Carbon-Aware Workload Management in Data Centers: A Multi-Energy Integration Approach

Data Centers (DCs) are significant contributors to global carbon emissions due to their high energy consumption. Addressing this challenge requires integrated approaches to optimize energy use and reduce carbon footprints. This study presents a carbon-aware energy management model for data centers, incorporating multi-energy integration to minimize operational emissions. The model, based on Mixed-Integer Linear Programming (MILP), is demonstrated on a 1MW hypothetical data center in Istanbul, Turkiye.

The model integrates various energy components besides the grid; on-site combined cooling, heating, and power systems, photovoltaic panels, batteries, thermal energy storage, heat pumps, and district heating connections. The objective is to reduce carbon emissions by optimizing the dispatch of energy by these components while accounting for variations in grid emission factors. Simulation results show that the model can reduce net daily CO2 emissions by up to 35% and lower energy costs by 44% when the data center is connected to a large district heating network, compared to conventional grid-reliant scenarios. These findings demonstrate the potential of multi-energy integration to reduce the environmental impact of data center operations while maintaining economic viability.