Modeling and Evaluation of Utility-Scale Microgrid Investment for Distribution System Planning

The electrical distribution network faces two great challenges for the immediate future. First, increased affordability of distributed energy resources (DERs)—and advancing control technologies of inverters that interface them to the grid—have driven a shift from a passive to an active distribution network (ADN), which heightens the complexity of system management. Second, the increased frequency of severe weather events and increased potential for a cybersecurity attack necessitate the need for a resilient infrastructure that can respond adaptively to shutdowns within the system. Microgrids (MGs) present a promising framework both to provide hierarchal control of DERs and to increase resiliency with grid-forming and grid-restoring functionality. Though much work has been done to validate the role of MGs in the distribution system, grid owners and utilities need effective methodologies to incorporate MGs into existing system planning frameworks to ensure that this technology is quickly and wisely adopted.

This thesis develops a two-stage optimization framework that models utility investment in medium-voltage microgrids (MVMGs) with consideration to normal and high-stress operating conditions. The problem is designed as a mixed-integer second-order-cone program (MISOCP) compatible with commercial solvers to obtain a global solution. The first stage models MG boundary selection as a multi-area power system splitting problem, co-optimizing network topology along with DER siting and sizing that results in optimal placement of MGs capable of prolonged self-sustainment. The second stage iterates through possible grid reconnection points for each MG to find the optimal point of common coupling (PCC) and optimizes islanding decisions for critical hours.

The proposed two-stage framework was optimized and tested on the IEEE 33-Bus System for baseline, one-area, and two-area cases to analyze and compare the capabilities of the method. The results of the first case study confirm that including MGs in the planning process can lead to heightened resilience against high-stress events that lead to economic savings. The second case study analyzes the value of islanding in a system planning context and classifies scenarios that could provide additional value streams to justify microgrid investment. Finally, suggestions to foster the continued improvement of utility microgrid planning are discussed in the conclusion.