Developing a collector system aggregation technique of a large-scale grid-connected wind farm

Abstract

The increasing complexity of power systems necessitates the development of aggregation techniques to address specific issues. Model reduction is crucial to expand simulation capacity, ensuring the stability and reliability of power system design. This paper presents a collector system full aggregation technique based on the per unit system model and power loss-based aggregation technique applied to the impedance model of a large-scale grid-connected Wind Farm (WF). The purpose of the proposed technique is to obtain a fully aggregated equivalent representation of the WF collector system, which includes High Voltage (HV) three-winding transformer(s) or more than one HV two-winding transformer, to reduce the simulation time and the system complexity while maintaining similar performance compared to the Detailed Model (DM) and high-accuracy results in both steady-state and dynamic conditions. To be validated, the presented Per Unit Full Aggregation (PUFA) technique is used to obtain a Full Aggregated Equivalent Model (FAEM) of the existing Gabel El-Zeit WF, Egypt, Detailed Model (DM) that contains two HV three-winding transformers. The PUFA model is tested using the DIgSILENT PowerFactory software package, leading to a significant reduction in model order and simulation time with respect to the DM. The PUFA technique is compared to the DM and other aggregated models in terms of power flow using Quasi-Dynamic Simulation (QDS), dynamic response using Low Voltage Right Through (LVRT) simulation, Modal Analysis (MA), and small signal simulation. The comparison between the different models and the proposed one has confirmed that it realizes very competitive results with high accuracy in both steady-state and dynamic conditions with the least simulation time.