Conference PaperPDF Available

Structure and daylighting performance comparisons of Heinz Isler’s roof shell based on variations in parametrically derived multi opening topologies

Abstract and Figures

The Design of building envelopes consists of two major phases in two different disciplines: structural design, and climatic design. Additional performance criteria such as acoustics may also be considered based on the program of the space. In general, a mono-disciplinary approach to design evaluates performance in each discipline detached from other disciplines and sometimes in a hierarchical order. This may be due to disjointed parameters that affect the design and optimization process, as well as different expertise of the designers and engineers in each field. Considering some structures such as post and beam systems, this method seems appropriate since each discipline has little influence on other fields. This provides relatively adequate freedom for the designers to decide about different design variables, such as size and orientation of the apertures, material of the cladding and the structure. However, there are some other building envelopes in which making a design decision in one field largely affects the performance in other fields. These building envelopes which cover the architectural space have a high potential for an interdisciplinary design approach. Heinz Isler is an engineer who has designed extremely efficient shells with excellent performance over time. Considering his works, there are a few shell structures which have one or multiple apertures, mainly designed to introduce daylighting into the space. But how have these apertures influenced the force flow and structural performance of the shell? What daylighting levels have they provided for the space? And by manipulating the size and number of these apertures, how may the structural and daylighting performance of a shell vary? This paper intends to look at a perforated concrete shell designed by Heinz Isler and assess its structural and daylighting performance. Then, the size, number and location of the openings is altered in order to observe the effect on the structural and daylighting performance of the shell. Rhino and Grasshopper are used as the modelling platform, while Karamba, which is a plugin for Rhino, is employed for assessing the structural performance, and the DIVA plugin for Rhino is employed to assess the daylighting performance. Finally, a comparison between different topologies is made using different numeric indicators. For structural performance, deflection, weight and maximum von Mises stress levels are considered, along with Daylight Autonomy on horizontal and vertical planes as the daylighting numeric indicator. The goal of this comparative study is to demonstrate tradeoffs among various performance criteria, regarding the relation between topology, structural performance and daylighting performance, and may be used by designers who consider multiple performance criteria in early design phases. Keywords: Shell structures, integrated design, parametric design, interdisciplinary design, structural performance, daylighting performance.
Content may be subject to copyright.
Proceedings of the International Association for Shell and Spatial Structures (IASS)
Symposium 2015, Amsterdam
Future Visions
17 - 20 August 2015, Amsterdam, The Netherlands
Structure and daylighting performance comparisons of
Heinz Isler’s roof shell based on variations in
parametrically derived multi opening topologies
Niloufar EMAMI*
*PhD Candidate, University of Michigan
2000 Bonisteel Blvd., Ann Arbor, MI
nemami@umich.edu
Abstract
The Design of building envelopes consists of two major phases in two different disciplines: structural
design, and climatic design. Additional performance criteria such as acoustics may also be considered
based on the program of the space. In general, a mono-disciplinary approach to design evaluates
performance in each discipline detached from other disciplines and sometimes in a hierarchical order.
This may be due to disjointed parameters that affect the design and optimization process, as well as
different expertise of the designers and engineers in each field. Considering some structures such as
post and beam systems, this method seems appropriate since each discipline has little influence on
other fields. This provides relatively adequate freedom for the designers to decide about different
design variables, such as size and orientation of the apertures, material of the cladding and the
structure. However, there are some other building envelopes in which making a design decision in one
field largely affects the performance in other fields. These building envelopes which cover the
architectural space have a high potential for an interdisciplinary design approach.
Heinz Isler is an engineer who has designed extremely efficient shells with excellent performance
over time. Considering his works, there are a few shell structures which have one or multiple
apertures, mainly designed to introduce daylighting into the space. But how have these apertures
influenced the force flow and structural performance of the shell? What daylighting levels have they
provided for the space? And by manipulating the size and number of these apertures, how may the
structural and daylighting performance of a shell vary? This paper intends to look at a perforated
concrete shell designed by Heinz Isler and assess its structural and daylighting performance. Then, the
size, number and location of the openings is altered in order to observe the effect on the structural and
daylighting performance of the shell. Rhino and Grasshopper are used as the modelling platform,
while Karamba, which is a plugin for Rhino, is employed for assessing the structural performance, and
the DIVA plugin for Rhino is employed to assess the daylighting performance. Finally, a comparison
between different topologies is made using different numeric indicators. For structural performance,
deflection, weight and maximum von Mises stress levels are considered, along with Daylight
Autonomy on horizontal and vertical planes as the daylighting numeric indicator. The goal of this
comparative study is to demonstrate tradeoffs among various performance criteria, regarding the
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
relation between topology, structural performance and daylighting performance, and may be used by
designers who consider multiple performance criteria in early design phases.
