Hafencity Universität, Resource Efficiency in Architecture and Planning, Hamburg, Germany gustavo.siqueira@hcu-hamburg.de
Udo Dietrich
Hafencity Universität, Resource Efficiency in Architecture and Planning, Hamburg, Germany udo.dietrich@hcu-hamburg.de
ABSTRACT: In the recent years great amount of research has been undertaken in the fields of thermal comfort indoors. It is largely accepted that in natural ventilated buildings where occu-pants are allowed to directly influence their environment, the comfort expectations tend to fol-low outdoor conditions. The adaptive comfort model relies on these observations and generated two major standards: EN 15251 focused on Europe and the ASHRAE 55-2004 worldwide ap-plicable.
The first part of this paper introduces the ASHRAE 55-2004 and its background field research.
The second part discusses the vantages and disadvantages of dividing the available comfort lim-its by climate groups for the benefit of climate responsive design. The last part presents evi-dence for a more suitable comfort limits' description through a single-curve function instead of the available braked-in-two-seasons linear model.
Fig. 1 – Comparison between two adaptive comfort limits in the major comfort standards (De Dear 2011) ture. Both of the comfort models have been developed on the base of research in office build-ings, however they are intended to be applied to all sedentary activities including dwelling.
Despite the different calculation methods both standards present a comfort zone divided into two seasons: the naturally ventilated (or “free-running”) and the heated or cooled season. Dur-ing the naturally ventilated season the comfort zone’s pattern is inclined whilst the heated or cooled season presents a flat profile. Note in Figure 1 the similarity of the resulted formulae.
(For more detailed description see Humphreys et al 2010, Nicol & Humphreys 2010, and De dear 2011).
3 HYPOTHESIS
This paper works on two hypotheses valid only for naturally ventilated buildings, in which in-door environment is influenced by outin-door climate:
1. Since seasonal changes are fluid and do not suffer abrupt changes it is probable that a curve would be more adequate to represent the comfort function than the two lines model of both existing adaptive comfort models.
2. As stated by the adaptive theory, people tend to adapt their comfort expectations ac-cording to given conditions. Therefore, it is sensible to believe that people in different climate types present different comfort patterns.
4 METHODS
The proof of the first hypothesis begins with analysis of the RP-884 database.
This database is available online in a group of separated files classified in many categories such as: location, building type, year and climate type. The given climate classification was substituted by the more widespread Köppen-Geiger classification. Moreover only the major climate groups were used and the cold climates were put together resulting in:
A- humid-hot B- arid-hot C- temperate D- cold
The aim of this paper is to evaluate the comfort patterns of users who are used to adapt them-selves dynamically to climate variations, hence only the data from naturally ventilated buildings
were considered. This explains the absence of the D category in the experiment, as none of the buildings in research projects located in the cold zone (D) were naturally ventilated.
Each file in the raw data represents a large amount of parameters, and the most important were: user’s comfort vote (in a 7-point scale ranging from -3 “too cold” to +3 “too hot”), in-doors temperature (real time), outdoor temperature (day mean) and air speed.
To find the comfort temperature only the votes between -1 and +1 were considered and represented in combination with the indoor temperature in which they were applied. The represented indoor temperature is an adjusted operative temperature, which is the given opera-tive temperature, after the influence of air speed upon comfort sensation has been reduced. This reduction is described in the following formula (EN 15251).
⎟⎠
V=Air speed, θin=adjusted indoor temperatur, θi=measured operative Temperatur (raw database) The comfort temperature is commonly represented at the y-axis.
The x-axis is used to represent the outdoor temperature. In this case, it is based upon the rounded mean outdoor temperature recorded on the raw file. The rounding helped to cluster the comfort votes in 1K intervals and to form a grid. Thereby it is possible to define, for each out-door temperature value, the mean comfort vote and the standard deviation.
Subsequently, as shown in Figure 2, the mean comfort curve is calculated as a polynomial re-gression upon the mean comfort votes. The mean standard deviation is used to define the com-fort range, i. e. the comcom-fort limits above and below the mean comcom-fort curves.
An enhancement (Dietrich 2010) was the substitution of the polynomial functions by Arc Tan functions. It came to be very useful since, unlike the polynomial functions, which started to change the direction as the local maximum and local minimum are achieved, the Arc Tan curves just change the direction twice.
Fig. 2 - The different patterns for the three main climate groups: A, B, C.
The resulted formulae for A:
[ ]
⎟⎟⎠⎞Where: θi= comfort temperature, θo=mean outdoor temperature.
5 DISCUSSION
After the comfort limits for the three different climate groups have been defined as shown be-fore, they can be used for an evaluation of the thermal performance during the design process.
This process is presented schematically in Figure 3 and will be described in the following.
The first step would be the climate classification itself into the climate groups A, B, C or D (Fig. 3, #1). The second step is the linkage between climate and the basic design statements.
These design statements refer to the Mahoney tables (Königsberger et al 1971) and to Eproklid (de Siqueira 2010). The direct linkage to the different climate zones has been described less ex-plicitly than in the Mahoney tables, however it is adequate for its present purpose, being only a quick start-up for further optimization. (Fig. 3, #2)
Subsequently the indicted design strategies are related to appropriate passive conditioning strategies as shown by Givoni (1998) (Fig. 3, #3).
At this point follows the first evaluation due to a thermal simulation combined with the pro-posed comfort limits discussed in paragraph 4. (This simulation requires detailed definitions of parameters such as air change ratio, internal gains, user’s profile etc., which could be object of another research.) If the result of this evaluation turns out as “hot” or “cold”, either the available design and the passive strategies can be optimized, or, if necessary, one of the low energy con-ditioning strategies introduced by Givoni (2011) can be applied (Fig. 3, #4).
