- Email: [email protected]
An experience on integrating monitoring and simulation tools in the design of energy-saving buildings
Energy and Buildings 40 (2008) 987–997 www.elsevier.com/locate/enbuild
An experience on integrating monitoring and simulation tools in the design of energy-saving buildings S. Flores Larsen a,1,*, C. Filippı´n b,2, A. Beascochea b,2, G. Lesino a,1 b
a INENCO, Universidad Nacional de Salta, CONICET, Buenos Aires 177, (4400) Salta, Argentina Universidad Nacional de La Pampa, CONICET, Spinetto 785, (6300) Santa Rosa, La Pampa, Argentina
Received 11 May 2007; received in revised form 31 July 2007; accepted 2 August 2007
Abstract In this paper we describe the design and thermal behaviour of a bioclimatic Auditorium at the National University of La Pampa, used for teaching activities in Santa Rosa, La Pampa (Argentina). The building was monitored in winter and simulated with SIMEDIF for Windows, a code developed at the Non Conventional Energy Research Institute (INENCO, Argentina). Then, a new project of a similar building was proposed for General Pico city, and the obtained physical model was used to simulate the building under the summer temperatures of the new city. The building was redesigned and passive solar strategies were applied to reduce heating and cooling loads. The final layout and the monitored thermal behaviour of the new building in winter and summer are described. Without additional costc, the new building savings were 50% in heating requirements respect to the conventional layout, and 70% in the requirements of conventional energy for cooling. # 2007 Elsevier B.V. All rights reserved. Keywords: Energy-efficient buildings; Passive solar design; Energy-saving; Thermal simulation
1. Introduction The challenge of reducing the emission of greenhouse gases at local and global levels requires behavioural changes in life styles and energy consumption patterns in people, and the use of more energy efficient production, processing and distribution technologies. The improvement of building techniques is an alternative way to increase the energy efficiency and reduce gas emissions. Passive solar design exploits the building’s orientation, shape, materials, windows, and external landscape, in combination with other energy efficiency strategies, to create a pleasant environment which is less dependent on fossil fuelbased energy. It has been extensively shown that it is possible to build passive houses with very low energy use and to normal costs. Important energy-savings as high as 60% compared to conventional houses have been found in cold [1], tropical [2], Mediterranean [3] and hot summer/cold winter climates [4]. Some successful applications of passive design are social * Corresponding author. Tel.: +54 387 4255489; fax: +54 387 4255489. E-mail addresses: [email protected], [email protected] (S.F. Larsen), [email protected] (C. Filippı´n), [email protected] (G. Lesino). 1 Tel.: +54 387 4255578; fax: +54 387 4255489. 2 Tel.: +54 2954 434222; fax: +54 2954 434222. 0378-7788/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2007.08.004
housing [2,5] and office buildings [6,7], where the energy demand for air conditioning can be significantly reduced. In this context, simulation programs have become important tools to improve building designs and energy consumption. These programs can calculate the thermal behaviour of buildings and change different variables, as the climatic conditions, geometry, materials, etc., to evaluate their thermal response. Thus, a feedback is carried out until an adequate final project is achieved [8]. If experimental data validates the physical model used in the simulation, a deeper insight of the building physics and good energy performance can be achieved, as shown in many research works [3,6]. Nowadays, a wide variety of simulation programs is available, of different complexity levels going from steady-state situation to very sophisticate CFD simulation [9]: TRNSYS [7], ESP-r [10,11], DOE-2, BLAST, Energy Plus [12], TAS [5], FLUENT [13], DEROB-LTH [14] between other common programs, can be mentioned. Methodologies for validation of building energy simulation programs have also been developed [11]. In Argentina, a lot of effort was directed since 1984 to develop a code for the simulation of transient thermal behaviour of passive multiroom buildings. SIMEDIF model and program was made by researchers of INENCO (Non Conventional Energy Research Institute), Argentina. The first version was
988
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
developed for DOS by Casermeiro and Saravia [15] in 1984 and the code has been largely validated since then throughout years of experimental work in Argentina. Numerous groups continuously used it for research [16], design, and simulation of the thermal behaviour of lightweight [17,18] and massive [19] buildings from the desert climate of the Puna (North of Argentina) to La Pampa. Since 2000 the code was adapted for Windows environment [20] and two modules for passive cooling systems were added: a module to calculate earth-to-air heat exchangers and a module for evaporative cooling systems. These models were validated with monitored data on existing buildings with buried pipes [21] and a greenhouse with evaporative panels [22]. This software was used in this paper, so a brief description of the thermal model was included in the related section. As mentioned, the growing demand for air-conditioned buildings and the resultant demand for electrical energy has prompted research into passive cooling, such as the European Union funded PASCOOL programme [23] and recommendations, guidelines and laws for energy-saving in buildings [24]. In Argentina, the local laws are not deeply concerned with lowering energy consumption of buildings. Local building designers have largely ignored passive design strategies, which can effectively reduce the building energy consumption. Only isolated exceptions can be mentioned ([25–27]), mainly architects and researchers of universities, who tried to change this situation and whose efforts are becoming clear since the last years. In La Pampa, the situation is even worse, because there is a lack of non-traditional materials and qualified workmanship. It is very difficult to convince the local authorities that energy conservation devices, passive solar heating, and bioclimatic design are beneficial and can be used without additional cost. On the other hand, designers still lack confidence to apply these techniques in absence of information about the performance of existing bioclimatic buildings in Argentina. In the last years, some effort was made to change this idea, by building energy-saving apartments for low-income students [28], a solar school with buried pipes for earth-to-air heat exchange [29], and a College of Agronomic Sciences (National University of La Pampa) [21], whose design combines different passive solar strategies to reduce up to 50% the natural gas consumption, when compared with conventional buildings. This paper trends to fill these holes and to bring some confidence to designers and construction workers of developing countries. In this context, socio-economic, educational and environmental reasons have driven the design of two Auditoriums with
Fig. 1. Location of Santa Rosa (36.578S, 64.45W, 189 m over sea level) and General Pico (35870 S, 63880 W, 141 m over sea level).
reduced energy consumption, for the National University of La Pampa. The local government insisted that the economic costs were similar to that of a conventional building, which due to local economic reasons is in general low. A big effort was made in the building design in order to accomplish this requirement. Both buildings are located in the center of La Pampa province, in Santa Rosa and General Pico cities, in a temperate semi-arid agricultural region of central Argentina (Fig. 1). The climatic data for summer and winter are summarized in Table 1 for both cities. According to the Olgyay’s bio-climogram, people could take advantage of solar radiation and reach the comfort area in 60% of the annual period. Passive solar systems, thermal inertia, natural ventilation, thermal insulation, external shading, building orientation and a compact design can be used to improve the thermal comfort along the year. The first building was designed and built between 1998 and 2000 in Santa Rosa city, where the climate is cold-temperate. It was monitored and simulated during a winter period to evaluate its thermal behaviour and to obtain a physical model of the building. Then, the possibility of a new similar project in General Pico city arose, where the winter conditions are more temperate (1204 vs. 1545 heating degrees days) but the cooling loads in summer are higher than in Santa Rosa (473 vs. 128 cooling degrees days), as shown in Table 1. The previous experience on Santa Rosa’s building was used and the simulation of the building under the summer temperatures of the new city was done. When overheating was confirmed, the redesign of the building was afforded in order to load the heating and cooling energy requirements. In the first part of this
Table 1 Meteorological data for Santa Rosa and General Pico, La Pampa, Argentina (provided by the National Meteorological Service) Winter
Summer
Temperatures (8C) (June)
Santa Rosa General Pico
Mean minimum
Mean
Swing
1.8 2.7
8.2 8.9
11.9 6.6
Heating degrees days base 18 8C 1545 1204
Temperatures (8C) (December) Mean maximum
Mean
Swing
30.3 30.9
22.3 23.4
14.9 14.6
Cooling degrees days base 23 8C 128 473
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
989
Fig. 2. Plan and section of the building in Santa Rosa, location of the temperature sensors and subdivision of the Auditorium into three thermal zones.
paper we describe the original design of the Auditorium in Santa Rosa. Then, a short description of SIMEDIF program and an analysis of the experimental and simulated indoor temperatures for a winter period are presented. The second part describes the summer simulation of the same building in General Pico, and the passive and hybrid design strategies applied to the building in order to deal with overheating and to reduce heating loads. Finally, the monitored winter and summer thermal behaviour of the new building are analysed, and conclusions for future applications and research are presented. 2. First project: Auditorium in Santa Rosa 2.1. Brief description of the building The bioclimatic building of Santa Rosa includes an Auditorium for 200 people, an access hall, and services (with two bathrooms and a little storeroom), with a total covered area of 316 m2. Fig. 2 shows a plan of the building and a vertical section of the Auditorium. The Auditorium is occupied typically by 150–200 students from 8:00 to 13:00 h and 14:00 to 20:00 h, so a quick response of the indoor temperature to the heating system in winter is desirable, together with a reduction of heating and cooling loads. Thus, the building is designed as a lightweight insulated structure and the following strategies are applied: Heating season: the transmission and infiltration heat losses are minimized. The infiltration heat losses are prevented by using hermetic carpentry. Windows have double glazing and walls and roof are well insulated. The external walls are composed of three layers (Fig. 3), from outside to inside: a waterproof cement/sand plaster layer on a 0.18 m thick brick, a 0.02 m thick cement/sand plaster layer, a 0.02 m of glass wool with black veil layer, a 0.03 m of glass wool with Kraft paper layer, and a pine tongued and grooved board. The Rvalue of the wall is 2.13 m2 8C/W. The south-facing wall has an additional layer of thermal insulation (0.025 m of expanded polystyrene) on its internal side, as shown in the picture of Fig. 3, and its R-value is 2.8 m2 8C/W. Higher R-
values imply thicker walls and/or thicker insulation, resulting in higher construction costs and longer return times of investment. In this case, thicker insulation is not recommended: much saving is not obtained because in these sites, winter outdoor temperatures are not as cold as in developed northern countries, where 0.1–0.3 m thick insulation are common. The indoor partition walls are each composed of a 0.2 m thick brick wall and a cement/sand plaster layer on one side. The access hall and services have slab roofs with thermal insulation (0.05 m of expanded polystyrene), while the Auditorium roof has a sloped 0.0007 m galvanized steel sheet (Fig. 3) with thermal insulation (a 0.03 m thick glass wool layer and a 0.04 m thick expanded polystyrene layer). The Auditorium ceiling finish is a pine tongued and grooved board, and the services and access hall ceilings are finished by a 0.05 m thick gypsum plaster with emulsion paint. The floor slab is 0.2 m thick reinforced concrete and it is finished with a
Fig. 3. Wall and roof scheme (left) and southern wall picture (right) with additional insulation: (1) pine wood; (2) glass wool with black veil (0.02 m); (3) glass wool with Kraft paper (0.03 m); (4) 0.18 m thick brick layer with waterproof cement/sand plaster on both sides; (5) brick laid on edge with waterproof cement/sand plaster layers on both sides; (6) wood tie beam; (7) wood board; (8) glass wool with Kraft paper (0.03 m); (9) expanded polystyrene (0.04 m, density: 20 kg/m3); (10) galvanized steel sheet (0.0007 m).
