A Parallel Sectionalized Restoration Scheme for Resilient Smart Grid Systems

Abstract

A parallel automated resilience-based restoration methodology is presented in the power system to minimize impact due to emergency power outages. In this power restoration process, a black start unit is assigned to a small region (i.e., a section) on an as-needed basis. A mixed integer nonlinear programming model is developed in order to optimally sectionalize the region of interest all the while maximizing the resiliency in terms of load shedding, restoration time, and network connectivity. For solving this large scale optimization model, a bi-level programming approach is proposed. This approach consists of two optimization levels. The sectionalization problem (upper level) is a mixed integer programming model and finds the optimal section set. The restoration problem (lower level) is a linear model and determines the dc optimal power flow and restoration time for the optimal section set identified in the upper level. We use the pre-emptive method of goal programming to deal with multiple conflicting objectives in the model. Our proposed solution approach outperformed mathematical programming with equilibrium constraints and found near optimal solutions. Numerical results and sensitivity analysis from two case studies (6- and 118-bus IEEE test systems) are further discussed to demonstrate the efficiency of the solution approach.