Keywords: Shell structures, integrated design, parametric design, interdisciplinary design, structural
performance, daylighting performance.
1. Introduction
Robert le Ricolais' influential structural engineering work from the 1960's was guided by his belief
that “The art of structure is where to put the holes”. The paradox describes the removal of material
that weakens a structure but also makes it lighter, thus making it more efficient. In the case of
continuous shell structures, a perforated shell poses a unique opportunity for interdisciplinary
integrated design since providing the apertures introduces daylight to the building, but at the same
time alters the force flow in the structure and affects its structural performance. With regards to
integrating architecture with engineering, Berger states that the structural form defines the structural
shape of the building on the outside and the form of the space on the inside. There is no longer any
distinction between engineering and architecture”. (Bertin [3])
In this research, the perforated concrete shell is referred to as a building envelope which encloses the
inside structurally as well as providing natural daylight for the interior space. This paper intends to
simulate the COOP shell designed by Isler, and assess its structural and daylighting performance
simultaneously. Then it aims to model design variations for the same envelope and compare the
performance of the original shell with design alternatives. In a study by Emami et al., a perforated
concrete panel for a façade was analyzed in terms of its daylighting and structural performance
adopting a generative design approach with ParaGen. (Emami, Khodadadi, and von Buelow[7]).
Looking at both structural and climatic performance, it is required to converge towards geometric
solutions that are high performance overall and at the same time for each individual discipline.
Emphasis is placed on the connections between daylighting aspects and structural morphology
regarding the geometry as the key interdisciplinary interface. For each field, measurable indicators
have been identified. It is worth noting that this research acknowledges the necessity of domain
specialization
and expertise, and builds upon earlier design integration research.
2. Heinz Isler
Isler built many projects, and his works have mainly been studied in terms of their form finding
methods and excellent structural performance. However, the role of available natural daylighting has
been underestimated in the study of his projects. In most of his projects, the natural daylighting enters
the interior through vertical, clear, glass walls, which leaves the main structure un-effected. In some
other projects, apertures appear in the shell surface and act as a skylight (Ramm[10]). Among Isler’s
perforated shell structures, Sicli Factory in Geneva built in 1969 (Figure 1) is an example in which
Isler made superior use of experimentation with different materials and cutting patterns resulting in
structures (Adriaenssens et al.[2]). Another project is Steinkirche” in Cazis in which Isler was
involved in the project’s development (Figure 2). He was not asked to find the form of the shell, but
was given a predetermined form by the architect. Although Isler was aware of the lack of structural
optimization of the forms, but with time, came to accept. Due to his belief in the power of nature to
create efficient structural surfaces, in this case through growth, he designed his shells based on
observation of expansion of foam (Chilton, [5]).
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
Figure 1: Form finding of Sicli Factory (Ramm[10]). Sicli Factory in Geneva (Segal[13])
Figure 2: Steinkirche, Cazis, (Chilton,[5])
One of his constructed shells is COOP Warehouse Wangen bei Olten (1960) which was form found by
inflation (Abel and Chilton[1]). This building of 54.6 x 58.8 meters has a shell rising at the center to 9
meters above the supports while the highest point of the shell is at over 15 meters above ground level
(Figure 3). Six intermediate supports are provided along each edge of the shell giving a column free
area of 3200 square meters, while 17 domed roof lights that represent 4% of the total roof area are
arranged to bring adequate and even daylight in to the space (Chilton, Heinz Isler[4]).