990
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Table 2 Thermal properties of building materials used in SIMEDIF for windows Material
Density (kg/m3)
Specific heat (J/kg K)
Thermal conductivity (W/m K)
Concrete Cement/sand plaster Brick Glass wool Expanded polystyrene Galvanized steel
2400 1184 1800 24 20 2700
653 836 920 920 1420 900
2.16 1.4 0.72 0.04 0.04 202
dark vinyl tile flooring. The properties of the various materials used in the building are summarized in Table 2. An auxiliary heating system was installed, consisting of two gas-fired units of 76500 W each one, with an air flow around 1.67 m3/s and a maximum air velocity in the central duct around 5 m/s, which is the allowed velocity to avoid noise problems. The heating requirements were estimated to obtain an indoor temperature around 20 8C with 30 occupants in winter. The heating units were connected to intelligent indoor thermostats, but during the monitoring described in this paper they were turned on and off manually, to control exactly their functioning periods. Cooling season: to prevent against overheating in summer, direct solar gain through windows is minimized: only one window with a double glass panel (2.70 m 1.5 m, U = 2.0 W/m2 8C) was included in the south wall, to provide natural daylighting and visual connection with outdoors. Natural ventilation is promoted through aeolian suction pipes. High thermal capacity is added by burying a part of the building: North and south-facing walls are 30% buried, while a 0.7 m berm surrounds an additional 20% of the external walls. 2.2. Monitoring equipment and location The Auditorium was monitored every 15 min during a week in winter holidays (July 13–19, 2000). Meteorological conditions (wind velocity and direction, outdoor drybulb temperature, and solar irradiance on horizontal surface) and indoor drybulb temperatures were monitored. A data-acquisition system was used, consisting of two NUDAM modules connected to a laptop PC (eight channels with T-thermocouples for temperature measurements and eight channels with voltage inputs). Inside the building, sensors were located in the access hall and at different heights in the Auditorium (one sensor at 0.6 m from the floor, five sensors at 2 m from the floor and parallel to the North wall, and six sensors at 2 m from the floor and parallel to the South wall). The locations of the sensors are shown in Fig. 5. The measurements of the five sensors parallel to North wall were averaged due to the little temperature difference between measurements (less than 0.3 8C), and the same method was used for the six sensors parallel to South wall. The building was unoccupied due to holidays, so windows and doors remained closed. To analyse the indoor temperature rise, the heating system was turned on manually from 8:00 to 19:00 a.m. during 3 consecutive days (July 17th, 18th, and 19th).
2.3. The simulation program SIMEDIF for Windows SIMEDIF needs the building to be divided into thermal zones. A zone is a building space that can be considered isothermal (i.e., a zone can be a group of various rooms in the building or a part of a room). Thus, each zone has a unique temperature whose temporal evolution is determined by the code, using the building data, materials, location, orientation, ambient temperature, and solar irradiance on horizontal surface in the simulation period. Those zones are connected to each other and with the environment by elements that have any of the following thermal characteristics: Storage and heat transfer by conduction, such as brick walls, adobe walls, concrete walls, etc. These elements are called massive walls. Storage into a uniform temperature mass such as water walls where the water is supposed to convect and has a uniform temperature. Heat transfer by conduction without storage, such as wood, expanded polystyrene, etc. In SIMEDIF, these elements are called lightweight walls. Heat transfer by conduction with two heat transfer coefficients (day and night), such as windows with night insulation. These elements are called windows, and they do not describe to solar gain or ventilation. Heat transfer by convection, such as openings in the walls that allow a convective air exchange. These elements are doors and vents in SIMEDIF. An energy balance is expressed at each node for which the temperature is to be determined. These nodes are of two different types: massive nodes (i.e., the nodes on massive walls and water walls) and non-massive nodes (i.e., the nodes on air and lightweight walls). The program calculates the temperatures Ti(t) inside and on massive wall surfaces by dividing the wall, that can be made of different materials (characterized by their conductivity k, density r, and specific heat to constant pressure cp), into a number of layers defined by the user, and applying the heat transfer equations in one-dimension to each layer. The building surfaces are described by an absorption coefficient a, and they can receive solar radiation I(t) on all its surface or on a smaller area AR. The heat transfer between a building surface and the surrounding air is described with a global time-invariant heat transfer coefficient h (W/m2 8C), which includes convection and radiation. The finite-difference explicit method is used to replace the time and spatial derivatives by finite differences. The result is a set of equations for m massive nodes. The temperature at time t + Dt can be evaluated from temperatures at the previous time t. This set of equations constitutes an initial value problem. Initial conditions, consisting of the starting storage mass temperatures and air temperatures, must be specified. The room drybulb temperature Tr(t) at t + Dt is computed from the global heat balance equation valued in the previous instant t. In this balance equation, the air renewals in the room,
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997 Table 3 Characteristics of the different zones of the Santa Rosa’s building Zone of Auditorium
Volume (m3)
Heating system
Access Hall Services Northern zone Southern zone Bottom zone
27.2 94.3 624.5 624.5 364.2
No No Yes
inner heat gains, and heat transfer due to the different elements connecting the room with other zones in the building and with outdoors, are considered. A detailed expression for this balance equation can be found in [20]. To simulate the hourly temperature with SIMEDIF for Windows, the building was divided into three zones (Auditorium, access hall and services), and the Auditorium was subdivided into three extra zones: northern, southern and bottom zones, to account for thermal differences inside this space. These thermal zones are shown with dashed lines in Fig. 2. The code calculates the hourly zone temperature by using the building data, materials, site, orientation, outdoor temperature, and solar irradiance on horizontal surface. The geometrical data were obtained from the construction drawings, the thermal properties of materials from tables commonly available in the heat transfer literature (see Table 2), and the hourly outdoor temperature and solar irradiance were monitored and recorded in a text file that the program read. The volumes of each zone are summarized in Table 3. Convective-radiative heat transfer coefficients were estimated through the dimensional equation [30]: h ¼ 5:7 þ 3:8v
(1)
where v is the wind speed in m/s and h is the heat transfer coefficient in W/m2 8C. This equation includes the effects of free convection and radiation, as Duffie and Beckman [30] pointed out. The monitored mean wind velocity was 1.6 m/s, thus a heat transfer coefficient of 12 W/m2 8C was imposed on external surfaces. On internal surfaces, convective-radiative heat transfer coefficients of 8 and 6 W/m2 8C (for surfaces with and without solar gain respectively) were imposed. These values are found to give accurate adjustments of monitored data, based on the authors’ own experience on building thermal simulation.
991
Table 4 Experimental mean temperatures of each zone, considering only the occupancy period (from 8:00 a.m. to 19:00 p.m.), for the period without heating system and with the heaters turned on Zone
Mean temperature (8C) (heaters off)
Mean temperature (8C) (heaters on)
Access Hall Northern zone Southern zone Bottom zone Outdoor air
13.5 10.5 10.0 8.4 5
14.4 18.0 12.9 10.9 6
2.4. Results Monitored data are shown in Fig. 4. Table 4 summarizes the mean drybulb temperatures of each zone, considering only the occupancy period (from 8:00 a.m. to 19:00 p.m.), for the period without heating system and with the heaters turned on. The records show that access hall was the warmest zone, due to a northern single pane glazed door (1.90 m width and 2.05 m height) that allows the direct solar gain to be absorbed by the dark floor. The outdoor temperature swing of 11.6 8C was attenuated to 3.6 8C in the Auditorium. Inside it and during the occupancy period (from 8:00 to 19:00 a.m.), the area next to the North wall (northern zone) had slightly higher temperatures than the South one (around 0.5 8C above the northern zone). Thermal vertical stratification was observed, from 1.6 to 2.1 8C between the sensors at 0.6 and at 2 m from the floor. This difference grew up to 4.5 8C when the heating system was turned on, caused by the heating system outlets located near the roof. Thus, the warmer air provided by the ducts increased the thermal stratification, already favoured by the hotter steel roof. This undesirable situation was discussed with the team who designed the heating system installation, to avoid similar problems in the future. This stratification was pointed out in a survey applied on students and professors, who referred to the coldness perceived at sitting level and the warmness at the teaching platform. As expected in a lightweight insulated building, the indoor temperatures increased when the heating system was turned on, satisfying the quick response requirement. The experimental and simulated data sets agree with a mean deviation between 0.5 and 1 8C, as shown in Fig. 5, where the
Fig. 4. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperatures from July 13th to 19th, 2000.
