Simulation Settings and Start
Define the simulation period, network and solver settings and start the dynamic annual simulation
Overview
On the Dynamic Simulation page, you define the simulation period, network-specific model options and solver settings and start the annual simulation. Before the start, VICUS Districts automatically checks the network for typical error sources (open ends, missing pump characteristic curves, outdated simultaneity results) and exports the project into the solver’s calculation model.
Access
Simulation button in the main toolbar, then the Dynamic Simulation page in the list on the left. On the left are the Network Settings and Performance Options tabs, and to the right of them the simulation period and start buttons.
Simulation Period
Start of Simulation area:
| Field | Format / Unit | Meaning |
|---|---|---|
| Start date: | dd.mm.yyyy hh:mm:ss | Beginning of the simulation period |
| End date: | dd.mm.yyyy hh:mm:ss | End of the simulation period |
| Duration: | value + time unit (e.g. h, d, a) | Length of the period |
End date and duration are coupled: if you change the duration, the end date is adjusted automatically, and vice versa.
Network Settings
Network Settings tab, Model Options for Hydraulic Networks area:
| Option | Default | Unit | Meaning |
|---|---|---|---|
| Network to simulate: | active network | – | Selection when there are multiple networks in the project |
| Nominal temperature difference: | from network parameters | K (0–100) | Nominal temperature difference between supply and return of the network |
| Position of the pressure maintenance: | – | – | At node inlet or At node outlet of the reference node; the source nodes (energy plants) are available as reference nodes |
| Pressure at pressure maintenance: | from network parameters | bar (0–100) | Reference pressure that is held at the pressure-maintenance node |
| Length of the pipe discretization: | from network parameters (depending on network type, e.g. 10 m) | m | Maximum length of the fluid volumes into which each pipe is thermally subdivided (see Model Representation) |
| Initial value fluid temperature: | empty = automatic | °C (−100–500) | Initial temperature of the fluid in the entire network at the start of the simulation |
The Options for the Coupling with the Ground area only appears if the network contains pipes with ground heat exchange (ground model) or geothermal components with a coupled ground model:
| Option | Unit | Meaning |
|---|---|---|
| Coupling time step: | min | Communication time step between the network model and the ground models; a maximum of 20 min is recommended for numerical stability |
| Discretization into several ground models along the network | – | Generates several ground models along the routes instead of one model per laying situation (default: off) |
| Temperature interval of the fluid: | K | Temperature step size for the split into ground models (only with discretization enabled) |
| Number of ground models: | – | Display of the resulting number of models after clicking Calculate the discretization of the ground model |
Solver Settings
Performance Options tab. The default values are suitable for heat networks; adjustments are only necessary in case of convergence problems or very large networks.
| Option | Default | Meaning |
|---|---|---|
| Relative tolerance: | 10⁻⁵ | Relative error tolerance of the time integration |
| Integrator: | CVODE | Time-integration method with adaptive step size; alternatively Implicit/Explicit Euler with a fixed step size |
| First time step: | 0.1 s | Initial time step of the integration |
| Maximum time step size: | 1 h | Upper limit of the adaptive step size |
| Solver for linear equation systems: | KLU (sparse) | Alternatively Dense, GMRES or BiCGStab |
| Solver for hydraulic networks | – | Newton convergence tolerance:, initial conditions of the Newton method (Use results of the previous time step as initial condition) as well as optionally Compute system only in fixed time steps: |
The collapsible Command-line solver options area additionally provides:
| Option | Default | Meaning |
|---|---|---|
| Solver message detail: | Normal | Output depth in the console window (Minimum to Very detailed) |
| Number of parallel threads: | empty = automatic | Fixed number of threads (1–64) |
| Write statistics for each internal solution step | off | Generates detailed solver statistics files |
| Close console window after simulation end | off | Close the console window automatically after the run |
| Copy command line to clipboard | – | Copies the complete solver call, e.g. for batch runs |
You configure the outputs of the simulation on the Outputs page; for heat networks, the option Generate default network-related outputs automatically generates all common result quantities (see Model Representation and Result Quantities).
Start and Checks

Start Simulation starts the simulation run with a progress display and a Cancel button.
Test Initialization checks the model initialization for input errors without calculating the simulation – recommended before long runs.
Log opens the solver log file
<ProjectName>/log/screenlog.txt.
Before the start, VICUS Districts automatically performs the following checks:
- Network cleanup: Fittings are corrected and open line ends are cleaned up (as an undoable change).
- Ground coupling: If the coupling time step is larger than 20 min, a dialog offers to reduce it to 20 min (Reduce to 20 min / Keep).
- Controlled pumps without limiting polynomial: Warning with a list of the affected nodes and components; without a limiting polynomial, the simulation may fail. Recommendation: assign the pump characteristic curve from the database via the steady-state calculation.
- Simultaneity: If the simultaneity results are outdated, the optimization is automatically re-run before the start.
The project is then exported into the solver’s calculation model (progress dialog) and the run is started. If the network contains coupled ground models, the simulation automatically runs as a co-simulation of the network and ground models.
Practical tip:
Never start a long annual run unchecked. With Test Initialization you check the model for input errors in seconds without calculating the simulation – this spares you the abort after many minutes of computation time. Also take the pre-start checks seriously: a warning about controlled pumps without limiting polynomial can actually make the run fail; in that case assign the pump characteristic curve beforehand from the database via the steady-state calculation.