Data-driven modeling and optimal compensation of FACTS devices for real-world performance assessment in the Adama-II wind farm

Abstract Wind power systems routinely face power quality challenges due to the stochastic nature of the wind resource. Effective reactive power compensation is therefore essential to ensure grid code compliance and maintain stable operation. This study presents a real-world performance analysis of FACTS devices, specifically Magnetically Controlled Reactor-type Static Var Compensator (MCR-SVCs) installed at the Adama-II Wind Farm in Ethiopia. It introduces TopFlow, a validated Python-based optimization framework for assessing the real-world performance of reactive power support in wind power systems. This model was cross-verified through PandaPower and DIgSILENT PowerFactory and further validated using SCADA data from the operational wind farm. The proposed model was employed to identify the discrete optimal reactive power compensation requirements to ensure compliance with the steady-state voltage limits of 0.95–1.05 p.u. and to validate the compensation threshold practice of the farm. The deployment of MCR-SVCs enhanced bus voltage by up to 1.4% and reduced reactive losses by more than 42.86% compared to the uncompensated operation. Model predictions closely reflected that the compensation threshold of 80 MW was a viable decision. By integrating real operational data with a rigorously validated analytical model, this study offers one of the few empirically grounded assessments of FACTS-based reactive compensation in operational wind power systems. The results confirm the effectiveness of MCR-SVCs in stabilizing wind farm operation and demonstrate the model’s robustness in identifying discrete optimal compensation thresholds and guiding improved grid-support strategies.