992
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 5. Monitored outdoor temperature, solar irradiance on horizontal surface, and averaged indoor temperature from July 13th to 27th, 2000. The simulation with SIMEDIF is shown with solid line.
temperatures of the North, South and Bottom zones were averaged to obtain a hourly indoor temperature of all the Auditorium space and to reduce the visual information in the graphics. The predicted indoor thermal swings and the effect of the heating system (on July 17th to 19th) agreed with measurements. During the first period without additional heating (July 13th to 16th), the outdoor mean temperature during the occupancy period was around 5 8C, while the mean temperature inside the Auditorium was around 9.6 8C, which is outside the comfort zone and evidences the need of auxiliary heating. When the heaters were turned on (July 17th to 19th, with an outdoor mean temperature around 6 8C), the mean temperature in the Auditorium was 13.9 8C during the occupancy period, with maxima around 20 8C and a measured daily energy consumption of 1.1 kWh/m2 day (4 MJ/m2 day). The building is outside the comfort zone when it is unoccupied. The simulations showed that the indoor temperature grows up to 5 8C when the metabolic heat gain of 200 students is considered. In this situation, that is a very common use pattern, the building is inside the comfort zone in the most part of the day. As a concluding remark: the thermal behaviour of the building in winter was satisfactory, but two aspects can be improved: the vertical thermal stratification and the reduction of heating loads. The temperature gradient could be lowered if the heating system is correctly installed. In a future design, a
reduction of the heating load can be obtained in winter by two strategies: an increase of the direct solar gains and the use of a solar passive system, as Trombe walls and solar air collectors. The increase of glazing and shading devices needs to be carefully designed, to prevent against glare and overheating in summer. 3. Second project: Auditorium in general pico 3.1. Improving the original building for the new location Based on the described experience for Santa Rosa, during 2003 a new Auditorium was designed for the National University of La Pampa, in General Pico city, and it was inaugurated in June 2005. The University asked for a building for 200 students with the minimum conventional energy consumption, and the lowest economic cost. The climatic conditions in General Pico are different than in Santa Rosa, as previously shown in Table 1. In summer, the outdoor temperatures are higher, with mean temperatures oscillating between 22 and 24 8C, mean maxima temperatures around 31 8C, and solar irradiance on horizontal surface of 23.4 MJ/ m2. The winter periods are cold-temperated, with mean temperatures between 8 and 10 8C, minima mean temperatures between 1.8 and 3 8C, solar irradiance on horizontal surface of 7.2 MJ/m2, and clearness index Kt around 0.52. The analysis of
Fig. 6. Simulated hourly thermal behaviour of the Auditorium placed in General Pico in a summer period, with and without students (occupancy schedule: 8 to 12 a.m. and 14 to 18 p.m.).
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
993
Fig. 7. Simulated hourly surface temperatures of the exterior North, South, and West walls, for a summer period in General Pico.
the climatic conditions indicates that winter conditions are more temperate (1204 vs. 1545 heating degrees days) but the cooling loads in summer are higher than in Santa Rosa (473 vs. 128 cooling degrees days). So for the winter period we decided to lower the heating load by using direct solar gain through glazing and indirect solar heating through solar air collectors. For the summer it was crucial to determine if there was overheating, which was analysed with SIMEDIF. As a first approach we considered the original building, without design changes, sited in General Pico. A simulation for five days in summer with and without students was done (Fig. 6): when outdoor mean temperature is around 23.6 8C, the mean indoor temperature is 32.7 8C (without students) and 36.7 8C (with 200 students), clearly out of the comfort zone. The simulations of the exterior surface temperature of the North, South, and West
walls are shown in Fig. 7: the maxima temperatures of the wall surfaces exceeds 40 8C, with the western wall being the warmest surface. Thus, in addition to high resistance roof and walls, shading, natural ventilation, higher air changes per hour in summer was included in the improved building. Also it was necessary to lower the solar energy absorbed by the walls, so the use of deciduous vegetation, clear painting colours, and shading devices were considered. An auxiliary cooling system was needed, and it was sized once the final design was concluded. Some aspects of the original building remained unchanged: the vertical envelope (0.18 m brick wall finished by pine wood in the internal side, with thermal insulation), the roof technology (a parabolic galvanized steel sheet, finished with thermal insulation and pine wood in the internal surface),
Fig. 8. Plan view of the improved building in General Pico and location of the temperature sensors.
994
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 9. Sectional and longitudinal views of the building in General Pico.
double glazing and hermetic carpentry. The additional design changes included: To minimize winter loads:
of conventional energy in summer. To avoid overheating, a cooling system of 5 kW is needed to maintain the mean indoor temperature around 23 8C and maxima values below 25.5 8C.
Seven meter square of double glazing, incorporated to the North wall for daylighting and direct solar gain in winter. Indirect solar gain through three air collectors by natural convection, to decrease the thermal load in winter. The air collectors, of 4 m2 each one, have a polycarbonate cover and they were installed in the North wall. In summer, the collection area is shaded by blackout sunscreens.
3.2. Monitoring results
A ventilated air camera and an insulated roof, to prevent the heat transfer between the hot metal roof and the indoor air. Shading devices, thicker insulation of the West wall, deciduous vegetation to shade North and South walls, and higher ventilation levels through three aeolian suction pipes. The aeolian suction pipes are 0.6 m in diameter and they have a mechanical system (a manual crank) to regulate the air changes per hour. For a mean wind velocity of 10 km/h, the manufacturer specifications indicates 11 air changes per hour for this particular building, which is higher than the minimum recommended by the norm. The berm protecting the walls was increased up to 2.3 m to provide thermal mass.
The new Auditorium was inaugurated in June 2005. The building was monitored from June 21th, 2005 to February 2, 2006. Wind velocity and direction, outdoor air temperature, solar irradiance on horizontal surface, and indoor temperatures were monitored. HOBO sensors were used for temperature measurements: 12 sensors were located in different zones inside the Auditorium (three sensors parallel to the North wall, three parallel to the South wall, three at center, one at the teacher platform and two next to the access door). The locations of the sensors are shown in Fig. 8. The air temperatures at inlet and outlet of the solar collectors were also measured. The temperature measurements of the sensors parallel to North wall were very similar, so they were averaged. The same method was used for the sensors at the center and the three HOBOS parallel to the South wall. The vertical thermal stratification, very significant in the Santa Rosa’s building was reduced to 1 8C between the teaching platform and the lowest sitting level. Only two representative periods were selected: one in winter holidays, when heating system was turned off and the building was unoccupied (July 15th to 18th), and one in summer, when the building was under complete occupation (November 15th to
The global losses coefficient G was estimated around 0.89 W/m3 8C, which is lower than the maximum admissible by the norms (Norma IRAM 11604/86). The total covered area is around 252 m2, and the volume is 1240 m3. Plan and sectional views of the building and a wall scheme are shown in Figs. 8– 10. Interior and exterior pictures of the indirect solar air collectors are shown in Fig. 11. More details of the building design and technology are in [31]. This final layout was achieved through successive simulations. To quantify the improvements of the new design, the simulated summer thermal behaviour of the unmodified (original) and final building are shown in Fig. 12, with 200 students in real use conditions. Monitored outdoor temperature data and solar irradiance on horizontal surface for General Pico city were used in the simulations. The results show that with diurnal mean outdoor temperatures around 25 8C, the mean indoor temperature is lowered from 36 8C (unmodified building) to 27 8C (improved building). This significant decrease of the cooling load reduces in a 70% the requirements
Fig. 10. Wall and floor scheme: (1) concrete subfloor; (2) waterproof floor insulation; (3) reinforced concrete subfloor; (4) concrete filling; (5) Floor; (6) Pine wood board; (7) glass wool with black veil; (8) glass wool; (9) waterproof insulation; (10) brick laid on edge with double waterproof insulation; (11) external plaster.
To minimize summer loads:
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
995
Fig. 11. Indirect solar air collectors (exterior and interior views).
Fig. 12. Simulated hourly thermal behaviour of the improved Auditorium in a summer period, with students.
18th). An extra period of four sunny winter days was selected to show the performance of the air solar collectors (August 30th to September 2nd). The results are shown in Figs. 13–15. Fig. 13 shows the measurements in a winter holiday period (July 15th to 18th), when the building was unoccupied and the heating system was turned off. The outdoor mean temperature during the occupancy period (8:00 a.m. to 19:00 p.m.) was around 9 8C, while the mean temperature inside the Auditorium was around 178. When the metabolic heat gain of the students is considered the mean temperature is inside the comfort zone during all the occupancy period. This can be analysed in Fig. 14 with the Auditorium under occupancy and the heating system function-
ing with an automatic thermostat (set in 21 8C). In Fig. 14, the air temperature at inlet and outlet of the air collector and solar irradiance on horizontal surface are shown. At the collector inlet the air temperature is the temperature of the Auditorium, around 24 8C. This air is warmed while it flows inside the collector, reaching temperatures as high as 50 8C in the solar midday, when the solar irradiance reaches its maximum value. This hotter air returns back to the Auditorium by natural convection and warms the indoor space. When compared with Santa Rosa building, we conclude that the difference between the mean outdoor and indoor temperature was increased from 4.6 8C (in the original building) to 8 8C (in the modified building), with the subsequent lowering of the heating load. A
Fig. 13. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperature for a winter holiday period (July 15th to 18th, 2005), when the heating system is turned off and the building is unoccupied.
996
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 14. Monitored temperature at the inlet and outlet of solar air collector and solar irradiance on horizontal surface, for a winter period (August 30th to September 2nd, 2005), when the building is used by teachers and students.
