A bi-level robust planning model for active distribution networks considering uncertainties of renewable energies. (February 2019)

• Record Type: Journal Article
• Authors: Wu, Ming; Kou, Lingfeng; Hou, Xiaogang; Ji, Yu; Xu, Bin; Gao, Hongjun
• Is Part Of: International journal of electrical power & energy systems. Volume 105(2019)
• Page Start: 814
• Page End: 822
• Publication Date: 2019-02
• Subjects: Active distribution system -- Second-order cone programming (SOCP) -- Column and constraint generation (CCG) -- Renewable uncertainties -- Robust optimization; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2018.09.032
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A bi-level robust planning model for active management elements is proposed. Variables in both investment level and operation level are associated together. The planning model can be formulated as a mixed integer second order conic problem. The renewable uncertainties in four seasons is addressed by a typical budget uncertainty set. The two-stage robust mathematical model is solved by CCG algorithm. Abstract: The distribution companies will face huge challenges in upgrading the existing network due to the uncertain integration of distributed renewable generations. In this paper, we propose a bi-level robust planning model for active management elements (AMEs) including on-load tap changer (OLTC), electrical storage system (ESS), capacitor bank (CB), and static VAR compensation (SVC) in order to accommodate uncertain development of wind power and photovoltaic power. The planning problem is constructed in two levels which are investment level and operation level. To overcome the poor convergence of the bi-level model, variables in both investment level and operation level are associated together. After equivalent transformation for non-linear terms the planning model can be formulated as a mixed integer second order conic programming (MISOCP) problem by some special means such as second order conic relaxation (SOCR) and big-M approach. We address the renewable uncertainties in four different seasons by a typical budget uncertainty set with adjustable budget. Then a two-stage robust mathematical model is proposed to decide a robust AME deployment scheme and solved by column and constraint generation (CCG) algorithm.