reinforcement learning scheme has been implemented. The conceptualized strategy has been tested on the IEEE-69 bus system; a comparative analysis with the baseline VVC approach indicates the effectiveness of the presented strategy, resulting in a 15.645% and 16.015% improvement for two-critic deep deterministic policy gradient (TC-DDPG) and two-critic soft actor-critic (TC-SAC) compared to the baseline scenario, respectively, showing the effectiveness of the presented strategy. Hence, contributing to the optimal integration of DERs in ADNs.