Fig. 15. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperature for a summer period (November 15th to 18th, 2005), when the building is occupied by teachers and students.
daily energy consumption of 0.94 kWh/m2 day (3.4 MJ/m2 day) was measured, with mean indoor temperatures around 20.6 8C. Fig. 15 shows the results for summer (November 15th to 18th), when the building was under complete occupation and use. The indoor temperature during the occupancy period (8:00 a.m. to 19:00 p.m.) was always below the outdoor temperature and, the most part of the day, inside the comfort zone for developing countries. Mean indoor temperatures around 23 8C were common, which is a successful result considering that metabolic heat contributes with values as high as 5 8C to the indoor temperature. With higher outdoor temperatures, which are common in summer, the cooling system will supply fresh air. If we compare this summer behaviour with that of the original unmodified building, we conclude that the difference between mean outdoor and indoor temperatures was lowered from 10 8C (for the original building) to 3.4 8C (for the modified building), with a considerable lowering of the cooling loads in summer. Besides, for days with maxima temperatures around 30 8C, the indoor space is inside the comfort zone and the use of the cooling system can be avoided. This contributes with a reduction of 70% in the requirements of conventional energy. 4. Conclusions Numerical programs greatly facilitate the study of thermal transients in buildings, which is very important both for
evaluating the heating and/or cooling energy requirements and for achieving environmental comfort conditions. The study of building thermal performance by means of software simulation packages is crucial to face real world problems in the design of energy efficient buildings. A previous simulation stage was essential to predict the winter and summer thermal behaviour of the building, helping to choose among different materials and technology for the building envelope, glazed areas, shape, orientation, and so on. The analysis of the building thermal behaviour shows that the energy behaviour during winter months is encouraging, but warmer periods seem to be the ones that present greater difficulties. Our contribution is applicable to regions of developing countries where non-traditional materials and qualified workmanship are not available. The insulated envelope, direct and indirect solar gains, and air collectors, warrants the winter thermal comfort, with predicted savings of 50% in heating requirements respect to the conventional layout, without additional cost. When compared with Santa Rosa building (without heating and without students), we conclude that the difference between the mean outdoor and indoor temperature was increased from 4.6 8C (in the original building) to 8 8C (in the modified building). Besides, with similar daily energy consumptions around 1 kWh/m2 day, the mean indoor temperature was increased from 14 8C (original) to 24 8C (modified building). In summer, if we compare the
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
both buildings, we conclude that the difference between mean outdoor and indoor temperatures was lowered from 10 8C (for the original building) to 3.4 8C (for the modified building), which implies a reduction of 70% in the requirements of conventional energy for cooling. Besides, for days with maxima temperatures around 30 8C, the indoor space is inside the comfort zone and the use of the cooling system can be avoided. The qualitative and quantitative analyses carried out showed promising results related to the buildings energy behaviour, yet it cannot be denied that there are certain factors that have direct incidence on the optimal energy behaviour of buildings and their thermal comfort: (a) exogenous factors (the high variability of the external physical environment), (b) endogenous factors (constructive characteristics) and (c) socioenvironmental factors (life-styles and behaviour of users, associated to the active energy contribution and preferences of heating levels). The true challenge for solar energy architecture is to simplify the technology and construction processes so that costs would not be higher than those for conventional buildings, assuring users’ pro-active behaviour towards the building’s correct thermal management. To involve users in these improved practices is still another challenge. Acknowledgements The authors are grateful for support from projects PICT 2000 No. 13-09991, CIUNSa Nos. 1088 and 1332 of Universidad Nacional de Salta, Argentina, and PIP CONICET 6543. References [1] M. Wall, Energy-efficient terrace houses in Sweden Simulations and measurements, Energy and Buildings 38 (2006) 627–634. [2] F. Garde, L. Adelard, H. Boyer, C. Rat, Implementation and experimental survey of passive design specifications used in new low-cost housing under tropical climates, Energy and Buildings 36 (2004) 353–366. [3] N. Cardinale, F. Ruggiero, Energetic aspects of bioclimatic buildings in the Mediterranean area: a comparison between two different computation methods, Energy and Buildings 31 (2000) 55–63. [4] Y. Feng, Thermal design standards for energy efficiency of residential buildings in hot summer/cold winter zones, Energy and Buildings 36 (2004) 1309–1312. [5] M. Macias, A. Mateo, M. Schuler, E.M. Mitre, Application of night cooling concept to social housing design in dry hot climate, Energy and Buildings 38 (2006) 1104–1110. [6] J. Pfafferott, S. Herkel, M. Wambsganß, Design, monitoring and evaluation of a low energy office building with passive cooling by night ventilation, Energy and Buildings 36 (2004) 455–465. [7] H. Breesch, A. Bossaer, A. Janssens, Passive cooling in a low-energy office building, Solar Energy 79 (2005) 682–696. [8] S. Citherlet, J.A. Clarke, J. Hand, Integration in building physics simulation, Energy and Buildings 33 (2001) 451–461. [9] J. Kosny, E. Kossecka, Multi-dimensional heat transfer through complex building envelope assemblies in hourly energy simulation programs, Energy and Buildings 34 (2002) 445–454. [10] J.F. Karlsson, B. Moshfegh, Energy demand and indoor climate in a low energy building—changed control strategies and boundary conditions, Energy and Buildings 38 (2006) 315–326. [11] S.O. Jensen, Validation of building energy simulation programs: a methodology, Energy and Buildings 22 (1995) 133–144.
997
[12] http://www.eere.energy.gov/buildings/energyplus/. [13] T. Ayata, O. Yildiz, Investigating the potential use of natural ventilation in new building designs in Turkey, Energy and Buildings 38 (2006) 959–963. [14] M.L. Persson, A. Roos, M. Wall, Influence of window size on the energy balance of low energy houses, Energy and Buildings 38 (2006) 181–188. [15] M. Casermeiro, L. Saravia, Ca´lculo Te´rmico Horario de Edificios Solares Pasivos, in: Actas de 98 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, San Juan, Argentina, 1984, pp. 39–45. [16] A. Esteves, J. Ferna´ndez, M. Basso, J. Mitchel, C. de Rosa, Simulacio´n te´rmica de edificios: aplicacio´n de los modelos Quick y SIMEDIF, in: Actas de la 178 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, Rosario, Argentina, 1994, pp. 543–550. [17] A. Herna´ndez, G. Lesino, Ana´lisis de la performance te´rmica de un prototipo de vivienda liviana: monitoreo y simulacio´n macrodina´mica. Parte I, in: Actas de la 168 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, La Plata, Argentina, 1993, pp. 167–174. [18] A. Herna´ndez, G. Lesino, Simulacio´n mediante SIMEDIF del comportamiento te´rmico de un prototipo de vivienda liviana construido en la Universidad Nacional de Salta, Avances en Energı´as Renovables y Medio Ambiente 4 (2) (2000) 08.29–08.34. [19] S. Flores Larsen, A. Herna´ndez, G. Lesino, N. Salvo, Measurement and simulation of the thermal behaviour of a massive building with passive solar conditioning, in: Proceedings of VII International Building Simulation Congress 2001, Rı´o de Janeiro, Brazil, (2001), pp. 183–190. [20] S. Flores Larsen, G. Lesino, A new code for the hour-by-hour thermal behaviour simulation of buildings, in: Proceedings of VII International Building Simulation Congress 2001, Rı´o de Janeiro, Brazil, (2001), pp. 75–82. [21] S. Flores Larsen, C. Filippı´n, G. Lesino, Buried pipes for passive cooling: two cases of educational buildings in argentina, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [22] V. Garcı´a, A. Iriarte, G. Lesino, S. Flores Larsen, C. Matı´as, Special greenhouse for walnut grafting in arid regions, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [23] N. Baker, M. Standeven, Thermal comfort for free-running buildings, Energy and Buildings 23 (1996) 175–182. [24] M. Wall, Distribution of solar radiation in glazed spaces and adjacent buildings. A comparison of simulation program, Energy and Buildings 26 (1997) 129–135. [25] G. Lesino, R. Requena, L. Saravia, L. Fontanilla, R. Echazu´, Edificios solares en La Puna: estado de avance de su construccio´n y modelizacio´n, in: Actas de 78 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, Rosario, Argentina, 1981, pp. 138–145. [26] C. De Rosa, M. Basso, A. Esteves, A. Pattini, J. Ferna´ndez, L. Cortegoso, J. Mitchell, M. Guisasola, G. Lesino, L. Saravia, Escuela solar te´cnico agraria en Mendoza. Ana´lisis del beneficio invernal de la incorporacio´n de un invernadero adosado, Energı´as Renovables y Medio Ambiente 1 (1996) 11–19. [27] G. Lesino, R. Requena, R. Caso, A. Herna´ndez, N. Salvo, Vivienda liviana aislada con servicios solares, Energı´as Renovables y Medio Ambiente 3 (1997) 47–54. [28] C. Filippı´n, S. Flores Larsen, A. Beascochea, G. Lesino, Response of conventional and energy-saving buildings to design and human dependent factors, Solar Energy 78 (3) (2005) 455–470. [29] C. Filippı´n, A. Beascochea, S. Flores Larsen, G. Lesino, A solar school building in a temperate region in Argentina, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [30] J.A. Duffie, W.A. Beckman, Solar Engineering of Thermal Processes, second ed., Wiley Interscience, New York, 1991. [31] C. Filippı´n, A. Beascochea, S. Flores Larsen, Auditorio bioclima´tico en la region central de Argentina, in: IV Congreso Latinoamericano sobre Confort y Comportamiento Te´rmico de los Edificios, Me´xico, 2005.
