An islanding partition method of active distribution networks based on chance-constrained programming. (15th May 2019)

• Record Type: Journal Article
• Authors: Zhao, Jinli; Zhang, Mengzhen; Yu, Hao; Ji, Haoran; Song, Guanyu; Li, Peng; Wang, Chengshan; Wu, Jianzhong
• Is Part Of: Applied energy. Volume 242(2019)
• Page Start: 78
• Page End: 91
• Publication Date: 2019-05-15
• Subjects: Active distribution network -- Islanding partition -- Uncertainty -- Chance constrained -- Mixed-integer quadratically constrained programming; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2019.03.118
• Languages: English
• ISSNs: 0306-2619

Abstract

Graphical abstract: Relationship between restored loads and chance level at different fault time. Highlights: The uncertainties of islanding operation in ADNs were modelled. A chance-constrained programming based model was built for islanding partition. The effects of the chance level on the islanding partition strategies were explored. Abstract: With the high integration of uncontrollable distributed generators (NDGs) comprising photovoltaic arrays (PVs) and wind turbines (WTs), many technical issues have become increasingly prominent in the islanding operation of active distribution networks (ADNs). Such problems including branch overloading and voltage violations threaten the secure operation of distribution systems and a continuous power supply. To address the uncertainties of NDG outputs when implementing islanding partition strategies, this paper proposes a new method of the islanding partition of ADNs based on chance-constrained programming. First, power generation scenarios and probability distributions for these scenarios are generated according to historical data considering the time series characteristics and uncertainties of NDGs. Then, the chance-constrained description of islanding operation is expanded to consider the effect of uncertainties related to NDGs. Moreover, an islanding partition model of ADNs based on chance-constrained programming is established. By applying convex relaxation and introducing auxiliary variables, the model is converted to a mixed-integer quadratically constrained programming model that can be effectively solved. Finally, case studies involving the modified IEEE 33-node system, IEEE 123-node system and an actual distribution system are conducted to verify the effectiveness and scalability of the proposed method.