Intelligent Building Envelopes - Fad or Future? Øyvind Aschehoug, Professor NTNU, Trondheim, Norway; [email protected], http://www.ab.ntnu.no/bht/ansatte/aschehoug.html Inger Andresen, dr.ing., Senior Scientist SINTEF, Trondheim, Norway [email protected], http://www.sintef.no/content/page4.aspx?id=4278&EmpId=202 Tommy Kleiven, dr.ing., Research Scientist SINTEF, Trondheim, Norway [email protected], http://www.sintef.no/content/page4.aspx?id=4278&EmpId=745 Annemie Wyckmans, PhD Candidate NTNU, Trondheim, Norway; [email protected], http://www.ab.ntnu.no/bht/ansatte/wyckmans.html KEYWORDS: building, envelope, intelligence, SUMMARY: The annual solar energy received at the envelope surfaces of a building is in the same order as the energy needed for operating the building, even at high latitudes. With better utilisation of this energy, one should potentially be able to create buildings that are self-sufficient with energy. Much interest has recently been focussed on intelligent facades or intelligent envelopes, that by adaptive or responsive actions will make it possible to utilise more of this energy for building purposes, such as heating, cooling, ventilation, lighting and electricity supply .One example of such an intelligent building envelope is a double façade which incorporates natural ventilation systems. The term intelligent is, however, often used without any really deep understanding of the complexity required beyond common descriptions such as interactive, adaptive and responsive. It is easily understood that an intelligent envelope system must be dynamic, i.e. able to change its main functional parameters according to the dynamic demands of the changing situations. But in order to be truly intelligent, the system must also be able to learn from earlier experience and use this knowledge to cope with new situations. This paper summarizes research work within several areas related to the development of intelligent building envelopes. The intelligence concept as related to buildings has been studied in detailed, and applied to daylighting by specifying how the envelope should create optimal daylight conditions indoors, based on input from the external conditions and occupant preferences. Occupant responses to double envelope facades have been surveyed in two buildings in Norway, one with a simplified ventilation system incorporated in the facade, one with an integrated PV system. System analysis has been performed for high latitude climates to develop a procedure for developing optimal design through analysis of prioritized optimization criteria.

1. The intelligence concept Frequently used in building design, the term intelligent building envelope has become a common denominator for a certain type of built form that uses artificial intelligence to provide the indoor environment with dynamic heating, cooling, lighting and air supply, aiming to procure an optimal balance between occupant comfort and energy use. There exists a multitude of definitions and synonyms for the concept of ‘intelligent building envelope’ – (Wigginton & Harris 2002) for example list over 30 definitions of intelligence related to buildings and building envelopes. Building envelopes that response intelligently to the dynamic character of the available external energies and internal occupant demands should potentially be able to give substantial reductions in energy use and peak power demand. This concept for sustainable development has caught great interest in many countries. The basic concept is in the literature given many different names and definitions. Very general terms for the new class of facades are:

Advanced facades (distinguished from conventional ones) High-performance facades (assuming conventional facades are low-performance) Innovative facades (conventional facades are per definition no innovations) More in line with the dynamic character of such facades, these terms are commonly used: Smart facades (implying automated computer-based controls) Intelligent facades (normally understood as identical to “smart”) Active facades (which only means dynamic in character) Interactive facades (implying reactions to external situation and user demands) Responsive facades (normally understood as identical to “interactive”) The term “intelligent”, however, involves the danger of ending up with a meaningless quality label when applied to the building envelope. In order to answer this question, the path of human intelligence has been explored. According to (Piaget 1967) and others, intelligent behaviour can be explained as a form of adaptation or structured interaction with the environment, involving the use of mental models; this adaptation enables a subject to solve problems and cope with new situations that occur in the environment. Adaptation occurs by means of a continuous range of processes varying from basic sensori-motor adaptation to learning skills and sudden insight. Mental models are constructed and updated continuously in interaction with the environment, providing regularity in the connection between stimulus and response. Intelligence can be regarded as a form of interaction with and adaptation to the environment, induced by a variety of psychical processes ranging from trial-and-error to insight. Transferring this description to the context of building design, an intelligent building envelope can be defined in the following manner: An intelligent building envelope adapts itself to its environment by means of perception, reasoning and action. This innate adaptiveness enables an intelligent building envelope to cope with new situations and solve problems that arise in its interaction with the environment. This operational definition, based on the psychological development of intelligent behaviour in human beings, is chosen because it relates the term ‘intelligence’ to concrete psychical processes rather than to the more subjective and diffuse concepts of rationality, sensibility and good judgment often associated with human intelligence. In order to systematically identify intelligent behaviour for a building envelope, a model of double-loop learning has been applied (Sterman 2000). This type of system analysis is used to map areas of conflict and synergy related to the intelligent behaviour of a building envelope, and to assess the consequence patterns this generates. An intelligent building envelope’s contribution to an indoor environment supportive of human needs can be analysed in five steps, corresponding to stages in double-loop learning: • • • • •

