Validation of a data-driven multi-ancillary services framework for photovoltaic/Storage units via power hardware-in-the-loop testing

Abstract

This paper presents the experimental validation of a unified multi-ancillary services (AS) architecture for distributed photovoltaic–battery energy storage (PV-DBESS) systems using power hardware-in-the-loop (PHiL) testing. The focus is on voltage regulation (VR) and voltage unbalance mitigation (VUM), with full decoupling between the two control schemes to ensure interference-free operation. Experiments are conducted on a small-scale distribution network using a hybrid PHiL setup that combines real-time simulation with physical hardware, bridging the simulation-to-practice gap. A detailed three-phase four-leg converter model with embedded VR and VUM algorithms is implemented in RSCAD, and simplified controllable current source models are also evaluated to balance computational efficiency with accuracy. The results demonstrate that the proposed control strategies effectively mitigate overvoltage and unbalance events across varying operating conditions and network characteristics. The VUM scheme leverages reactive power through virtual damping susceptances, while the VR scheme coordinates active and reactive power to regulate the positive-sequence voltage. Interoperability with conventional constant-power converters is also verified. The study confirms that the proposed AS architecture provides a robust and practical solution for enhancing reliability and hosting capacity in active distribution networks, while indicating its potential suitability for future large-scale real-time studies and integration with advanced grid operation systems, particularly through the use of simplified converter models.