Given the size of the shell for COOP storage center, Isler was very concerned about its buckling
behavior due to the fact that the parts of the large-span shell were close to cylindrical in form, a shape
that has a much lower buckling resistance than Isler’s normal double-curved surfaces. Based on
similar forms, Isler knew that the loads would be carried in arch-like sections of the shell, from corner
to center along the diagonals and these arches lean against each other. He expected local buckling to
occur in the surface underneath the arches near the edge beam because that was nearly cylindrical
(thus weaker in buckling), and not in the double-curved arch sections (Chilton, Heinz Isler[4]).
Designing the shell thickness, Isler thickened it to 150 mm in the critical areas to get a safety factor of
3, assuming that all factors were unfavorable for the shell. When the shell was constructed, everything
was in favor (the weather, concrete and scaffolding), therefore the actual safety factor increased to 5.
The pattern of 4 and 4.5 meter diameter openings is so organized that the corner to corner arching of
the shell is not interrupted. This bubble shell is made of high quality concrete with double-layer
reinforcement (Chilton, Heinz Isler[4]).
Figure 3: COOP storage and distribution center, (Ramm[10]), (Chilton, Heinz Isler[4])
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
3. Research design and methodology
This research intends to simulate the COOP center shell designed by Isler, and assess its structural and
daylighting performance simultaneously. The original placement of roof domes is presented in figure
4, and the arches on which the apertures are placed have been tentatively drawn on the plan view
(Figure 4). The geometric properties of the shell are summarized in Figure 5.
Fig. 4: elevation (left)(Chilton, Heinz Isler), the plan view presented with superimposed arches(right)
Length
(m)
Width (m) Maximum
Shell
thickness
(mm)
Shell rising
above the
supports
(m)
Shell rising
above
ground (m)
Number
of circular
apertures
Radius of
circular
apertures
(m)
Calculated
Perforatio
n ratio
58.8 54.6 150 9 15 17 2 and 2.25 4%
Figure 5: Geometric properties of Coop Warehouse, (Chilton, Heinz Isler[4])
This research studies the model of a shell with the same dimensions as the Coop Warehouse
maintaining the same perforation ratio, while the number and the diameter of the apertures are altered.
In the designed alternatives, the global geometry of the shell including its span-to-rise ratio and shell
thickness are not varied; however, the topology is being altered. The variables and constants are
presented in Figure 6. On a separate note, although the perforation ratio in the Coop Warehouse is
mentioned to be 4%, the calculated perforation ratio considering the projected area of the curves (and
not the flat curves) divided by the shell surface area is calculated to be 6.5%. This has been opted as
the target perforation ratio for design alternatives. The meshing and boundary conditions are also
identical in all cases. Based on the constants and variables, nine cases are designed as alternatives to
conduct a numerical study (Figure 7).
Constants Span-to-rise ratio: 54.6 m to 9 m / 58.8 m to 9 m
Shell thickness: 15 cm
Perforation ratio: 6.55% ± 0.08%
Variables Number of circular apertures (ranges from 1 to 31)
Radius of apertures (ranges from 1.5m to 8.5 m)
Figure 6: Design parameters including constants and variables
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
Case 1 Case 2 Case 3 Case 4
(rhombic
pattern)
Case 5
(circular
pattern)
Case 6 Case 7
(original
shell)
Case 8 Case 9
Apertures
Placement
Aperture no 1 5 7 9 9 13 17 21 31
radius (m) 8.5 3.75 3.1 2.75 2.75 2.3 2 1.8 1.5
Perforation
ratio (%)
6.54% 6.53% 6.58% 6.46% 6.63% 6.58% 6.56% 6.51% 6.52%
Figure 7: Design alternatives
Rhinoceros, which is a Nurbs modeling software, along with Grasshopper and Kangaroo, have been
opted as the modeling and form finding platform. This allows parametric design and alteration of
different design variations. Karamba which is a structural analysis plugin for Grasshopper has been
chosen for the structural analysis. It is worth noting that the results have been calibrated with ANSYS
which uses a FEM method. DIVA for Grasshopper has been used for daylighting analysis (figure 8).