Sensory perception Mental model Assessment of information and feedback Strategic thinking Implementation

Given its ability to adapt to the environment by means of perception, reasoning and action, an intelligent building envelope may be expected to fulfil three objectives: to cope with a variable environment, to cope with a conflictive environment, and to cope with human behaviour. Adaptiveness enables an intelligent building envelope to cope with new and varying situations in its environment. This environment can be regarded as composed of three main elements: an outdoor element with climate and site conditions, an indoor element contained within the shell of the building envelope, and a third element consisting of the building users, their preferences and behaviour. An intelligent building envelope needs to solve problems that occur in its interaction with the environment; the tasks required of the envelope, may sometimes be conflicting. The envelope thus needs to make trade-offs according to a prevalent set of priorities, and find an optimal solution to all of the tasks, rather than the perfect solution to one particular task. The building occupant forms a particular point of focus in the variable and conflictive environment an intelligent building envelope is confronted with; the acceptance of an envelope’s behaviour by the building occupant is of the utmost importance. In studies on user satisfaction with lighting control systems, it is often found that the inability to

overrule the system is the users’ most important complaint. Dissatisfied users may actively counteract the envelope’s strategies and even try to sabotage its operation. The use of glass in the building envelope brings forth variable and potentially conflictive demands of transparency versus privacy, of openness versus insulation, of access to daylight versus solar shading. This poses architectural and technical challenges in material use, form and composition. An intelligent building envelope is expected to manage these apparent contradictions under diverse circumstances with a desirable outcome.

2. System analysis of intelligent envelopes In order to find the most optimum façade configuration for a building in a given context, a lot of different design criteria need to be taken into account. Among the most important ones are indoor comfort requirements such as thermal conditions, lighting and noise, costs, architectural issues such as scale and proportion, and energy use for operating the building. However, all of the studies reported in the literature only include a limited selection of these optimisation criteria. As such, they seem to confirm that a numerical global optimisation is not possible. There are too many variables, some of which are difficult to quantify, and it is not possible to make a numerical model that includes them all. Also, there is almost always a trade-off between conflicting criteria (e.g. cost versus comfort), and the decision-maker needs to make priorities in order to reach a solution. It is also quite obvious that there exists no global optimum facade configuration that could be applied to any building anywhere in the world. Every building is unique; it has a unique set of users and it is set in a unique context.

2.1 The system analysis approach Systems analysis is not a well-defined methodology or distinct field, but several authors describe a procedure that involves some distinct steps as illustrated in figure 1. The analysis starts out with formulating the problem, identifying and describing boundaries and constraints, as well as objective and values. It then proceeds with identifying, designing and screening of alternatives. The next step involves building and using models for predicting the consequences of each alternative solution, and at the end the alternatives are compared and ranked. During the analysis, new issues may come up that makes it necessary to go back and revise some of the previous assumptions or conditions. The initial analysis is characterized by craftsmanship, expert knowledge, experience and hunches. Only in the last fine tuning phase, one may make good use of mathematical models. At this stage, the problems are usually well defined and setting the details may be ideal for formal procedures for optimisation.

