Energy management system for enhanced resiliency of microgrids during islanded operation

• Published in: Electric power systems research Vol. 137; pp. 133 - 141
• Main Authors: Balasubramaniam, Karthikeyan, Saraf, Parimal, Hadidi, Ramtin, Makram, Elham B.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.08.2016
• Subjects: Demand response; Energy management system; Load shedding; Microgrid; Non-linear programming; Resilience; Demand response; Microgrid; Non-linear programming; Energy management system; Load shedding; Resilience
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2016.04.006

•Exact representation of reactive power demand of the system through standard power flow equations.•Operational constraints are included in the optimization model.•Demand response is modeled through a fleet of PHEVs and adjustable loads.•Uncertainty of renewable generation and load is quantified using probability distribution.•Optimization model captures both spatial and temporal variations of critical and non-critical loads. This paper proposes a method to enhance resiliency of microgrids through survivability. Survivability in this context is to minimize load shed for the duration the microgrid is in islanded mode following a disturbance event. During islanded operation, microgrid loads are prioritized as critical and non-critical loads. The key decision is to ascertain whether to provide energy to non-critical loads after supplying the critical loads or to store excess energy for future dispatches. This task is formulated as a non-linear programming problem. The objective is to minimize the amount of critical load shed while maximizing the amount of non-critical load served for a projected restoration time while adhering to relevant operational and physical constraints. For this extended time-scale problem, uncertainty of renewable generation and load forecast is quantified with probability distribution and confidence levels are used to establish likelihood of forecast error. Distributed generation such as solar and wind farm along with battery energy storage system are modeled. Demand response is implemented through adjustable loads and a fleet of plug in hybrid electric vehicles that can be operated in both grid to vehicle and vehicle to grid mode. Test cases are studied on a modified CIGRE microgrid benchmark test system and results are compared with a temporal decomposition scheme based energy management system.