A MILP Model for the Operational Planning of Multi-Energy Systems Accounting for variable Delivery/Return Temperatures and Non-Isothermal Mixing in Headers

This work presents a MILP formulation for the optimal operation planning of Multi-Energy Systems serving district heating networks. In particular, the work focuses on Multi-Energy Systems (MES) featuring thermal generation units connected in series and/or parallel, units with limitations on the operating temperature range of inlet/outlet water, and headers in which occurs the non-isothermal mixing of water flows. The model accounts for both dispatchable (e.g. CHP engines) and non-dispatchable (e.g. thermal solar panels) generation units, as well as for the presence of stratified thermal storages. To avoid the nonlinearity and nonconvexity associated to the variable stream temperatures and non-isothermal mixing, each water flow is represented as a linear combination of at most two virtual flows at discrete temperature levels, imposing a Type 2 Special Ordered Set (SOS2) condition to identify the two flows with closest temperatures. The model is used to optimize the operation of an integrated MES designed to serve the electric and thermal loads of a University Campus.