Measurement Feedback-Driven Load Restoration Enhancement of Active Distribution Networks with Soft Open Points

Increasing distributed generator (DG) penetration poses significant challenges in the restoration performance of active distribution networks (ADNs). Load restoration after topology reconfiguration has not been adequately considered. With the development of flexible interconnection technology, restoration performance can be further improved by electronic devices, such as soft open points (SOPs). Model-based methods may not align with SOPs' rapid regulatory capability, and the pre-training process may constrain machine learning-based methods. To address these issues, a measurement feedback-driven method is proposed to enhance continued load restoration in ADNs with SOPs. First, a dynamic mapping matrix (DMM) is extracted to depict the relationship between the restoration strategy and the system states of the outage area. Then, a measurement feedback-driven model is established to effectively implement the continued restoration strategy, thereby avoiding impacts from iterative interaction with ADNs. A rectification mechanism of DMM is further designed to improve the adaptability to DG fluctuations. The load restoration capability of the SOP is fully exploited to enhance the restoration performance. The case studies are conducted on a practical distribution network with the four-terminal SOP. Results show that the proposed method can effectively leverage the benefits of the SOP to restore affected loads while adapting to DG fluctuations.