A MILP framework for the integration of CO2-based District heating and cooling networks in multi-energy systems

Multi-Energy Systems (MES) are a promising solution to contribute to the decarbonization of the energy sector, which is crucial to achieving global climate goals and limiting further temperature increases. By enhancing synergies between different energy forms and technologies, MES enable a more efficient use of available primary energy sources, and thus a reduction of greenhouse gas emissions. The main goal of this thesis is to develop a multi-objective optimization model for the design and operation of a MES that can adaptively integrate a 5th generation District Heating and Cooling Network (DHCN) using carbon dioxide (CO2) as heat transfer fluid. The objective is to minimize both the total annualized investment and operating costs and the greenhouse gas (GHG) emissions by determining the optimal sizes of the system components, including energy conversion and storage units and network branches, and their optimal power flows with hourly resolution. The optimization problem is formulated as a Mixed-Integer Linear Programming (MILP) model and implemented in Python using Gurobi as optimizer. Key features of the model include: (i) the integrated optimization of the topology, design, and operation of the CO2 network within the MES framework; (ii) the explicit modeling of the system topology through a multi-nodal approach; (iii) the flexible selection of users to be connected to the network without imposing them as inputs; (iv) the contextual optimization of the CO2 temperature to minimize energy consumption, taking into account all possible operating modes of the conversion units connected to the network; and (v) the regulation of the trade-off between user comfort needs and overall system performance by defining tolerances to the potential comfort loss resulting from satisfying the cooling demand with direct heat exchange at the available temperature. The model is tested on a virtual case study representing a small urban district with 18 nodes, considering different scenarios with different energy conversion and storage technologies, thermal networks, and user demand profiles. The results show that the progressive connection of users to the CO2-based DHCN leads to a reduction of GHG emissions of up to 26.6–52.8% compared to a case without the network, while the impact on total costs is strongly influenced by the thermal demand profiles and cooling comfort needs of the users, ranging from a cost reduction of 7.0% to an increase of 23.1%. In conclusion, the developed optimization model proves to be a valuable tool for the early planning stages of sustainable MES incorporating CO2-based DHCNs by allowing the evaluation of trade-offs between costs, emissions, and user comfort in a wide range of different cases.