Expansion planning of an existing grid-connected microgrid via coordinated solar and hybrid storage: A study on Deakin University township toward net-zero target

Abstract

Existing renewable microgrids often require expansion to accommodate rising demand and more stringent decarbonization objectives, yet most sizing studies focus on new systems rather than upgrades to operating assets. This study develops an optimization-based framework for expanding an existing grid-connected solar-battery microgrid through the coordinated addition of hydrogen storage, battery capacity, and solar panels in a solar farm. The framework integrates discrete component sizing, hourly energy management, power balance and storage constraints, and LCOE minimization and is validated using five years of hourly SCADA data from the Deakin University Waurn Ponds campus microgrid. Three cumulative expansion strategies are compared using five optimization algorithms: mixed-integer linear programming, differential evolution, particle swarm optimization, covariance matrix adaptation evolution strategy, and Bayesian optimization. The strategies consist of: (1) adding hydrogen storage to the existing system; (2) increasing battery capacity alongside hydrogen storage; and (3) expanding the solar farm together with hybrid battery-hydrogen storage. Under differential evolution, the respective LCOEs are 2.81, 0.61, and 0.16 AUD/kWh, with self-sufficiency ratios of 41.8%, 44.3%, and 50.0%. The results indicate that storage-led expansion alone cannot compensate for insufficient renewable generation: hydrogen conversion losses and limited asset utilization increase cost, whereas solar farm expansion directly increases usable renewable production and improves storage utilization. Sensitivity analysis identifies solar farm size and electrolyzer capacity as the dominant economic drivers, with solar farm size offering the more favorable cost–coverage trade-off. These findings highlight that renewable generation and storage must be planned jointly when upgrading existing grid-connected microgrids.