An experience on integrating monitoring and simulation tools in the design of energy-saving buildings S. Flores Larsen a,1,*, C. Filippı´n b,2, A. Beascochea b,2, G. Lesino a,1 b
a INENCO, Universidad Nacional de Salta, CONICET, Buenos Aires 177, (4400) Salta, Argentina Universidad Nacional de La Pampa, CONICET, Spinetto 785, (6300) Santa Rosa, La Pampa, Argentina
Received 11 May 2007; received in revised form 31 July 2007; accepted 2 August 2007
Abstract In this paper we describe the design and thermal behaviour of a bioclimatic Auditorium at the National University of La Pampa, used for teaching activities in Santa Rosa, La Pampa (Argentina). The building was monitored in winter and simulated with SIMEDIF for Windows, a code developed at the Non Conventional Energy Research Institute (INENCO, Argentina). Then, a new project of a similar building was proposed for General Pico city, and the obtained physical model was used to simulate the building under the summer temperatures of the new city. The building was redesigned and passive solar strategies were applied to reduce heating and cooling loads. The final layout and the monitored thermal behaviour of the new building in winter and summer are described. Without additional costc, the new building savings were 50% in heating requirements respect to the conventional layout, and 70% in the requirements of conventional energy for cooling. # 2007 Elsevier B.V. All rights reserved. Keywords: Energy-efficient buildings; Passive solar design; Energy-saving; Thermal simulation
1. Introduction The challenge of reducing the emission of greenhouse gases at local and global levels requires behavioural changes in life styles and energy consumption patterns in people, and the use of more energy efficient production, processing and distribution technologies. The improvement of building techniques is an alternative way to increase the energy efficiency and reduce gas emissions. Passive solar design exploits the building’s orientation, shape, materials, windows, and external landscape, in combination with other energy efficiency strategies, to create a pleasant environment which is less dependent on fossil fuelbased energy. It has been extensively shown that it is possible to build passive houses with very low energy use and to normal costs. Important energy-savings as high as 60% compared to conventional houses have been found in cold [1], tropical [2], Mediterranean [3] and hot summer/cold winter climates [4]. Some successful applications of passive design are social * Corresponding author. Tel.: +54 387 4255489; fax: +54 387 4255489. E-mail addresses: [email protected], [email protected] (S.F. Larsen), [email protected] (C. Filippı´n), [email protected] (G. Lesino). 1 Tel.: +54 387 4255578; fax: +54 387 4255489. 2 Tel.: +54 2954 434222; fax: +54 2954 434222. 0378-7788/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2007.08.004
housing [2,5] and office buildings [6,7], where the energy demand for air conditioning can be significantly reduced. In this context, simulation programs have become important tools to improve building designs and energy consumption. These programs can calculate the thermal behaviour of buildings and change different variables, as the climatic conditions, geometry, materials, etc., to evaluate their thermal response. Thus, a feedback is carried out until an adequate final project is achieved [8]. If experimental data validates the physical model used in the simulation, a deeper insight of the building physics and good energy performance can be achieved, as shown in many research works [3,6]. Nowadays, a wide variety of simulation programs is available, of different complexity levels going from steady-state situation to very sophisticate CFD simulation [9]: TRNSYS [7], ESP-r [10,11], DOE-2, BLAST, Energy Plus [12], TAS [5], FLUENT [13], DEROB-LTH [14] between other common programs, can be mentioned. Methodologies for validation of building energy simulation programs have also been developed [11]. In Argentina, a lot of effort was directed since 1984 to develop a code for the simulation of transient thermal behaviour of passive multiroom buildings. SIMEDIF model and program was made by researchers of INENCO (Non Conventional Energy Research Institute), Argentina. The first version was
988
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
developed for DOS by Casermeiro and Saravia [15] in 1984 and the code has been largely validated since then throughout years of experimental work in Argentina. Numerous groups continuously used it for research [16], design, and simulation of the thermal behaviour of lightweight [17,18] and massive [19] buildings from the desert climate of the Puna (North of Argentina) to La Pampa. Since 2000 the code was adapted for Windows environment [20] and two modules for passive cooling systems were added: a module to calculate earth-to-air heat exchangers and a module for evaporative cooling systems. These models were validated with monitored data on existing buildings with buried pipes [21] and a greenhouse with evaporative panels [22]. This software was used in this paper, so a brief description of the thermal model was included in the related section. As mentioned, the growing demand for air-conditioned buildings and the resultant demand for electrical energy has prompted research into passive cooling, such as the European Union funded PASCOOL programme [23] and recommendations, guidelines and laws for energy-saving in buildings [24]. In Argentina, the local laws are not deeply concerned with lowering energy consumption of buildings. Local building designers have largely ignored passive design strategies, which can effectively reduce the building energy consumption. Only isolated exceptions can be mentioned ([25–27]), mainly architects and researchers of universities, who tried to change this situation and whose efforts are becoming clear since the last years. In La Pampa, the situation is even worse, because there is a lack of non-traditional materials and qualified workmanship. It is very difficult to convince the local authorities that energy conservation devices, passive solar heating, and bioclimatic design are beneficial and can be used without additional cost. On the other hand, designers still lack confidence to apply these techniques in absence of information about the performance of existing bioclimatic buildings in Argentina. In the last years, some effort was made to change this idea, by building energy-saving apartments for low-income students [28], a solar school with buried pipes for earth-to-air heat exchange [29], and a College of Agronomic Sciences (National University of La Pampa) [21], whose design combines different passive solar strategies to reduce up to 50% the natural gas consumption, when compared with conventional buildings. This paper trends to fill these holes and to bring some confidence to designers and construction workers of developing countries. In this context, socio-economic, educational and environmental reasons have driven the design of two Auditoriums with
Fig. 1. Location of Santa Rosa (36.578S, 64.45W, 189 m over sea level) and General Pico (35870 S, 63880 W, 141 m over sea level).
reduced energy consumption, for the National University of La Pampa. The local government insisted that the economic costs were similar to that of a conventional building, which due to local economic reasons is in general low. A big effort was made in the building design in order to accomplish this requirement. Both buildings are located in the center of La Pampa province, in Santa Rosa and General Pico cities, in a temperate semi-arid agricultural region of central Argentina (Fig. 1). The climatic data for summer and winter are summarized in Table 1 for both cities. According to the Olgyay’s bio-climogram, people could take advantage of solar radiation and reach the comfort area in 60% of the annual period. Passive solar systems, thermal inertia, natural ventilation, thermal insulation, external shading, building orientation and a compact design can be used to improve the thermal comfort along the year. The first building was designed and built between 1998 and 2000 in Santa Rosa city, where the climate is cold-temperate. It was monitored and simulated during a winter period to evaluate its thermal behaviour and to obtain a physical model of the building. Then, the possibility of a new similar project in General Pico city arose, where the winter conditions are more temperate (1204 vs. 1545 heating degrees days) but the cooling loads in summer are higher than in Santa Rosa (473 vs. 128 cooling degrees days), as shown in Table 1. The previous experience on Santa Rosa’s building was used and the simulation of the building under the summer temperatures of the new city was done. When overheating was confirmed, the redesign of the building was afforded in order to load the heating and cooling energy requirements. In the first part of this
Table 1 Meteorological data for Santa Rosa and General Pico, La Pampa, Argentina (provided by the National Meteorological Service) Winter
Summer
Temperatures (8C) (June)
Santa Rosa General Pico
Mean minimum
Mean
Swing
1.8 2.7
8.2 8.9
11.9 6.6
Heating degrees days base 18 8C 1545 1204
Temperatures (8C) (December) Mean maximum
Mean
Swing
30.3 30.9
22.3 23.4
14.9 14.6
Cooling degrees days base 23 8C 128 473
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
989
Fig. 2. Plan and section of the building in Santa Rosa, location of the temperature sensors and subdivision of the Auditorium into three thermal zones.
paper we describe the original design of the Auditorium in Santa Rosa. Then, a short description of SIMEDIF program and an analysis of the experimental and simulated indoor temperatures for a winter period are presented. The second part describes the summer simulation of the same building in General Pico, and the passive and hybrid design strategies applied to the building in order to deal with overheating and to reduce heating loads. Finally, the monitored winter and summer thermal behaviour of the new building are analysed, and conclusions for future applications and research are presented. 2. First project: Auditorium in Santa Rosa 2.1. Brief description of the building The bioclimatic building of Santa Rosa includes an Auditorium for 200 people, an access hall, and services (with two bathrooms and a little storeroom), with a total covered area of 316 m2. Fig. 2 shows a plan of the building and a vertical section of the Auditorium. The Auditorium is occupied typically by 150–200 students from 8:00 to 13:00 h and 14:00 to 20:00 h, so a quick response of the indoor temperature to the heating system in winter is desirable, together with a reduction of heating and cooling loads. Thus, the building is designed as a lightweight insulated structure and the following strategies are applied: Heating season: the transmission and infiltration heat losses are minimized. The infiltration heat losses are prevented by using hermetic carpentry. Windows have double glazing and walls and roof are well insulated. The external walls are composed of three layers (Fig. 3), from outside to inside: a waterproof cement/sand plaster layer on a 0.18 m thick brick, a 0.02 m thick cement/sand plaster layer, a 0.02 m of glass wool with black veil layer, a 0.03 m of glass wool with Kraft paper layer, and a pine tongued and grooved board. The Rvalue of the wall is 2.13 m2 8C/W. The south-facing wall has an additional layer of thermal insulation (0.025 m of expanded polystyrene) on its internal side, as shown in the picture of Fig. 3, and its R-value is 2.8 m2 8C/W. Higher R-
values imply thicker walls and/or thicker insulation, resulting in higher construction costs and longer return times of investment. In this case, thicker insulation is not recommended: much saving is not obtained because in these sites, winter outdoor temperatures are not as cold as in developed northern countries, where 0.1–0.3 m thick insulation are common. The indoor partition walls are each composed of a 0.2 m thick brick wall and a cement/sand plaster layer on one side. The access hall and services have slab roofs with thermal insulation (0.05 m of expanded polystyrene), while the Auditorium roof has a sloped 0.0007 m galvanized steel sheet (Fig. 3) with thermal insulation (a 0.03 m thick glass wool layer and a 0.04 m thick expanded polystyrene layer). The Auditorium ceiling finish is a pine tongued and grooved board, and the services and access hall ceilings are finished by a 0.05 m thick gypsum plaster with emulsion paint. The floor slab is 0.2 m thick reinforced concrete and it is finished with a
Fig. 3. Wall and roof scheme (left) and southern wall picture (right) with additional insulation: (1) pine wood; (2) glass wool with black veil (0.02 m); (3) glass wool with Kraft paper (0.03 m); (4) 0.18 m thick brick layer with waterproof cement/sand plaster on both sides; (5) brick laid on edge with waterproof cement/sand plaster layers on both sides; (6) wood tie beam; (7) wood board; (8) glass wool with Kraft paper (0.03 m); (9) expanded polystyrene (0.04 m, density: 20 kg/m3); (10) galvanized steel sheet (0.0007 m).
