Heating demand and supply¶

Domestic heating demand and supply simulation requires a set of modules to be integrated and jointly executed. Below the list of modules available in demod

for simulating household indoor temperature settings:

CREST thermostat setting simulator
Living Lab thermostat setting simulator

for building thermal behavior:

Low-order building thermal model (4R3C)

for domestic hot water demand:

CREST domestic hot water demand

for hot water tank thermal behavior:

Low-order hot water tank thermal model (1R1C)

for integrated heat demand for space heating and domestic hot water:

Integrated heat demand simulator

for heating system controllers:

Thermostats
Heating system controller

for heating system operation and consumption:

CREST heating system

for integrated heating demand and supply simulation:

CREST five modules heating simulator

Household indoor temperature settings¶

Two modules are currently available to simulate how each household set indoor temperature set point and switch on/off periods of the heating system.

CREST thermostat setting simulator¶

API: This simulator is currently not available. It will be reliased in future version at CRESTcontrols.
Description: This module implements the approach developed in CREST. First, indoor air temperature set point is stochastically assigned based on empirical discrete distributions. Then, timer setting (i.e., on and off periods) are stochastically simulated using a first order Markov chain model, which uses empirical data for weekdays and weekends. If the timer is set on, the heating system keep indoor air temperature within the deadband of $\pm 2^{\circ}C$ .
Availability: This module uses empirical data from CREST, which are derived from a UK study. No equivalent data are currently available for Germany.

Living Lab thermostat setting simulator¶

API

For details about the implementation of this simulator you can visit VariableThermostatTemperatureSimulator.

Description

This module is inspired by [Sovacool2020] and attempts to simulate heating system control by defining six different usage patterns. These six different patterns aim to give relevance to the heterogeneous behaviour of different households in terms of heating periods (i.e., regularity and dependence on the presence of active residents) and target temperatures.

The six profiles can be briefly described as follows:

Cool Conservers, often adjust temperature to try and cut bills.
Steady and Savvy, rarely adjust their heating as they are fine with 18-20°C.
Hot and Cold Fluctuators, often adjust temperature to get comfortable.
On-Demand Sizzlers, some like it hotter or want to spend more than others in their home.
On-off Switchers, turn it on and off to try and make sure home is only warm when someone is in.
Toasty Cruisers, love feeling cosy and prefer not to put clothes on if they are cold.

Availability

This module is inspired by empirical observations of an UK-based research [Sovacool2020], but makes use of guessed parameters.

Compatibility

This module is compatible with all heating system simulators that accept exogenous indoor temperature set point profiles as inputs.

Building thermal behavior¶

Demod employs simplified lumped-capacitance models to simulate building and heating system thermal behaviour.

Low-order building thermal model (4R3C)¶

API

For details about the implementation of this simulator you can visit BuildingThermalDynamics.

Description

This module simulates the thermal behavior of the building using an equivalent low-order electric circuit as in CREST (see Fig. 3). Six building typologies are available: detached house, semi-detached house and apartment either renovated or not renovated.

The name 4R3C refers to three thermal capacitances representing the thermal masses of the building $C_{b}$ , indoor air $C_{ia}$ , and heat emitters and buffer $C_{em} + C_{buf}$ and the four thermal transmittance account for heat transfer between (i) walls and indoor air $u_{bi}$ , (ii) walls and outdoor air $u_{bo}$ , (iii) emitters and indoor air $u_{em}$ , and (iv) air ventilation between indoor and outdoor $u_{v}$ .

Here are the equivalent equations:

$T_{ia}^{t+1}=T_{ia}^t + \frac{dt}{C_{ia}}[u_{ia,em}(T_{em}^t-T_{ia}^t)-u_{ia,b}(T_{ia}^t-T_{b}^t)-u_{v}(T_{ia}^t-T_{oa}^t)+g^t]$

$T_{b}^{t+1}=T_{b}^t + \frac{dt}{C_{b}}[u_{ia,b}(T_{ia}^t-T_{b}^t)-u_{oa,b}(T_{b}^t-T_{oa}^t)]$

$T_{em}^{t+1}=T_{em}^t + \frac{dt}{C_{em}}[Q^t-u_{ia,em}(T_{em}^t-T_{ia}^t)]$

The emitters currently available in demod are a radiator system. More details on their sizing and characteristics can be found in [McKenna2016].