2.2 Design criteria The first task of the system analysis involves selecting, describing and prioritising design criteria. Money is usually the prime driving force of human beings. However, we do not normally have access to costs for all items in an analysis. It can be argued that the economic of the façade will appear through the productivity of the people behind the façade. Therefore, the main design criterion is human well-being. The PROBE-study (Post-Occupancy Review of Building Engineering) that was undertaken in the UK during 199598 is perhaps the most comprehensive occupant survey that has been undertaken in recent years (Leaman and Bordass, 1999). The PROBE study contains several indications that façade design is very important for occupant well-being. For example, the study concluded that high levels of occupant satisfaction were easier to achieve when the following features were present (Leaman and Bordass, 1999): • shallower plan forms and depths of space (workstations typically 6 m or less from a window); • thermal mass (provided the acoustics are satisfactory); • stable and comfortable thermal conditions; • freedom from distracting noise; • air infiltration under control; • operable windows close to the users; • views out; • effective controls with clear, usable interfaces.

in itia t i o n

F o r m u la t in g t h e p r o b le m

b o u n d a r ie s c o n s t r a in t s

o b je c tiv e s

v a lu e s c r i t e r ia

Id e n t i f y i n g , d e s ig n in g a n d s c r e e n in g o f a lt e r n a t i v e s

a lte r n a t i v e s

B u i l d i n g a n d u s i n g m o d e ls f o r p r e d i c t i n g t h e consequences

F o r e c a s tin g fu t u r e c o n te x t s

consequences ( im p a c t s )

C o m p a r i n g a n d r a n k in g a l t e r n a t i v e s

c o m m u n ic a t i n g re s u lts

FIG. 1: The systems analysis procedure with iteration loops (Findeisen and Quade 1985). The PROBE study proves that daylight and view is very important for occupant well-being and productivity. Also access to control of own environment, i.e. thermal comfort, is a major factor. In addition comes the more technical design parameters, such as operation, flexibility and energy use.

2.3 A case study analysis A case study façade was analysed in order to test the system analysis approach. The building is an office located in a Nordic climate, Oslo 59oN, with one exterior façade facing south. In order to structure the analysis, it was centred around 7 design issues that together is believed to cover the PROBE parameters: 1: Window-to-wall area ratio 2: Window geometry and location 3: Type of glazing and shading 4: Thermal insulation of wall 5: Energy production 6: Energy storage 7: Control integration Due to the space constraints of this paper, we have to jump to the last part of the analysis where the first 4 design issues have already been evaluated, and a preliminary potential solution has been specified. The preliminary facade design looks like the one illustrated in FIG. 2. The facade has quite a large window area that covers the entire with of the facade. This layout is a result of the high emphasis on view, daylight and flexibility. The window-to-wall area ratio is 0.75. The windows have a total U-value of 0.7 W/(m2K), the total solar energy transmission is 0.46 and the daylight transmission is 0.70 (triple glazing with two LE-coatings and krypton gas filling). The opaque wall has a Uvalue of 0.22 (20 cm of mineral wool). The windows are separated into two zones; the lower view zone is equipped with automated exterior venetian blinds with user override. The upper daylight zone has a daylight system consisting of reflecting venetian blinds that redirects the daylight towards the ceiling. This option offers maximum flexibility, maximum energy savings, and good visual quality in the room. Thyholt et al (1999) have shown that lighting energy savings in the order of 60% could be achieved with such a system combined with a dimming control strategy. Space cooling is eliminated.

Upper zone: Daylight

Middle zone: View + glare control

Lower zone: Opaque

FIG. 2: The preliminary facade layout. The energy use for different purposes is then analysed with the simulation program SCIAQ Pro for a conventional façade, an optimised façade according to FIG.2, and an optimised double façade incorporating natural ventilation. The energy demands are illustrated in FIG. 3.

180 160

Eq u ipment

140

Lighting

120

Fans

100

DHW

80

Coolin g c o il

60

Coolin g

40

Heatin g c o il Room heating

20 0 c o n v e n tional

optimiz e d s in g le

optimiz e d d o u b le

FIG. 3: Yearly energy loads (kWh/m2 floor area) for the office with an ”optimised” facade and an ”optimised” double facade, compared to a standard new office facade. The energy results for the double façade are quite promising, and a daylighting analysis shows that the impact on daylight conditions is quite acceptable. The lower façade zone can be used for energy generation purposes, which would mean incorporating photovoltaic cells. The cost efficiency of this, in fact for the whole double façade concept, is of course the important unanswered questions. In order to fulfill all the important design criteria, the façade needs an advanced control system, possibly based on fuzzy logic, in order to maintain both optimal operation and occupant satisfaction.

