One of the key input parameters for any geothermal simulation is the heating, cooling and domestic hot water demand of the building. While these values are sometimes known based on detailed simulations or measurements, more often than not, especially in the early stages, they need to be estimated. This chapter sheds some light on this topic and explores several approaches to addressing this challenge.
Before we start this chapter, we would like to address some important terminology. When talking about the building demand, the terms ‘heating’, ‘cooling’ and ‘domestic hot water’ are used. Given the efficiency of the heat pump (which we will discuss in the next chapter), this gets converted into the ‘extraction’ and ‘injection’ load, which is the ground load.
Thermal building demand
Within GHEtool, various types of load profile can be used. First, there is a distinction between hourly and monthly resolutions, with 8,760 and 12 timesteps per year, respectively. In addition to the resolution, the load is usually specified for one full year, based on the assumption that each year of the simulation period is identical. However, if this is not the case — for example, if your project is phased — you can use multi-year input.
The next sections will discuss both the monthly and hourly resolutions.
Monthly resolution
If no hourly data is available, a monthly simulation can be performed. Four key inputs are required for this: the peak power for heating and cooling, and the annual energy demand for heating and cooling. If domestic hot water demand is included, a fifth input is required. Each of these is discussed in more detail below.
Peak power for heating
The heating peak corresponds to the maximum output of the building’s heat pump. If the heat pump has a capacity of 10 kW, the building cannot draw more than this, so this value should be set as the heating peak. Depending on regional building standards, the capacity of the heat pump is usually determined by a static heat loss calculation based on a reference outdoor temperature (e.g. –8 °C in Belgium).
The situation is sligthly different when modulating heat pumps are used. Let’s say the calculated static heat loss of a building is 7 kW. An installer might choose to install an 8 kW heat pump to allow for some oversizing. However, this would lead to an oversized borefield, since the borefield must be able to cope with the peak demand. If the system uses a modulating heat pump, you can often set a lower peak power in the heat pump controller — for example, limiting the 8 kW heat pump to operate at a maximum of 7 kW — thus avoiding unnecessary oversizing of the borefield. The same reasoning applies if the emission system (e.g. radiators or underfloor heating) cannot utilise the full capacity of the heat pump.
Early stage
If the specifications of the heat pump are not yet known, the peak heating demand can be estimated using established rules of thumb. These are based either on the capacity of the emission system or on general guidance relating to the building’s type and age. A sample table showing typical values is provided below. However, be aware that these figures are region-specific and will need to be adapted to your local context.
| Type | Emission power in heating | Emission power in cooling |
|---|---|---|
| Floor heating | 40-80 W/m² | 15-25 W/m² |
| Climate ceiling | 30-55 W/m² | 25-50 W/m² |
| Wall heating | 30-70 W/m² | 25-60 W/m² |
| Fan coil unit (non-condensing) | 200-2000 W | 250-500 W |
| Fan coil unit (condensing) | 200-2000 W | 1000-2000 W |
(Data obtained from the Cooling 2.0 project)
| Type | Heat demand |
|---|---|
| Residential building (after 2002) | 45 W/m² |
| Office building (old) | 89 W/m² |
| Office building (new) | 59 W/m² |
| School (old) | 109 W/m² |
| School (new) | 60 W/m² |
| Retail (old) | 56 W/m² |
| Retail (new) | 54 W/m² |
(Data obtained from nPro, values for the climate of Berlin)
Simultaneity
When multiple heat pumps are connected to a central, shared borefield, the resulting peak power is not simply the sum of the individual units. Instead, a simultaneity factor must be applied. The problem of simultaneity and its solutions will be discussed later in this course.
As well as the peak power itself, the duration of this peak is also important. When working with an hourly load profile, this is embedded in the profile itself. However, in the case of a monthly load profile, it must be specified explicitly. The peak duration can be defined as the longest period in which the heat pump works at maximum power. For example, if the heat pump switches on and runs at maximum power for eight hours straight, the peak duration is eight hours.
Peak power for cooling
The approach to defining the peak cooling power depends on whether the borefield is actively designed for cooling or whether cooling is considered a secondary benefit, a ‘nice to have’.
Design for cooling
In warmer climates, where cooling plays a more significant role than heating, the borefield is typically limited by the peak injection temperature. In such cases, the cooling demand is usually determined in accordance with local building regulations, taking into account factors such as the area of glazing, the g-value and the U-value. The building’s cooling demand is then derived in a similar manner to the heating demand, based on these calculations.
Nice to have
In regions where cooling is not a primary concern, the emission system is usually not designed to ensure thermal comfort in summer. Here, cooling is often treated as a ‘nice to have’ feature — an added benefit of installing a geothermal borefield. In such cases, peak cooling power is typically based on the capacity of the emission system, which was originally designed for heating. Refer to the table of typical emission system capacities provided earlier.
Early stage
In the early stages of a project, the peak cooling demand, like the heating demand, can be estimated using basic rules of thumb. This can be done by either drawing on the emission system’s capacity or by using values from similar reference buildings.
Energy demand for heating
While peak power is one important aspect of a building’s energy profile, the other key parameter for borefield design is the annual energy demand for heating. This can be estimated in several ways, such as by using full load hours, rules of thumb or heating degree days.