If the next evaluation shows that the comfort state can be achieved by no means, an active conditioning strategy has to be chosen (Fig. 3, #5).
In the following the process itself will be described for each of the three climate groups.
As the outdoor climate in the A-group (humid-hot) usually stays close to acceptable comfort limits it is basically necessary to prevent extra thermal stresses. Therefore sun protection and light materials are used to avoid thermal loads. The large opening ratio required allows perma-nent cross ventilation as a passive conditioning strategy. Those recommendations guarantee the enhancement of the comfort. If necessary, an indirect evaporative system can be used as low energy conditioning. Givoni and Gonzales showed positive results in an experiment in Maracai-bo, Venezuela in these climatic conditions (Givoni 2011).
For the B-group (arid-hot), as both annual and daily swings are usually large, it is reasonable to combine high thermal mass with good sun and wind regulation. In the cold season sun radia-tion is allowed to warm up indoors and daytime ventilaradia-tion is preferred, whilst in the hot season sun radiation should be completely avoided and nocturnal ventilation prevails. These passive conditioning strategies could be complemented with direct or indirect evaporative cooling sys-tems, if comfort state cannot be achieved in the hot season.
For the C-group usually similar strategies as for the B-Group can be applied except that more thermal insulation and less thermal mass are needed mainly in the coldest zones of this climate type.
Fig. 3 - Schematic presentation of the design process using proposed comfort limits
6 CONCLUSIONS
The paper presents some evidence that support both hypotheses.
Firstly, since all the functions show high level of confidence (r²>0.9), it can be stated that a curve is suitable to describe a comfort function.
Secondly, as the pattern’s differences between the three curves are remarkable, the paper’s second hypothesis is also verified. Note in Figure 4 that the A-function is very short and in-clined whilst the B-function is both longer and flatter.
Moreover, it has been shown that by using the proposed climate comfort limits an effective and practical design-tool can be created, which here has only been explained schematically. The detailed description of this design-tool could be subject for a further research.
Two points are still unclear:
1. The adopted comfort ranges are compromising: 1.5K (A), 3K (B) and 2.7K (C). De Dear (2011) shows how to quantify the enhancement of comfort limits due to elevation of air speed.
It is arguable though, if the comfort ranges are constant parallel to the mean comfort.
2. As the RP-884 is focused on surveys in office buildings, it is desirable to investigate if the patterns originated in other functions e.g. dwellings would diverge.
Fig. 4 - The Arc tan curves overlay for comparison of those patterns
REFERENCES
de Dear, R.J. 2011. Recent Enhancements to the Adaptive Comfort Standard in ASHRAE 55 - 2010 . 45th Annual Conference of the Architectural Science Association, ANZAScA 2011. The University of Sydney.
de Dear, R.J. Brager, G.S. & Cooper, D. 1998. Developing an adaptive model of thermal comfort and pre-ference. ASHRAE RP-884 Final Report.
de Siqueira, G.L. 2010. Entwurfsprozess mit Klimadaten. Masterthesis. Hafencity Universität Hamburg.
Dietrich, U. 2010. Internal discussion.
Humphreys, M.A. 1978. Outdoor temperatures and comfort indoors. Building Research and Practice 6:
92-105.
Humphreys, M.A., Rijal, H.B. & Nicol, J.F. 2010. Examining and developing the adaptive relation be-tween climate and thermal comfort indoors. Windsor, UK.
Nicol, J. F. & Humphreys, M.A. 2010. Derivation of the adaptive equations for thermal comfort in free-running buildings in European standard EN15251. Building and Environment, 45.
Koenigsberger, O.H., Mahoney, C. & Evans, M. 1971. Climate and House Design. New York: United na-tions
Givoni, B. 1998. Climate Considerations in Building and Urban Design. New York: Van Nostrand Rein-hold Co.
Givoni, B. 2011. Indoor temperature reduction by passive cooling systems. Solar Energy 85: 1692-1726.
1 INTRODUCTION
The “20-20-20” strategy for smart, sustainable and inclusive growth defines the energy and en-vironmental targets that the European Union is committed to achieve by 2020 (EC 2012). The targets are: a 20% reduction of EU greenhouse gases emissions compared to the 1990 levels; the reaching of a 20% share of EU gross final energy consumption supplied from renewable energy sources; a 20% improvement in the EU energy efficiency through a 20% reduction of primary energy consumption. The Eurostat statistical data currently available (Eurostat 2012) indicate that in 2010 the levels reached for the three targets are respectively 15.0%, 12.5%, 5.4%. The target of a 20% reduction of primary energy consumption is the most far from reach.
The building sector accounts for 40% of total energy consumption in the European Union (EEA 2012). So the 2020 targets achievement is directly dependent on the energy efficiency ef-forts in this sector. The energy saving potential to 2020 is 27% for the residential building sector and 30% for the services building sector (European Commission 2006). The CO2 emission re-duction potential for the residential and services building sector is 20% in the reference scenario and 25% in the alternative efficient technology scenarios (European Commission 2011a). In 2010 the EPBD Directive (European Union 2002) has been recasted, in order to lead the build-ing sector toward European 2020 targets achievement. The EPBD recast Directive (European Union 2010) establishes that the building energy performance should be expressed by a primary energy index based on Primary Energy Factors (PEF) per energy carrier, which can be derived from national or regional annual average.
The aim of the paper is to show that the PEF can play a central role for leading the building sector towards the European 2020 targets achievement. The results of the building energy per-formance assessment are directly dependent on the PEF values. Therefore they can direct the choices among different energy carriers used to meet the building energy needs.