3. Post occupancy evaluations of double facades The implementation of double facades in both new and retrofit buildings has seen broad application in recent years. A great deal of research has been carried out that focus on the technical and economical aspects concerned with this type of facade. Little research has focused on the user satisfaction, however. The user experiences from two different office buildings with double facades in Norway have recently been studied. The two buildings investigated here are the Hamar Town Hall (2001) in Hamar and a retrofit of an office building at the university campus of NTNU (2000) in Trondheim, Norway.

3.1 The Hamar Town Hall The Hamar Town hall was designed by .Snøhetta Architects AS and completed in 2001. The building is 5 stories tall and has a gross floor area of 10 500 square metres. The depth of the plan is approximately 17m and the floor to ceiling height is 3m. The town hall has two double facades that face south and north. The main reason for choosing a double façade in the north façade was to screen the interior from the traffic right outside this façade. The south

façade is a double façade to preheat ventilation air during some parts of the year, and to protect the solar blinds in the cavity.

FIG. 4: The northern double facade (left) and the southern double façade (right) of the Hamar Town Hall (2001) designed by the Norwegian architects Snøhetta AS. There were two sources of information for this investigation: a questionnaire was sent out to all occupants sitting next to one of the two double facades, and open-ended interviews were carried out with the architect and the caretaker. The questionnaire was divided into 6 parts, covering: the office in general, windows and view to the out side, direct sunlight in the office, daylight conditions, the double façade, and personal data. Of the 22 occupants located next to the double facades, 19 participated in the survey. The greater majority of the respondents are satisfied with the indoor climate and the daylighting. There were some negative remarks about the solar shading system. On the north side, all the occupants were satisfied with the protection against traffic noise, and sound propagation through the cavity was no problem. The most surprising result was that about 3/4 of the respondents experienced no difference between working behind a double facade, compared to a conventional facade. They did not feel that they had a lower degree of contact with the exterior. North facade

South facade

Not considered

Not considered

Less contact with the exterior More contact with the exterior

Less contact with the exterior More contact with the exterior

FIG. 5: The pies show the response to the following question for the north- (left) and south facade (right) respectively: How will you rate your physical contact to the outside (the possibility to "reach the outside", i.e. by opening windows) now that you are located adjacent to a double facade?

3.2 The BP Solar façade at NTNU BP Norway and BP Solar have commissioned research units at the Norwegian University of Science and Technology, NTNU and the Foundation for Scientific and Industrial Research at NTNU, SINTEF, to develop a building system for facades, based on photovoltaic (PV) solar cells. A prototype solar façade system, combining a double façade with a building-integrated PV system, was constructed on an existing university building at the Natural Sciences and Technology campus of NTNU in the spring of 2000. The façade area is 400 m2, of which around 100 m2 net area is covered with mono-crystalline PV cells laminated between 2 sheets of glass. The cavity behind can be vented, the hatches is controlled by cavity temperature. mere data????