990
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Table 2 Thermal properties of building materials used in SIMEDIF for windows Material
Density (kg/m3)
Specific heat (J/kg K)
Thermal conductivity (W/m K)
Concrete Cement/sand plaster Brick Glass wool Expanded polystyrene Galvanized steel
2400 1184 1800 24 20 2700
653 836 920 920 1420 900
2.16 1.4 0.72 0.04 0.04 202
dark vinyl tile flooring. The properties of the various materials used in the building are summarized in Table 2. An auxiliary heating system was installed, consisting of two gas-fired units of 76500 W each one, with an air flow around 1.67 m3/s and a maximum air velocity in the central duct around 5 m/s, which is the allowed velocity to avoid noise problems. The heating requirements were estimated to obtain an indoor temperature around 20 8C with 30 occupants in winter. The heating units were connected to intelligent indoor thermostats, but during the monitoring described in this paper they were turned on and off manually, to control exactly their functioning periods. Cooling season: to prevent against overheating in summer, direct solar gain through windows is minimized: only one window with a double glass panel (2.70 m 1.5 m, U = 2.0 W/m2 8C) was included in the south wall, to provide natural daylighting and visual connection with outdoors. Natural ventilation is promoted through aeolian suction pipes. High thermal capacity is added by burying a part of the building: North and south-facing walls are 30% buried, while a 0.7 m berm surrounds an additional 20% of the external walls. 2.2. Monitoring equipment and location The Auditorium was monitored every 15 min during a week in winter holidays (July 13–19, 2000). Meteorological conditions (wind velocity and direction, outdoor drybulb temperature, and solar irradiance on horizontal surface) and indoor drybulb temperatures were monitored. A data-acquisition system was used, consisting of two NUDAM modules connected to a laptop PC (eight channels with T-thermocouples for temperature measurements and eight channels with voltage inputs). Inside the building, sensors were located in the access hall and at different heights in the Auditorium (one sensor at 0.6 m from the floor, five sensors at 2 m from the floor and parallel to the North wall, and six sensors at 2 m from the floor and parallel to the South wall). The locations of the sensors are shown in Fig. 5. The measurements of the five sensors parallel to North wall were averaged due to the little temperature difference between measurements (less than 0.3 8C), and the same method was used for the six sensors parallel to South wall. The building was unoccupied due to holidays, so windows and doors remained closed. To analyse the indoor temperature rise, the heating system was turned on manually from 8:00 to 19:00 a.m. during 3 consecutive days (July 17th, 18th, and 19th).
2.3. The simulation program SIMEDIF for Windows SIMEDIF needs the building to be divided into thermal zones. A zone is a building space that can be considered isothermal (i.e., a zone can be a group of various rooms in the building or a part of a room). Thus, each zone has a unique temperature whose temporal evolution is determined by the code, using the building data, materials, location, orientation, ambient temperature, and solar irradiance on horizontal surface in the simulation period. Those zones are connected to each other and with the environment by elements that have any of the following thermal characteristics: Storage and heat transfer by conduction, such as brick walls, adobe walls, concrete walls, etc. These elements are called massive walls. Storage into a uniform temperature mass such as water walls where the water is supposed to convect and has a uniform temperature. Heat transfer by conduction without storage, such as wood, expanded polystyrene, etc. In SIMEDIF, these elements are called lightweight walls. Heat transfer by conduction with two heat transfer coefficients (day and night), such as windows with night insulation. These elements are called windows, and they do not describe to solar gain or ventilation. Heat transfer by convection, such as openings in the walls that allow a convective air exchange. These elements are doors and vents in SIMEDIF. An energy balance is expressed at each node for which the temperature is to be determined. These nodes are of two different types: massive nodes (i.e., the nodes on massive walls and water walls) and non-massive nodes (i.e., the nodes on air and lightweight walls). The program calculates the temperatures Ti(t) inside and on massive wall surfaces by dividing the wall, that can be made of different materials (characterized by their conductivity k, density r, and specific heat to constant pressure cp), into a number of layers defined by the user, and applying the heat transfer equations in one-dimension to each layer. The building surfaces are described by an absorption coefficient a, and they can receive solar radiation I(t) on all its surface or on a smaller area AR. The heat transfer between a building surface and the surrounding air is described with a global time-invariant heat transfer coefficient h (W/m2 8C), which includes convection and radiation. The finite-difference explicit method is used to replace the time and spatial derivatives by finite differences. The result is a set of equations for m massive nodes. The temperature at time t + Dt can be evaluated from temperatures at the previous time t. This set of equations constitutes an initial value problem. Initial conditions, consisting of the starting storage mass temperatures and air temperatures, must be specified. The room drybulb temperature Tr(t) at t + Dt is computed from the global heat balance equation valued in the previous instant t. In this balance equation, the air renewals in the room,
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997 Table 3 Characteristics of the different zones of the Santa Rosa’s building Zone of Auditorium
Volume (m3)
Heating system
Access Hall Services Northern zone Southern zone Bottom zone
27.2 94.3 624.5 624.5 364.2
No No Yes
inner heat gains, and heat transfer due to the different elements connecting the room with other zones in the building and with outdoors, are considered. A detailed expression for this balance equation can be found in [20]. To simulate the hourly temperature with SIMEDIF for Windows, the building was divided into three zones (Auditorium, access hall and services), and the Auditorium was subdivided into three extra zones: northern, southern and bottom zones, to account for thermal differences inside this space. These thermal zones are shown with dashed lines in Fig. 2. The code calculates the hourly zone temperature by using the building data, materials, site, orientation, outdoor temperature, and solar irradiance on horizontal surface. The geometrical data were obtained from the construction drawings, the thermal properties of materials from tables commonly available in the heat transfer literature (see Table 2), and the hourly outdoor temperature and solar irradiance were monitored and recorded in a text file that the program read. The volumes of each zone are summarized in Table 3. Convective-radiative heat transfer coefficients were estimated through the dimensional equation [30]: h ¼ 5:7 þ 3:8v
(1)
where v is the wind speed in m/s and h is the heat transfer coefficient in W/m2 8C. This equation includes the effects of free convection and radiation, as Duffie and Beckman [30] pointed out. The monitored mean wind velocity was 1.6 m/s, thus a heat transfer coefficient of 12 W/m2 8C was imposed on external surfaces. On internal surfaces, convective-radiative heat transfer coefficients of 8 and 6 W/m2 8C (for surfaces with and without solar gain respectively) were imposed. These values are found to give accurate adjustments of monitored data, based on the authors’ own experience on building thermal simulation.
991
Table 4 Experimental mean temperatures of each zone, considering only the occupancy period (from 8:00 a.m. to 19:00 p.m.), for the period without heating system and with the heaters turned on Zone
Mean temperature (8C) (heaters off)
Mean temperature (8C) (heaters on)
Access Hall Northern zone Southern zone Bottom zone Outdoor air
13.5 10.5 10.0 8.4 5
14.4 18.0 12.9 10.9 6
2.4. Results Monitored data are shown in Fig. 4. Table 4 summarizes the mean drybulb temperatures of each zone, considering only the occupancy period (from 8:00 a.m. to 19:00 p.m.), for the period without heating system and with the heaters turned on. The records show that access hall was the warmest zone, due to a northern single pane glazed door (1.90 m width and 2.05 m height) that allows the direct solar gain to be absorbed by the dark floor. The outdoor temperature swing of 11.6 8C was attenuated to 3.6 8C in the Auditorium. Inside it and during the occupancy period (from 8:00 to 19:00 a.m.), the area next to the North wall (northern zone) had slightly higher temperatures than the South one (around 0.5 8C above the northern zone). Thermal vertical stratification was observed, from 1.6 to 2.1 8C between the sensors at 0.6 and at 2 m from the floor. This difference grew up to 4.5 8C when the heating system was turned on, caused by the heating system outlets located near the roof. Thus, the warmer air provided by the ducts increased the thermal stratification, already favoured by the hotter steel roof. This undesirable situation was discussed with the team who designed the heating system installation, to avoid similar problems in the future. This stratification was pointed out in a survey applied on students and professors, who referred to the coldness perceived at sitting level and the warmness at the teaching platform. As expected in a lightweight insulated building, the indoor temperatures increased when the heating system was turned on, satisfying the quick response requirement. The experimental and simulated data sets agree with a mean deviation between 0.5 and 1 8C, as shown in Fig. 5, where the
Fig. 4. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperatures from July 13th to 19th, 2000.