Availability

The parameters for the capacitance and resistences are taken from CREST, and they refer to the UK building stock. An updated parameters for the German case will be released in future versions.

Fig. 3 Low-order building thermal model 4R3C¶

Domestic hot water demand¶

Currently in demod there is a module for simulating the demand for domestic hot water.

CREST domestic hot water demand¶

API

Simulation of domestic hot water demand is part of Occupancy-based appliance usage simulator and for details about the implementation you can visit OccupancyApplianceSimulator.

Description

Currently demod simulates domestic hot water demand following the approach of CREST, which simulates the use of water fixtures in the same way as household appliances:

first the number of water fixtures in the house is initialized;
then, the pdf of the activities washing or cooking is multiplied by a calibration scalar, whose value is assigned such that the simulated annual water consumption of each fixture matches a target value;
the water withdrawal event occurs if the probability exceeds a random draw;
finally, when a water withdrawal event occurs, the temperature of hot water and withdrawn volume are determined stochastically.

Availability

This module uses empirical data from CREST, which are derived from a UK study. No equivalent data are currently available for Germany.

Hot water tank thermal behavior¶

Currently in demod, a hot water tank can be simulated using the module BuildingThermalDynamics as a component of regular boiler Heating system controller. Dedicated modules will be released in future versions.

Low-order hot water tank thermal model (1R1C)¶

API

For details about the implementation of this simulator you can visit BuildingThermalDynamics.

Description

This module simulates the thermal behavior of the hot water tank using an equivalent low-order electric circuit as in CREST (see Fig. 4).

The name 1R1C refers to thermal capacitance representing the thermal mass of hot water $C_{tank}$ and the thermal resistences of the hot water tank insulation between hot water and indoor air $u_{tank}$ .

Here is the equivalent equation:

$T_{dhw}^{t+1}=T_{dhw}^t + \frac{dt}{C_{tank}}[Q_{dhw}-m_{dhw}^{t}cp_{dhw}(T_{dhw}^t-T_{dhw}^{in})-u_{tank}(T_{dhw}^t-T_{ia}^t)]$

where $m_{dhw}$ is the hot water mass flow, $cp_{dhw}$ is the thermal capacity of water ( $4.2 \: kJ/kg^{\circ}C$ ) and $Q_{dhw}$ refers to the heat provided from the heating system.

Availability

The parameters for the capacitance and resistences are taken from CREST.

1R1C low-order hot water tank thermal model

Fig. 4 Low-order hot water tank thermal model (1R1C)¶

Heat demand¶

These modules convert the demand for domestic energy services such as indoor thermal comfort and domestic hot water withdrawal into heating demand.

Integrated heat demand simulator¶

API

For details about the implementation of this simulator you can visit HeatDemand.

Description

This module estimates the heat demand for domestic hot water heating $Q_{dhw}$ and space heating $Q_{sh}$ in an integrated way.

For calculating the target heat demand required to deliver hot water at the appropriate temperature, the module use the following equations,

$Q_{tank}=\frac{C_{tank}}{dt}(T_{dhw}-T_{tank})$

$Q_{flow}=m_{dhw}cp_{dhw}(T_{tank}-T_{w,inlet})$

$Q_{loss}=u_{tank}(T_{tank}-T_{ia})$

$Q_{dhw}= Q_{tank} + Q_{flow} + Q_{loss}$

where the heat demand for domestic hot water $Q_{dhw}$ is estimated as the sum of three components: (1) $Q_{tank}$ the heat required to warm the entire tank up to the target temperature, (2) $Q_{flow}$ the heat required to heat up the water requested by the users and (3) $Q_{loss}$ the thermal losses of the tank.

To calculate the heat supply required to achieve the comfort temperature of the indoor air, the algorithm aims to keep the temperature of the emitters within the operating range $50^{\circ}C \: \pm 5^{\circ}C$ .

This algorithm is activated when the indoor temperature of the building is equal to or lower than the minimum limit of the the indoor air comfort range $T_{ia,target} - 2^{\circ}C$ . On the other hand, it is deactivated when the latter reaches the maximum limit $T_{ia,target} + 2^{\circ}C$ . In this way, typical alternating pattern of the heating system operation and oscillating temperature profiles are observed.

