Optimal energy storage configuration for power quality enhancement in active distribution networks with coordinated PV operation

The extensive integration of renewable energy sources and power electronic devices in active distribution networks (ADNs) has created substantial power quality challenges. Although energy storage systems (ESSs) are considered a key solution, determining optimal ESS locations and capacities remains an unresolved issue. To tackle this problem, this paper develops a bi-level ESS configuration framework. The upper-level model minimizes ESS configuration costs by optimizing location selection and capacity sizing, where the lower-level model formulates a multi-phase optimal harmonic power flow (MPOHPF) problem, aiming to minimize the power loss in the network and keep voltage deviations, harmonic distortions, and voltage unbalance within standard limits. However, the coupling effects between nodes in ADNs and the inclusion of harmonic variables significantly increase the complexity of solving the bi-level model. Therefore, this paper further introduces a candidate node selection strategy based on their contributions to power quality indices, which efficiently narrows down the configuration space. Furthermore, the lower-level model is transformed into a convex quadratic constrained quadratic program by substituting power flow equations with current injection equations. An alternating iterative algorithm is then applied to optimize and compute power flows iteratively, ensuring both convergence and computational efficiency. Finally, the proposed method is comprehensively verified using a 162-node three-phase four-wire distribution network.