992
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 5. Monitored outdoor temperature, solar irradiance on horizontal surface, and averaged indoor temperature from July 13th to 27th, 2000. The simulation with SIMEDIF is shown with solid line.
temperatures of the North, South and Bottom zones were averaged to obtain a hourly indoor temperature of all the Auditorium space and to reduce the visual information in the graphics. The predicted indoor thermal swings and the effect of the heating system (on July 17th to 19th) agreed with measurements. During the first period without additional heating (July 13th to 16th), the outdoor mean temperature during the occupancy period was around 5 8C, while the mean temperature inside the Auditorium was around 9.6 8C, which is outside the comfort zone and evidences the need of auxiliary heating. When the heaters were turned on (July 17th to 19th, with an outdoor mean temperature around 6 8C), the mean temperature in the Auditorium was 13.9 8C during the occupancy period, with maxima around 20 8C and a measured daily energy consumption of 1.1 kWh/m2 day (4 MJ/m2 day). The building is outside the comfort zone when it is unoccupied. The simulations showed that the indoor temperature grows up to 5 8C when the metabolic heat gain of 200 students is considered. In this situation, that is a very common use pattern, the building is inside the comfort zone in the most part of the day. As a concluding remark: the thermal behaviour of the building in winter was satisfactory, but two aspects can be improved: the vertical thermal stratification and the reduction of heating loads. The temperature gradient could be lowered if the heating system is correctly installed. In a future design, a
reduction of the heating load can be obtained in winter by two strategies: an increase of the direct solar gains and the use of a solar passive system, as Trombe walls and solar air collectors. The increase of glazing and shading devices needs to be carefully designed, to prevent against glare and overheating in summer. 3. Second project: Auditorium in general pico 3.1. Improving the original building for the new location Based on the described experience for Santa Rosa, during 2003 a new Auditorium was designed for the National University of La Pampa, in General Pico city, and it was inaugurated in June 2005. The University asked for a building for 200 students with the minimum conventional energy consumption, and the lowest economic cost. The climatic conditions in General Pico are different than in Santa Rosa, as previously shown in Table 1. In summer, the outdoor temperatures are higher, with mean temperatures oscillating between 22 and 24 8C, mean maxima temperatures around 31 8C, and solar irradiance on horizontal surface of 23.4 MJ/ m2. The winter periods are cold-temperated, with mean temperatures between 8 and 10 8C, minima mean temperatures between 1.8 and 3 8C, solar irradiance on horizontal surface of 7.2 MJ/m2, and clearness index Kt around 0.52. The analysis of
Fig. 6. Simulated hourly thermal behaviour of the Auditorium placed in General Pico in a summer period, with and without students (occupancy schedule: 8 to 12 a.m. and 14 to 18 p.m.).
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
993
Fig. 7. Simulated hourly surface temperatures of the exterior North, South, and West walls, for a summer period in General Pico.
the climatic conditions indicates that winter conditions are more temperate (1204 vs. 1545 heating degrees days) but the cooling loads in summer are higher than in Santa Rosa (473 vs. 128 cooling degrees days). So for the winter period we decided to lower the heating load by using direct solar gain through glazing and indirect solar heating through solar air collectors. For the summer it was crucial to determine if there was overheating, which was analysed with SIMEDIF. As a first approach we considered the original building, without design changes, sited in General Pico. A simulation for five days in summer with and without students was done (Fig. 6): when outdoor mean temperature is around 23.6 8C, the mean indoor temperature is 32.7 8C (without students) and 36.7 8C (with 200 students), clearly out of the comfort zone. The simulations of the exterior surface temperature of the North, South, and West
walls are shown in Fig. 7: the maxima temperatures of the wall surfaces exceeds 40 8C, with the western wall being the warmest surface. Thus, in addition to high resistance roof and walls, shading, natural ventilation, higher air changes per hour in summer was included in the improved building. Also it was necessary to lower the solar energy absorbed by the walls, so the use of deciduous vegetation, clear painting colours, and shading devices were considered. An auxiliary cooling system was needed, and it was sized once the final design was concluded. Some aspects of the original building remained unchanged: the vertical envelope (0.18 m brick wall finished by pine wood in the internal side, with thermal insulation), the roof technology (a parabolic galvanized steel sheet, finished with thermal insulation and pine wood in the internal surface),
Fig. 8. Plan view of the improved building in General Pico and location of the temperature sensors.
994
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 9. Sectional and longitudinal views of the building in General Pico.
double glazing and hermetic carpentry. The additional design changes included: To minimize winter loads:
of conventional energy in summer. To avoid overheating, a cooling system of 5 kW is needed to maintain the mean indoor temperature around 23 8C and maxima values below 25.5 8C.
Seven meter square of double glazing, incorporated to the North wall for daylighting and direct solar gain in winter. Indirect solar gain through three air collectors by natural convection, to decrease the thermal load in winter. The air collectors, of 4 m2 each one, have a polycarbonate cover and they were installed in the North wall. In summer, the collection area is shaded by blackout sunscreens.
3.2. Monitoring results
A ventilated air camera and an insulated roof, to prevent the heat transfer between the hot metal roof and the indoor air. Shading devices, thicker insulation of the West wall, deciduous vegetation to shade North and South walls, and higher ventilation levels through three aeolian suction pipes. The aeolian suction pipes are 0.6 m in diameter and they have a mechanical system (a manual crank) to regulate the air changes per hour. For a mean wind velocity of 10 km/h, the manufacturer specifications indicates 11 air changes per hour for this particular building, which is higher than the minimum recommended by the norm. The berm protecting the walls was increased up to 2.3 m to provide thermal mass.
The new Auditorium was inaugurated in June 2005. The building was monitored from June 21th, 2005 to February 2, 2006. Wind velocity and direction, outdoor air temperature, solar irradiance on horizontal surface, and indoor temperatures were monitored. HOBO sensors were used for temperature measurements: 12 sensors were located in different zones inside the Auditorium (three sensors parallel to the North wall, three parallel to the South wall, three at center, one at the teacher platform and two next to the access door). The locations of the sensors are shown in Fig. 8. The air temperatures at inlet and outlet of the solar collectors were also measured. The temperature measurements of the sensors parallel to North wall were very similar, so they were averaged. The same method was used for the sensors at the center and the three HOBOS parallel to the South wall. The vertical thermal stratification, very significant in the Santa Rosa’s building was reduced to 1 8C between the teaching platform and the lowest sitting level. Only two representative periods were selected: one in winter holidays, when heating system was turned off and the building was unoccupied (July 15th to 18th), and one in summer, when the building was under complete occupation (November 15th to
The global losses coefficient G was estimated around 0.89 W/m3 8C, which is lower than the maximum admissible by the norms (Norma IRAM 11604/86). The total covered area is around 252 m2, and the volume is 1240 m3. Plan and sectional views of the building and a wall scheme are shown in Figs. 8– 10. Interior and exterior pictures of the indirect solar air collectors are shown in Fig. 11. More details of the building design and technology are in [31]. This final layout was achieved through successive simulations. To quantify the improvements of the new design, the simulated summer thermal behaviour of the unmodified (original) and final building are shown in Fig. 12, with 200 students in real use conditions. Monitored outdoor temperature data and solar irradiance on horizontal surface for General Pico city were used in the simulations. The results show that with diurnal mean outdoor temperatures around 25 8C, the mean indoor temperature is lowered from 36 8C (unmodified building) to 27 8C (improved building). This significant decrease of the cooling load reduces in a 70% the requirements
Fig. 10. Wall and floor scheme: (1) concrete subfloor; (2) waterproof floor insulation; (3) reinforced concrete subfloor; (4) concrete filling; (5) Floor; (6) Pine wood board; (7) glass wool with black veil; (8) glass wool; (9) waterproof insulation; (10) brick laid on edge with double waterproof insulation; (11) external plaster.
To minimize summer loads:
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
995
Fig. 11. Indirect solar air collectors (exterior and interior views).
Fig. 12. Simulated hourly thermal behaviour of the improved Auditorium in a summer period, with students.
18th). An extra period of four sunny winter days was selected to show the performance of the air solar collectors (August 30th to September 2nd). The results are shown in Figs. 13–15. Fig. 13 shows the measurements in a winter holiday period (July 15th to 18th), when the building was unoccupied and the heating system was turned off. The outdoor mean temperature during the occupancy period (8:00 a.m. to 19:00 p.m.) was around 9 8C, while the mean temperature inside the Auditorium was around 178. When the metabolic heat gain of the students is considered the mean temperature is inside the comfort zone during all the occupancy period. This can be analysed in Fig. 14 with the Auditorium under occupancy and the heating system function-
ing with an automatic thermostat (set in 21 8C). In Fig. 14, the air temperature at inlet and outlet of the air collector and solar irradiance on horizontal surface are shown. At the collector inlet the air temperature is the temperature of the Auditorium, around 24 8C. This air is warmed while it flows inside the collector, reaching temperatures as high as 50 8C in the solar midday, when the solar irradiance reaches its maximum value. This hotter air returns back to the Auditorium by natural convection and warms the indoor space. When compared with Santa Rosa building, we conclude that the difference between the mean outdoor and indoor temperature was increased from 4.6 8C (in the original building) to 8 8C (in the modified building), with the subsequent lowering of the heating load. A
Fig. 13. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperature for a winter holiday period (July 15th to 18th, 2005), when the heating system is turned off and the building is unoccupied.
996
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
Fig. 14. Monitored temperature at the inlet and outlet of solar air collector and solar irradiance on horizontal surface, for a winter period (August 30th to September 2nd, 2005), when the building is used by teachers and students.
