Ideal Heating, Thermostat & Surface Heating

Thermostat control signals, ideal convective heating and cooling, ideal surface heating in the active layer and the ideal pipe register with return flow.

VICUS Buildings Page 12 of 12 in this chapter

Overview

The heat and cooling delivery to the zones is modeled via a control chain: the thermostat model generates normalized control signals from the setpoint and actual temperature, which the delivery models — ideal heating/cooling, ideal surface heating or ideal pipe register — convert into powers. All models are parameterized via the usage profile or the surface-heating assignment.

Thermostat

The thermostat model provides per zone the control signals HeatingControlValue and CoolingControlValue (each 0…1) as well as the current setpoints ThermostatHeatingSetpoint and ThermostatCoolingSetpoint [C]. Properties:

  • Setpoints — constant or schedule-controlled (heating and cooling setpoint separately).
  • Reference variable — room air temperature (AirTemperature) or operative temperature (OperativeTemperature).
  • Controller type —
    • Analog (P controller): the signal rises proportionally to the control deviation; the tolerance (TemperatureTolerance [K]) determines the steepness (kP=1/tolerancek_P = 1/\mathrm{tolerance}). At 0.1 K tolerance, the signal is fully modulated 0.1 K below the setpoint.
    • Digital (two-position controller with hysteresis): switches with the hysteresis band TemperatureBand [K] around the setpoint.
  • Pre-heating/pre-cooling (optimum start) — if a pre-heating or pre-cooling time > 0 h is set in the thermostat sub-model, VICUS Buildings modifies the heating or cooling setpoint schedule on export: the first setpoint step of a day towards the comfort value (the first rise for heating, the first drop for cooling) is brought forward by the given time and executed as a staircase of 1-hour steps; the comfort value is reached in the last hour before the original step. There is only one such lead per day, the ramp is clipped at midnight, and days without a setpoint step remain unchanged. The solver thus receives an ordinary setpoint schedule.

Pre-heating/pre-cooling (optimum start)

With the Pre-heating time or Pre-cooling time [h] of the thermostat sub-model (0 = off, max. 23 h), the first daily setpoint change is brought forward. This is done during export directly in the setpoint schedule; the solver’s thermostat model itself remains unchanged.

  • Pre-heating: in each daily cycle, only the first rise of the heating setpoint (setback value → comfort value) is brought forward.
  • Pre-cooling: as a mirror image, only the first drop of the cooling setpoint (higher setback value → lower comfort value) is brought forward.

The step is replaced by a ramp in 1-hour steps that reaches the comfort value in the last hour before the original change. In the kk-th hour before the change:

θk=θcomfort−(k−1) θcomfort−θsetbacktlead\theta_{k} = \theta_{\mathrm{comfort}} - (k-1)\,\frac{\theta_{\mathrm{comfort}} - \theta_{\mathrm{setback}}}{t_{\mathrm{lead}}}

with the lead time tleadt_{\mathrm{lead}} in hours. Example: heating setpoint 17 °C → 21 °C at 7:00 with a pre-heating time of 2 h gives 19 °C from 5:00 and 21 °C from 6:00. The ramp stops at midnight (no carry-over into the previous day); days without such a setpoint change remain unchanged. Later changes on the same day (e.g. after a midday setback) are deliberately not brought forward.

Ideal heating/cooling (convective)

The model converts the thermostat signals into an ideal convective heating or cooling load of the zone, limited by the maximum powers per floor area:

Q˙heat=min⁡ ⁣(1,  max⁡(0,  kP xH+kI ⁣∫xH dt))⋅qheat,max⋅A\dot{Q}_{\mathrm{heat}} = \min\!\big(1,\; \max(0,\; k_P\,x_H + k_I\!\textstyle\int x_H\,dt)\big)\cdot q_{\mathrm{heat,max}} \cdot A

with the heating control signal xHx_H, the zone floor area AA and MaxHeatingPowerPerArea or MaxCoolingPowerPerArea [W/m²]. Optionally, an I component acts in addition to the P component (Kp/Ki parameters, PI controller; the integral component does not accumulate any negative control deviation). The limitation to 0…1 prevents “cooling by the heater” and exceeding the maximum power. Results: IdealHeatingLoad and IdealCoolingLoad [W] (cooling load negative, as energy is removed from the zone), which are accounted for in the room energy balance. A zone without a cooling setpoint receives no cooling load (and vice versa).

Ideal surface heating (active layer)

The ideal surface heating delivers the power not to the room air, but directly into the active layer of a construction (see construction heat conduction):

Q˙active=xH⋅qheat,max⋅Aconstruction  −  xC⋅qcool,max⋅Aconstruction\dot{Q}_{\mathrm{active}} = x_H \cdot q_{\mathrm{heat,max}} \cdot A_{\mathrm{construction}} \;-\; x_C \cdot q_{\mathrm{cool,max}} \cdot A_{\mathrm{construction}}

The control signals xHx_H, xCx_C come from the thermostat of the assigned control zone — which can also be a different zone than the one in which the surface is located. The heat reaches the room only via the component layers, whereby the inertia of a floor or ceiling heating system is physically modeled. Result: ActiveLayerThermalLoad [W].

Ideal pipe register

The pipe register is the more detailed surface-heating model: instead of a power, a mass flow through a pipe register in the active layer is controlled.

  • The mass flow follows the thermostat signal: m˙=x⋅m˙max\dot{m} = x \cdot \dot{m}_{\mathrm{max}} (MaxMassFlux [kg/s]).
  • The supply temperature is constant (SupplyTemperature [C]) or schedule-controlled (e.g. from a heating curve via the supply system).
  • The heat delivery to the layer follows the analytical solution of a pipe in a constant ambient temperature:
Q˙=m˙ cp(1−e− UA/(m˙∥ cp))(Tsupply−Tlayer)\dot{Q} = \dot{m}\, c_p \left(1 - e^{-\,UA/(\dot{m}_{\parallel}\, c_p)}\right)\left(T_{\mathrm{supply}} - T_{\mathrm{layer}}\right)

The UA value is formed from the pipe length, inner diameter, length-related U-value of the pipe wall (UValuePipeWall [W/mK]) and the inner-side heat transfer; the latter is computed flow-dependently via Reynolds, Prandtl and Nusselt correlations from the fluid properties. m˙∥\dot{m}_{\parallel} is the mass flow per parallel pipe strand. A simultaneous heating and cooling signal is not permitted (runtime error).

Results per active layer: MassFlux [kg/s], ActiveLayerThermalLoad [W] and ReturnTemperature [C] — the return temperature results from the supply temperature and the delivered power.

Good to know

For heating/cooling load determination and energy demand, the ideal heating/cooling with a sufficiently high maximum power is usually enough — the outputs IdealHeatingLoad/IdealCoolingLoad then directly provide the demand (the cooling load with a negative sign). The pipe register is worthwhile when supply temperatures, return temperatures or the self-regulation of a floor heating system are to be investigated. A very small thermostat tolerance makes the controller stiff and can significantly increase the computation time.

Stay up to date

New features, tutorials and updates delivered to your inbox.