$Q_{sh}=C_{em}(T_{em,target}-T_{em})+u_{em}(T_{em}-T_{ia})$

Compatibility: This module is flexible and allows to use alternative comfort temperature and heating switch on profiles. Once these profiles are generated or empirically measured, they can be given as input to the thermal building model to estimate the heating demand.

Heating system controllers¶

This section presents some modules for controlling the heating system and its components.

Thermostats¶

API: For details about the implementation of this simulator you can visit Thermostats.
Description: The thermostat monitors the temperature of a thermal component and accordingly sends an ON or OFF signal to the heating system: once the temperature of a component reaches the minimum boundary, which corresponds to its target temperature minus a deadband, the thermostat is set on ON; on the contrary, if the temperature reaches the maximum, the thermostat is set to OFF.
Compatibility: Any thermal component, for which a target temperature and deadband are defined, can be controlled by this module.

Heating system controller¶

API

For details about the implementation of this simulator you can visit SystemControls.

Description

This module simulates an integrated heating system with a timer and thermostat, checking which controls should be sent to the HeatingSystem. It allows to manage in an integrated way the supply of heating for domestic hot water and space heating, prioritizing the first and ensuring that the heating system works within the recommended operating conditions.

This unit takes the indoor temperature of the building as input and compares it to thermostat setting to estimate the space heating thermal demand. Moreover, thanks to the temperature monitoring of the emitters, the controller avoids that they can reach temperatures higher than the safety temperature of 55 °C.

There are currently two heating system configurations available for which the control system uses two different control methods: ‘combi’ boiler or normal boiler.

Combi boiler or combination boiler is both a water heater and central heating boiler in a single unit. Combi boilers heat water directly from the mains when households turn on a tap, so a hot water storage cylinder is not required.

Regular boiler provides heat both for space heating and domestic hot water, too. However, the hot water system is connected to a separate hot water cylinder, which allows hot water to be stored without the need for the heating system to be activated every time households turn on a tap.

Compatibility

this module is currently compatible with the heating system implemented in HeatingSystem.

Heating systems¶

Currently demod implements a set of heating systems, following the approach developed in [McKenna2016].

CREST heating system¶

API

For details about the implementation of this simulator you can visit HeatingSystem.

Description

It simulates the energy consumption (i.e., gas or electricity) of the heating system for providing the requested heat demand.

The algorithm used here is relatively simple and estimates the consumption of electricity or gas $m_{fuel}$ on the basis of the nominal fuel flow rate $m_{fuel,n}$ , the heat demand $Q_{th}$ and the heat supply at nominal conditions $Q_{th,n}$ , as follows:

$m_{fuel}= m_{fuel,n} \frac{Q_{th}}{Q_{th,n}}$

where the heat supply at nominal conditions is estimated using the fuel calorific value $CV$ and the thermal efficiency $\eta_{th}$ ,

$Q_{th,n} = m_{fuel,n} {CV} \eta_{th}$

If the heat requested is larger than the maximum heat providable, the heat outputs will be capped.

For more details on the data used and the different system and fuel options available, you can refer to CREST.

Compatibility

For the moment, simple heating system are implemented. Regular gaz boilers and a simple heat pump model are available.

Integrated heating demand and supply¶

The modules in this section combine a set of the modules presented above and simplify their use. In this way, all links between modules are already implemented and only one module needs to be launched to calculate heating demand and supply.

CREST five modules heating simulator¶

API

For details about the implementation of this simulator you can visit FiveModulesHeatingSimulator.

Description

This module estimates domestic heating demand and supply. It is based on the CREST model, but simplifies the operation of thermostat control setting by the users.

As shown in Fig. 5, the following 5 components are integrated in this module:

CREST heating system:
Heating system controller
Integrated heat demand simulator
Thermostats
Low-order building thermal model (4R3C) and Low-order hot water tank thermal model (1R1C)

In Fig. 5, the modules and their connections are shown schematically.

Compatibility

This simulator is also compatible with external simulated components.

For instance, the desired indoor temperature can be passed in the step method through external_target_temperature

Fig. 5 CREST integrated heating demand and supply model¶