Brought to you by
Brought to you by
Which parameter can be used as a suitable indicator for the sustainability of buildings from an energy standpoint? The so-called PER factors (Primary Energy Renewable) were first introduced in the final presentation at last year’s International Passive House Conference, as future-oriented sustainability assessment criteria [Feist 2014]. With the new release of the Passive House Planning Package (PHPP version 9), PER is being introduced as a worldwide certification criteria for Passive Houses. This article describes the methodology used to derive the factors integrated into the PHPP.
The methodology used to derive PER factors is based on the ideas that have previously been published in [Feist 2013] and [Feist 2014]. The same approach was further developed, applied and analysed internationally. With an hourly resolution load profiles of the energy demand are simulated in the context of a future scenario - where the energy is supplied solely by renewable energy (RE) sources, including all necessary storage facilities (Figure 1). The individual calculations are based on climate data from various sources, the resulting PER factors describe how much more renewable energy must be supplied in order to cover the final energy consumed at the building, including all losses incurred along the way.
The PER factor is determined by the simultaneity of available energy resources and the energy demand, as this dictates how much energy needs to be temporarily stored before it is used. Short-term storage can technically be achieved fairly efficiently, whilst longer term seasonal storage will always cause higher energy losses. Depending on the load profile, the energy demand will partially be covered (a) directly from the renewable supply, (b) with energy that has temporarily been buffered, or © with energy from a seasonal storage. As a logical consequence of the temporal correlations, heating – which occurs only during seasons with low RE availability – is highly energy intensive in the envisioned future supply chain. For cooling, on the other hand, a larger proportion of the associated energy demand can be used directly without need for temporary storage and losses. Load profiles that occur throughout the year (e.g. domestic electricity use, hot water) lie within these two extreme scenarios.
The modelled future energy supply network is based purely on electricity from renewable sources. Three different sources are taken into account: Photovoltaics, wind turbines and hydropower. Biomass needs to be treated differently in the calculations as it can easily be stored and used on a demand basis. In fact, all systems based on secondary energy (e.g. district heat) are taken into account independently of the supply network, directly in the PHPP, with appropriate parameters for the respective system.
The hourly electricity production through solar power is calculated based on a photovoltaic system oriented toward the equator. The power output is calculated based on the solar radiation information from the respective climate data set, taking into account a temperature-sensitive efficiency of the photovoltaic module. The model’s inclination is determined such that the highest possible annual energy yield is reached at the considered location.
The hourly electricity produced through wind energy is calculated based on a smoothed profile of hourly wind velocities. The same climate data set is used, in order to take account of local climatic correlations between wind and radiation or temperature. However, such localised wind data is rarely representative of the potential of wind energy in the surrounding region. The original wind velocity data is therefore calibrated based on long-term measured averages of the region [SSE] and extrapolated to a hub height of 150 m. The actual power output then depends on the turbine’s power curve, which is modelled with a specific power output of 380 W/m² and 200 W/m² for regions with strong and weak wind, respectively (based on [Mono et.al. 2014]). If the average wind speed at hub height is lower than 4 m/s, a significant development of wind energy in the area is unlikely for economic reasons. The contribution of wind energy to the total mix in the calculations is therefore limited to 0.5 % of the annual yield. Offshore wind energy is not considered at this stage – the calculations are slightly on the safe side in this regard.
Last but not least, the electricity produced from hydropower is taken into account based on the predicted contribution of this energy source to each country’s future total energy demand. The prediction for future hydropower generation is taken from [intpow 2009]. The future electricity demand is scaled for each country according to current and future population densities, current electricity consumption per capita and the vision of a “2000 Watt Society”. Within Europe it is assumed that countries with a hydropower surplus export part of this electricity. The profile of the hydropower electricity is fairly constant throughout the year and takes into account effects of rain, snow and glacial runoff.
At the receiving end we differentiate between five different load profiles for different types of energy consumption: household electricity, domestic hot water, heating, cooling and dehumi¬dification. Examples are shown in Figure 2.
The load profile of the household electricity (20 kWh/(m²a)) is based on a standard profile of the Federal Association of the German Energy and Water Industry (BDEW). For domestic hot water (approx. 15 kWh/(m²a)), an hourly consumption based on [Recknagel 2009] is used. The energy demand is then calculated based on the required temperature rise, with a cold water temperature based on the ground temperature of the local climate data.
The remaining profiles – heating, cooling and dehumidification – are determined with simpli¬fied dynamic simulations of a reference building in the respective climate. Several useful energy profiles are calculated with different building specifications as to account for variations in the heating and cooling period: A typical Passive House, as well as a building with up to eightfold of the energy demand. A heat pump is assumed as heating and cooling system to calculate the electrical load profile (end energy demand).
All load profiles are smoothed ±4 hours in order to account for variations caused e.g. by different user behaviour.
In order to determine the PER factor, the respective load profiles of electricity production and electricity consumption are each scaled to 100 kWh/a and compared on an hourly basis. Direct electricity consumption is assumed to the extent of supply and demand matching up. If the supply exceeds the demand, any excess electricity is fed into the storage facilities. Accordingly, any surplus energy demand during times of low RE availability, is covered with energy taken from the storage. The short-term storage is first used to its full capacity before filling or tapping the long-term seasonal storage. Figure 3 shows an example of the RE electricity load profile and simultaneous electricity demand, as well as the corresponding storage level of the short-term storage.
The short-term storage facilities as part of the power grid (e.g. pumped or other mechanical storage and/or batteries) are assumed to have a capacity of 0.1 kWh for each 100 kWh/a energy demand. The storage efficiency is modelled with 70%. Additional capacity is added for the storage potential of domestic hot water tanks (20 kWh thermal storage, η = 95 %), as well as for the thermal mass of the building (22 kWh thermal storage, η = 90 %). These additions are not scaled according to the demand i.e. a building with higher energy consumption will have less storage buffer relative to the peak load and total demand.
