Design of district heating networks

District heating and cooling networks are invaluable technologies to provide energy to urban areas in a cost- and environmental effective way. topotherm is an optimization suite for district heating network design, that is developed and maintained by our research group. It is a Python-based mixed-integer linear programming district heating network model that scales well into larger districts for single and multiple time steps, making it especially suited for renewable energy sources with weather-dependent behavior and variable heating loads. It has been benchmarked against multiple other open-source models [1].

• topotherm features Sseveral optimization models, which share the same core but have differing capabilities and strength:

  • topotherm design: Network design case (that is, the peak demand case) topology and piping optimization for a single demand point of each potential consumer.

  • topotherm RES: developed for weather-dependent renewable energy supply and district heating operation with variable heating demands for a set of time steps.

  • topotherm storage: multiple time steps with daily storage included

  • topotherm robust: design case which is robust against the uncertainty of chosen parameters, such as demand uncertainty or variable electricity prices

  • topotherm nonlinear: non-linear optimization of the system for maximal precision

• Additionally, topotherm supports quasi-dynamic district heating network simulations with the full thermo-hydraulic non-linear equation system to ensure the optimization outcomes are accurate.

Topotherm provides decision making support for multiple producer types and multiple possible locations with variable characteristics, including solar thermal, heat pumps, deep hydrological geothermal, waste heat recovery.

All optimization algorithms minimize the total annualized costs or maximize profits from sold heat. Optionally, different stakeholder targets, such as greenhouse gas emission refuctions can be included with custom constraints and objective functions.