FIG. 6: The southern-aspect double facade of the office building at the NTNU campus (2000), designed by SINTEF/ NTNU. PV cells are integrated in the outer skin of the double façade at each floor slab. To find out how the occupants reacted to the new situation, questionnaire surveys where performed before construction and after construction, both during winter and summer conditions. The questionnaire covered seven major aspects of the indoor conditions in addition to a general question about the satisfaction with the indoor climate: indoor climate (a list of parameters was given), temperature, blinds, air quality, window venting, and daylighting. In the time span of the surveys, the number of building occupants relevant to the survey was reduced from 28 to 9 due to a relocation of their work stations. Therefore, even though the response rate was to the questionnaires was high (26/28, 16/23, 9/9), the statistical significance of the survey results is limited. The pre-construction survey About half of the respondents were dissatisfied with the indoor climate, most likely because a new ventilation system was under completion at the time of the survey and created a lot of disturbance and dust. About 25% of the occupants considered their room temperature to be a problem, and 30% of dissatisfaction was related to the air quality. The problem reported was too high temperatures, and the air quality problem is probably related to the temperature reactions. The post-construction survey winter The results from the winter survey seem to confirm that the new façade has improved some of the indoor conditions, while others are unchanged. Drafts, room temperature conditions, and air quality seem to have improved somewhat, but for most factors, the large majority of the occupants report no change. As one could expect, some occupants (20%) replied that the new facades impaired the view, and about the same number reported that shadow patterns, presumably from the PV cells, were occasionally an inconvenience. The daylighting, however, was judged to be sufficient by the same percentage as before construction The post-construction survey summer With only 9 replies it is difficult to draw very firm conclusions from this survey. On many questions the reactions are much the same for the winter and the summer season. This is the overall impression for daylighting, view and shadow patterns, glare due to direct sun, and the use of blinds. The inconvenience from noise from equipment and neighbouring offices is also much at the same level. For some other factors, the replies for the summer situations are much the same as for the pre-construction stage, and different from the winter results. One example is the inconvenience by noise from outdoor, probably because in the summer situation the façade vents are open most of the time. The most important problems are related to thermal comfort and indoor air quality. On the general question about satisfaction with the indoor climate, the percentage dissatisfied fell from 45% at the pre-construction stage to 5 percent for the winter condition, but came back to 35% for the summer condition. And 5 of the 9 occupants expressed that the new façade had impaired the indoor climate. About half of the few occupants that have experienced all the surveys report that during summer, the ventilation is insufficient and the building overheats. This is due to excess solar gains that under some conditions cannot be vented out through the cavity as designed. Under unfavourable conditions - a combination of direct sun, high external temperature, and wind from the north sector - the stack effect expected to vent the cavity is not strong enough to counter the wind blowing down into the cavity. This explanation is substantiated by the fact that the dissatisfied occupants were mainly situated on the 4th floor.

1a

100 % 90 % 80 % 70 % 60 % 50 % 40 % 30 % 20 % 10 % 0% pre-constr.

satisfied

post-constr. post-constr. (w) (s) neutral

dissatisfied

FIG. 7: Percentage distribution of the replies to the question:” Generally speaking, are you satisfied with the indoor climate in your office?”

4. Conclusions Research on intelligent building envelopes has proved that the concept offer many opportunities for exploiting external energies for the purpose of creating an optimal indoor environment for the occupants. The envelope surfaces can incorporate solar systems for thermal and electric energy, daylighting systems, and natural ventilation. A promising option is the double façade, where the outer glass skin cover a cavity that is integrated in the ventilation concept. There are, however, many problems to be solved before this type of technology becomes commonplace. There are no comprehensive analysis method available that covers all the important parameters, and the control logic needed is also a challenge. There are also very little information available about investment and operational costs .

5. References Andresen, I. and Ø. Aschehoug: “System analysis of smart facades”. Conference “Glass in Buildings”, Bath (UK), 2005-04-07/08. Aschehoug, Ø, Hestnes, A.G., Matusiak, B., Lien, A.G., Stang, J. and D. Bell, (2000) “BP Amoco Solar Skin – A Double Façade With PV”. Proceedings. Eurosun 2000, Copenhagen June 2000. Findeisen, W. and E.S. Quade (1985): “The Methodology of System Analysis: An Introduction and Overview”. Handbook of System Analysis. H.J.Miser and E.S.Quade. John Wiley & Sons, Chichester. Kleiven, T., Aschehoug, Ø. and A. Wyckmans: “Double facades at high latitudes - some user experiences”. Conference “Glass in Buildings”, Bath (UK), 2005-04-07/08. Leaman, A. and B. Bordass (1999). ‘The PROBE occupant survey and their implications’, CIBSE National Conference. Piaget, J (1967) ‘The psychology of intelligence’, Routledge & Kegan Paul Ltd, London. Sterman, J D (2000) ‘Business Dynamics: Systems Thinking and Modelling for a Complex World’, McGraw-Hill, Boston Wigginton, M, Harris, J (2002) ‘Intelligent skins’, Butterworth-Heinemann, Oxford. Wyckmans, A., Aschehoug, Ø. and A.G. Hestnes: “The intelligent building envelope - concept and qualifications”. Conference “Glass in Buildings”, Bath (UK), 2005-04-07/08.