Full load hours
Using full load hours is a simple way to estimate annual energy demand based on known (or estimated) peak power. If a system operates at peak load for x hours per year, the total energy demand is given by the following formula: $$energy = peak \cdot FLH$$The table below provides indicative full load hour values for various building types.
| Type | Full load hours |
|---|---|
| Nursing home | 1300-1900 |
| Hospitals | 1500-2000 |
| Offices | 900-1600 |
| Schools | 800-1300 |
| Residential | 1200-1500 |
| Others | 1000-2000 |
(Data obtained from SenterNovem, Cijfers en Tabellen 2007)
Rule of thumb
As with peak power, the annual heating demand can be estimated using empirical values based on floor area. The table below provides typical heat demand values for buildings in the Berlin climate region.
| Type | Heat demand |
|---|---|
| Residential building (after 2002) | 72 kWh/m² |
| Office building (old) | 125 kWh/m² |
| Office building (new) | 65 kWh/m² |
| School (old) | 120 kWh/m² |
| School (new) | 60 kWh/m² |
| Retail (old) | 95 kWh/m² |
| Retail (new) | 65 kWh/m² |
(Data obtained from nPro)
Heating degree days
Another method of estimating heating demand is to use Heating Degree Days (HDDs). HDDs quantify the extent to which, and for how long, the outside air temperature is below a certain base temperature, referred to as the balance point temperature. Below this temperature, the building requires heating. The total HDD is the sum of the differences between the base temperature and the actual outside temperature each day during the heating season.
Compared to FLH, HDD-based methods offer a more refined estimate because they account for building characteristics (e.g. insulation and solar gain) and climate conditions separately.
Energy demand for cooling
The same principles that apply to heating also apply to cooling. If detailed hourly data is unavailable, cooling demand can be estimated using either full load hours (FLH) or a rule of thumb.
Full load hours
For Belgium’s climate, the typical FLH value for cooling ranges between 500 and 1,000 hours. Once the peak cooling power has been determined, the energy demand can be calculated as follows:
$$energy = peak \cdot FLH$$
Rule of thumb
Alternatively, you can use the benchmark value for annual cooling demand per square metre. The table below provides these values.
| Type | Service sector | Residential sector | Average |
|---|---|---|---|
| Austria | 83 kWh/m² | 38 kWh/m² | 49 kWh/m² |
| Belgium | 50 kWh/m² | 23 kWh/m² | 28 kWh/m² |
| Germany | 74 kWh/m² | 33 kWh/m² | 46 kWh/m² |
| The Netherlands | 37 kWh/m² | 16 kWh/m² | 22 kWh/m² |
| Spain | 130 kWh/m² | 59 kWh/m² | 69 kWh/m² |
(Data obtained from the Heat Roadmap Europe)
Cooling degree days
Just as heating degree days (HDD) can be used to estimate heating demand, cooling degree days (CDD) can be used to estimate cooling requirements. CDDs are calculated based on the difference between a balance point temperature (usually 18°C) and the actual outdoor temperature whenever the latter is higher.
Two empirical formulas, developed by the European Commission (ENER/C1/2018-493, doi: 10.2833/158083) are
For space cooling in the residential sector:
$$FLH=96+0.85\cdot CDD$$
For space cooling in the tertiary sector:
$$FLH=475+0.49\cdot CDD$$
Energy demand for DHW
An increasingly important parameter in borefield design is the demand for domestic hot water. In most residential contexts, the typical value is around 1000 kWh per person per year. However, this figure can be significantly higher in buildings such as hotels and hospitals.
Hourly resolution
Designing a borefield using an hourly resolution will provide the most accurate results, as peak powers (and their durations) no longer need to be estimated. This has the advantage that complex problems such as the combination of active and passive cooling, and simultaneity (both of which will be discussed later in this course), are automatically overcome. Typically, hourly data is generated from dynamic building simulations (using software such as IESVE, DesignBuilder, VICUS Buildings, IDA ICE, etc.), or, in the case of a renovation, may be based on measurement data.
Create hourly load profiles in GHEtool
Since hourly load data adds considerable value to the design of borefields (this will become clear later in the course), but is often unavailable, GHEtool has been equipped with a method to create an hourly load profile. Inspired by the degree day method, this method can be used to scale up your estimated peak heating and cooling powers, as well as your yearly energy demands, to an hourly profile based on outside temperatures. The method involves three consecutive steps.
- By starting with a weather file and an initial threshold temperature, above which heating begins, an hourly profile is obtained. The higher values occur at the hours when the temperature is lower, and therefore the difference between the temperature and the threshold is the greatest. It is at these moments that we also expect the highest peak demand. The same reasoning can be applied to the cooling demand.
- Next, this hourly profile is scaled with the yearly energy demand (which is an input) to create an hourly load that has the same yearly demand as the building.
- Finally, the peak power should be checked. If the peak power of the profile is different from the peak demand of the building, the temperature threshold is adjusted so that the heating starts earlier or later, and the second step is repeated.
Conclusion
In this chapter we discussed the building demand and how it can be estimated in the early stages, using heating degree days or rules of thumb. A distinction was made between a monthly load and an hourly load and a methodology to create an hourly load was introduced. Since up until now, all the demands were building demands, an efficiency is required to translate it into ground loads. This is the subject of the next chapter.
Questions
References