Fig. 15. Monitored outdoor temperature, solar irradiance on horizontal surface, and indoor temperature for a summer period (November 15th to 18th, 2005), when the building is occupied by teachers and students.
daily energy consumption of 0.94 kWh/m2 day (3.4 MJ/m2 day) was measured, with mean indoor temperatures around 20.6 8C. Fig. 15 shows the results for summer (November 15th to 18th), when the building was under complete occupation and use. The indoor temperature during the occupancy period (8:00 a.m. to 19:00 p.m.) was always below the outdoor temperature and, the most part of the day, inside the comfort zone for developing countries. Mean indoor temperatures around 23 8C were common, which is a successful result considering that metabolic heat contributes with values as high as 5 8C to the indoor temperature. With higher outdoor temperatures, which are common in summer, the cooling system will supply fresh air. If we compare this summer behaviour with that of the original unmodified building, we conclude that the difference between mean outdoor and indoor temperatures was lowered from 10 8C (for the original building) to 3.4 8C (for the modified building), with a considerable lowering of the cooling loads in summer. Besides, for days with maxima temperatures around 30 8C, the indoor space is inside the comfort zone and the use of the cooling system can be avoided. This contributes with a reduction of 70% in the requirements of conventional energy. 4. Conclusions Numerical programs greatly facilitate the study of thermal transients in buildings, which is very important both for
evaluating the heating and/or cooling energy requirements and for achieving environmental comfort conditions. The study of building thermal performance by means of software simulation packages is crucial to face real world problems in the design of energy efficient buildings. A previous simulation stage was essential to predict the winter and summer thermal behaviour of the building, helping to choose among different materials and technology for the building envelope, glazed areas, shape, orientation, and so on. The analysis of the building thermal behaviour shows that the energy behaviour during winter months is encouraging, but warmer periods seem to be the ones that present greater difficulties. Our contribution is applicable to regions of developing countries where non-traditional materials and qualified workmanship are not available. The insulated envelope, direct and indirect solar gains, and air collectors, warrants the winter thermal comfort, with predicted savings of 50% in heating requirements respect to the conventional layout, without additional cost. When compared with Santa Rosa building (without heating and without students), we conclude that the difference between the mean outdoor and indoor temperature was increased from 4.6 8C (in the original building) to 8 8C (in the modified building). Besides, with similar daily energy consumptions around 1 kWh/m2 day, the mean indoor temperature was increased from 14 8C (original) to 24 8C (modified building). In summer, if we compare the
S.F. Larsen et al. / Energy and Buildings 40 (2008) 987–997
both buildings, we conclude that the difference between mean outdoor and indoor temperatures was lowered from 10 8C (for the original building) to 3.4 8C (for the modified building), which implies a reduction of 70% in the requirements of conventional energy for cooling. Besides, for days with maxima temperatures around 30 8C, the indoor space is inside the comfort zone and the use of the cooling system can be avoided. The qualitative and quantitative analyses carried out showed promising results related to the buildings energy behaviour, yet it cannot be denied that there are certain factors that have direct incidence on the optimal energy behaviour of buildings and their thermal comfort: (a) exogenous factors (the high variability of the external physical environment), (b) endogenous factors (constructive characteristics) and (c) socioenvironmental factors (life-styles and behaviour of users, associated to the active energy contribution and preferences of heating levels). The true challenge for solar energy architecture is to simplify the technology and construction processes so that costs would not be higher than those for conventional buildings, assuring users’ pro-active behaviour towards the building’s correct thermal management. To involve users in these improved practices is still another challenge. Acknowledgements The authors are grateful for support from projects PICT 2000 No. 13-09991, CIUNSa Nos. 1088 and 1332 of Universidad Nacional de Salta, Argentina, and PIP CONICET 6543. References [1] M. Wall, Energy-efficient terrace houses in Sweden Simulations and measurements, Energy and Buildings 38 (2006) 627–634. [2] F. Garde, L. Adelard, H. Boyer, C. Rat, Implementation and experimental survey of passive design specifications used in new low-cost housing under tropical climates, Energy and Buildings 36 (2004) 353–366. [3] N. Cardinale, F. Ruggiero, Energetic aspects of bioclimatic buildings in the Mediterranean area: a comparison between two different computation methods, Energy and Buildings 31 (2000) 55–63. [4] Y. Feng, Thermal design standards for energy efficiency of residential buildings in hot summer/cold winter zones, Energy and Buildings 36 (2004) 1309–1312. [5] M. Macias, A. Mateo, M. Schuler, E.M. Mitre, Application of night cooling concept to social housing design in dry hot climate, Energy and Buildings 38 (2006) 1104–1110. [6] J. Pfafferott, S. Herkel, M. Wambsganß, Design, monitoring and evaluation of a low energy office building with passive cooling by night ventilation, Energy and Buildings 36 (2004) 455–465. [7] H. Breesch, A. Bossaer, A. Janssens, Passive cooling in a low-energy office building, Solar Energy 79 (2005) 682–696. [8] S. Citherlet, J.A. Clarke, J. Hand, Integration in building physics simulation, Energy and Buildings 33 (2001) 451–461. [9] J. Kosny, E. Kossecka, Multi-dimensional heat transfer through complex building envelope assemblies in hourly energy simulation programs, Energy and Buildings 34 (2002) 445–454. [10] J.F. Karlsson, B. Moshfegh, Energy demand and indoor climate in a low energy building—changed control strategies and boundary conditions, Energy and Buildings 38 (2006) 315–326. [11] S.O. Jensen, Validation of building energy simulation programs: a methodology, Energy and Buildings 22 (1995) 133–144.
997
[12] http://www.eere.energy.gov/buildings/energyplus/. [13] T. Ayata, O. Yildiz, Investigating the potential use of natural ventilation in new building designs in Turkey, Energy and Buildings 38 (2006) 959–963. [14] M.L. Persson, A. Roos, M. Wall, Influence of window size on the energy balance of low energy houses, Energy and Buildings 38 (2006) 181–188. [15] M. Casermeiro, L. Saravia, Ca´lculo Te´rmico Horario de Edificios Solares Pasivos, in: Actas de 98 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, San Juan, Argentina, 1984, pp. 39–45. [16] A. Esteves, J. Ferna´ndez, M. Basso, J. Mitchel, C. de Rosa, Simulacio´n te´rmica de edificios: aplicacio´n de los modelos Quick y SIMEDIF, in: Actas de la 178 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, Rosario, Argentina, 1994, pp. 543–550. [17] A. Herna´ndez, G. Lesino, Ana´lisis de la performance te´rmica de un prototipo de vivienda liviana: monitoreo y simulacio´n macrodina´mica. Parte I, in: Actas de la 168 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, La Plata, Argentina, 1993, pp. 167–174. [18] A. Herna´ndez, G. Lesino, Simulacio´n mediante SIMEDIF del comportamiento te´rmico de un prototipo de vivienda liviana construido en la Universidad Nacional de Salta, Avances en Energı´as Renovables y Medio Ambiente 4 (2) (2000) 08.29–08.34. [19] S. Flores Larsen, A. Herna´ndez, G. Lesino, N. Salvo, Measurement and simulation of the thermal behaviour of a massive building with passive solar conditioning, in: Proceedings of VII International Building Simulation Congress 2001, Rı´o de Janeiro, Brazil, (2001), pp. 183–190. [20] S. Flores Larsen, G. Lesino, A new code for the hour-by-hour thermal behaviour simulation of buildings, in: Proceedings of VII International Building Simulation Congress 2001, Rı´o de Janeiro, Brazil, (2001), pp. 75–82. [21] S. Flores Larsen, C. Filippı´n, G. Lesino, Buried pipes for passive cooling: two cases of educational buildings in argentina, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [22] V. Garcı´a, A. Iriarte, G. Lesino, S. Flores Larsen, C. Matı´as, Special greenhouse for walnut grafting in arid regions, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [23] N. Baker, M. Standeven, Thermal comfort for free-running buildings, Energy and Buildings 23 (1996) 175–182. [24] M. Wall, Distribution of solar radiation in glazed spaces and adjacent buildings. A comparison of simulation program, Energy and Buildings 26 (1997) 129–135. [25] G. Lesino, R. Requena, L. Saravia, L. Fontanilla, R. Echazu´, Edificios solares en La Puna: estado de avance de su construccio´n y modelizacio´n, in: Actas de 78 Reunio´n de ASADES, Asociacio´n Argentina de Energı´a Solar, Rosario, Argentina, 1981, pp. 138–145. [26] C. De Rosa, M. Basso, A. Esteves, A. Pattini, J. Ferna´ndez, L. Cortegoso, J. Mitchell, M. Guisasola, G. Lesino, L. Saravia, Escuela solar te´cnico agraria en Mendoza. Ana´lisis del beneficio invernal de la incorporacio´n de un invernadero adosado, Energı´as Renovables y Medio Ambiente 1 (1996) 11–19. [27] G. Lesino, R. Requena, R. Caso, A. Herna´ndez, N. Salvo, Vivienda liviana aislada con servicios solares, Energı´as Renovables y Medio Ambiente 3 (1997) 47–54. [28] C. Filippı´n, S. Flores Larsen, A. Beascochea, G. Lesino, Response of conventional and energy-saving buildings to design and human dependent factors, Solar Energy 78 (3) (2005) 455–470. [29] C. Filippı´n, A. Beascochea, S. Flores Larsen, G. Lesino, A solar school building in a temperate region in Argentina, in: D.Y. Goswami, S. Vijayaraghaven, R. Campbell-Howe (Eds.), Solar World Congress ISES 2005, Orlando, USA, 2005, ISBN: 0-89553-177-1. [30] J.A. Duffie, W.A. Beckman, Solar Engineering of Thermal Processes, second ed., Wiley Interscience, New York, 1991. [31] C. Filippı´n, A. Beascochea, S. Flores Larsen, Auditorio bioclima´tico en la region central de Argentina, in: IV Congreso Latinoamericano sobre Confort y Comportamiento Te´rmico de los Edificios, Me´xico, 2005.