Demand response integration in hybrid renewable energy systems: A campus microgrid case study in Bangladesh

Institutional campuses in developing regions face simultaneous pressures of rising electricity tariffs, organic waste accumulation, and decarbonization targets, yet integrated energy planning frameworks that address all three remain scarce. This study develops a single-stage mixed-integer linear programming model that jointly optimizes the sizing and 8760-hour dispatch of a grid-connected hybrid renewable energy system comprising photovoltaic generation, modular battery storage, and a food-waste biogas generator, applied to a 15,892-student university campus in Kushtia, Bangladesh. An integrated anaerobic digestion chain converts measured campus food waste into a daily electricity potential with daytime-weighted injection scheduling, while combustion thermal output is reported as a co-benefit and is not dispatched within the electrical model. A demand response layer schedules seven flexible task categories, including electric vehicle charging, cooking, water pumping, and cooling, using binary activation variables with comfort penalties; cooling tasks are seasonally deactivated in winter, reflecting Bangladesh’s cooling-only climate. Activating demand response reduces the levelized cost of electricity by 21.6%, from $0.0458 to $0.0359 per kWh, and lowers the annualized system cost by $15,742 per year, while simultaneously reducing optimal battery capacity by 44%, from 640 kWh to 360 kWh, demonstrating that load flexibility substitutes for storage capital under time-of-use pricing. Grid peak demand falls by 19.9 kW (−3.5%) and grid-electricity displacement reduces annual CO 2 emissions by 551 t per year under the nominal demand response case; the anaerobic digestion system yields an additional 648 t CO 2 -eq per year in avoided landfill methane and digestate credits, reported separately as a system co-benefit. An adversarial robust analysis spanning 36 combinations of solar, demand, and biogas uncertainty confirms worst-case feasibility across all vertices and quantifies a value of robustness of 18.3%, equivalent to $0.0066 per kWh, in levelized cost of electricity. The framework is validated against the National Renewable Energy Laboratory’s REopt tool within 10% on levelized cost of electricity and renewable fraction, and offers a reproducible planning template for universities and public institutions seeking cost-effective, low-carbon energy solutions in South Asia.