Cost-Optimal Coordination of PV Generation and D-STATCOM Control in Active Distribution Networks

This paper presents an intelligent operational strategy that performs the coordinated dispatch of active and reactive power from PV distributed generators (PV DGs) and Distributed Static Compensators (D-STATCOMs) to support secure and economical operation of active distribution networks. The problem is formulated as a nonlinear optimization problem that explicitly represents the P and Q control capabilities of Distributed Energy Resources (DER), encompassing small-scale generation and compensation units connected at the distribution level, such as PV generators and D-STATCOM devices, adjusting their reference power setpoints to minimize daily operating costs, including energy purchasing and DER maintenance, while satisfying device power limits and the voltage and current constraints of the grid. To solve this problem efficiently, a parallel version of the Population Continuous Genetic Algorithm (CGA) is implemented, enabling simultaneous evaluation of candidate solutions and significantly reducing computational time. The strategy is assessed on the 33- and 69-node benchmark systems under deterministic and uncertainty scenarios derived from real demand and solar-generation profiles from a Colombian region. In all cases, the proposed approach achieved the lowest operating cost, outperforming state-of-the-art metaheuristics such as Particle Swarm Optimization (PSO), Sine Cosine Algorithm (SCA), and Crow Search Algorithm (CSA), while maintaining power limits, voltages and line currents within secure ranges, exhibiting excellent repeatability with standard deviations close to 0.0090%, and reducing execution time by more than 68% compared with its sequential counterpart. The main contributions of this work are: a unified optimization model for joint P–Q control in PV and D–STATCOM units, a robust codification mechanism that ensures stable convergence under variability, and a parallel evolutionary framework that delivers optimal, repeatable, and computationally efficient energy management in distribution networks subject to realistic operating uncertainty.