It is further assumed that the seasonal storage has the required capacity to store exactly the amount of energy storage required over the course of the year (supply = demand + losses). One possibility of a working seasonal storage system is the conversion of RE electricity into methane, for which a conversion efficiency of 57 % is assumed. The re-conversion from gas into electricity in a CCTG plant is modelled with an efficiency of 50 %. Electricity consumed via the seasonal storage, therefore has an overall efficiency of only approx. 30 %. Finally, 5 % distribution losses are added for all electricity transmission via the electrical grid.
The hourly RE profile depends on the proportion of solar and wind energy in the mix, in addition to the fixed percentage of hydropower. Both extremes (no solar or no wind), as well as four intermediate steps are modelled, which results in six calculations for the demand profile. The PER factor is determined based on the most favourable combination of wind and solar energy.
It is not possible to analyse the demand profiles independently, as the mix of different energy usages is decisive for the overall load profile and thus dictates the simultaneity with the energy supply. The PER algorithms are based on a baseline consumption with an incremental increase of the energy consumer profile in question. The increase ranges from zero to eight times the assumed typical PH load – including an increase in the heating/cooling period length for at a respectively higher annual energy demand. The baseline consumption is made up of all the remaining energy applications and represents the expected demand distribution / grid load over the course of the year under the given climatic conditions.
For each of the load profiles the PER routine calculates the required RE supply to cover the total energy demand, plus all storage losses. The PER factor for the energy application in question then equates to the slope of the changing RE supply over the increase in energy demand (examples are shown in Figure 4). In some cases this can lead to factors below one, which would mean that less energy needs to be additionally generated than will be consumed. This is the case only if the additional energy demand balances out seasonal disparities and thus reduces the need for seasonal storage, e.g. additional cooling in a heating dominated climate.
Figure 5 shows example PER factors for selected climates under very different climatic conditions (all with comparatively small hydropower contribution). The results for household electricity and domestic hot water don’t vary much, with typical values around 1.3 – i.e. 30 % more RE electricity needs to be supplied than can actually be used at the building. The factors for heating, cooling and dehumidification are more strongly influenced by the given local climatic conditions. Unfortunately, direct correlations with a climate zone are not given as the RE availability and concurrent energy demand differ worldwide – solar radiation, wind speeds and, of course, the availability of hydropower. Generally speaking, in climates where there is a heating demand, the primary electricity requirements for heating tend to be noticeably higher than for cooling. Locations with more hydropower potential in most cases feature lower PER factors, as this RE source is fairly constant throughout the year.
The methodology to derive PER factors, as described in this article, to begin with are only valid for the specific climate data set used. Calculations for the very same location but a different climate data set (e.g. different time period, different source) will lead to slightly different results. Furthermore, these calculations are purely local, meaning that the influences of RE generation in the nearby surroundings is not at all taken into account. In reality, electricity production and electricity consumption cannot be viewed as strictly local but must be seen in a regional context. Electrical grids are in many ways influenced by politics and developments, locally and worldwide, cannot be reliably foreseen. However, it is clear that a purely local energy supply, though technically possible, is needlessly complex and therefore an assumption that is too pessimistic.
The PER factors to be used in the PHPP are thus not based on individual local calculations but rather on a combination via a global Fourier approximation of the results calculated for over 700 locations worldwide. In addition, the minimum value used in the PHPP is 1 (supply = demand). Figure 6 shows the average value and variation of the PER factor for space heating of all locations currently integrated into the PHPP.
The assessment based on Renewable Primary Energy (PER) is a s new concept, which the PHPP user and Passive House designer will need some time to get used to. It is important to understand that the change does not affect the core definition of the Passive House in any way – the requirements for the useful energy demand remain unchanged. The new PER factors are merely an additional, highly effective and descriptive approach for assessing the overall sustainability of a building - including the used technology - from the energy supply perspective.
The explanations in this article describe core aspects of the methodology with which the PER factors were derived. It aims to promote a better understanding of the correlations and the meaning of these new values. Most important is the understanding the PER factors essentially simply indicate how much more primary electricity needs to be generated to cover the demand at the building. They directly represent the required resources for a given energy application, or more precisely, the renewable energy resources.
The results clearly illustrate that a high level of efficiency of heating is more important than for cooling when looking at using resources responsibly and sustainably - the reason being the simultaneity of RE availability and energy demand over the course of the year.
[Feist 2013] Feist, W.: Energy concepts – the Passive House in comparison. In: Conference Proceedings of the 17th International Passive House Conference, Frankfurt, April 2013.
[Feist 2014] Feist, W.: Passive House – the next decade. In: Conference Proceedings of the 18th International Passive House Conference, Aachen, April 2014.
[intpow 2009] World Hydro Potential and Development. intpow, Norwegian Renewable Energy Parnters 2009
[SSE] Monthly average wind data at one-degree resolution of the world. These data were obtained from the NASA Langley Research Center Atmospheric Science Data Center Surface meteorological and Solar Energy (SSE) web portal supported by the NASA LaRC POWER Project.
[Mono et.al. 2014] Mono, R., Glasstetter, P., Horn, F.: Ungleichzeitigkeit und Effekte räum¬licher Verteilung von Wind- und Solarenergie in Deutschland. Eine Untersuchung der 100 % erneuerbar Stiftung. April 2014.
[Recknagel 2009] Recknagel, Sprenger, Schramek: Taschenbuch für Heiz¬ung und Klima¬technik; 74. Auflage, München, Oldenbourg Industrieverlag 2009.