Figure 8: Simulation workflow
The metrics used to evaluate structural performance are “maximum deflection” and “buckling load
factor under uniform loading” as a stability criteria, and “maximum von Mises stress” as a strength
criteria. Since perforation ratio, thickness and span-to-rise ratio are not varied, weight is constant thus
not a criteria in this study. The criteria for daylighting performance is “Daylight Autonomy (DA)” on
horizontal and vertical planes. The structural and daylighting performance were assessed in different
stages, and then the results were analyzed. Finally, having the original shell as a reference point, other
scenarios are ranked below or above it based on performance criteria.
4. Mono-disciplinary performance assessment: Structural performance
In this phase, the structural performance of design alternatives namely maximum deformation,
maximum von Mises stress and buckling load factor under uniform load were assessed using both
Karamba in Grasshopper and ANSYS. This allowed a comparison of the results obtained from the two
programs. Reinforced concrete was assigned to the shell as the material with “density= 2000 kg/m^3”,
“Modulus of Elasticity= 2.1*10
4
MPa”, “Shear Modulus = 8.9*10
3
MPa”, “Poisson ratio= 0.2”,
Form-Finding
Rhinoceros
Grasshopper
Kangaroo
Analysis
DIVA for
Grasshopper
Karamba for
Grasshopper
Validation
ANSYS
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
“Compressive strength= 28 MPa” and “Tensile strength= 2.8 MPa”. The four edge lines were selected
as supports which were restricted in all three axes. Then the shell was subject to self-weight under
earth-gravity and a global pressure of -1 KPa in the z axis (vertical, downward).
4.1 Structural analysis of a shell with no perforation
The first modelled case for structural performance assessment was the shell with no aperture. This is
referred as “case 0” in forthcoming tables, and serves as a case for comparison. The aim of evaluating
“case 0” is to review the behavior of a typical form-found shell by looking at maximum deformation
and force flow lines. This was done in both Karamba and ANSYS, in order to observe different
numerical outcome and graphical representations for the same structure. The maximum deformation is
located at the center point of the shell, and is depicted with red color in ANSYS which then transits to
blue when the deformation is reduced. Looking at stress plots from ANSYS, maximum tension occurs
in the edge lines and in the lower layer of the structure, whereas the maximum compression occurs at
the four corners. The structural analysis of a non-perforated shell is presented in Figure 9.
Figure 9: from left to right, maximum principal, minimum principal and vector principal stress plots
4.2. Calibration of structural results among ANSYS and Karamba
In this phase, the performance of all nine cases were assessed in both ANSYS and Karamba. The
graphical representations of the outcomes have also been retrieved to better understand any related or
dissimilar results. The color scale of Karamba was altered to be in line with ANSYS (Figure 10)
KARAMBA
Deformation
ANSYS deformation
KARAMBA von
Mises stress
ANSYS von Mises
stress (top layer)
Figure 10: Graphical representation comparison among Karamba and ANSYS
Comparing results from Karamba and ANSYS, deformation and maximum von Mises stresses were
extracted and compared as “single-criteria” assessments (Figure 11). In terms of maximum deflection,
there was good agreement between the two; as the number of the apertures was increased, and their
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
radius was decreased, the maximum deflection was decreased. Comparing maximum von Mises stress
levels between two softwares, they are not quite compatible, neither in the absolute values nor in their
trend. More in depth study is required to examine the element type, meshing and calculation
methodologies in these two softwares. For this study though, only deformation which is validated
through ANSYS is considered. Considering buckling load factor, unfortunately Karamba’s
Eigenvalues do not correspond to the load-factors for buckling. However, the buckling load factor was
calculated in ANSYS for a more in depth study. The buckling load factor for the original shell
designed by Heinz Isler (case 7) is calculated to be 11.4, 11.5 and 11.6 in the first three buckling
modes. If divided by 5, a buckling safety factor of around 2 is retrieved. Calculating this factor for
other cases, all but one fall above 2 hence considered acceptable compared to the original case. The
only unacceptable case is the first case with a large oculus in the middle. It is noticeable that the shell
with no perforation has the highest buckling-load safety factor of 3, and case 9 is ranked second with a
safety factor of 2.5. (Figure 12)
Figure 11: Comparing max. deflection and max. von Mises stress levels in Karamba and ANSYS
Figure 12: Buckling load safety factor Figure 13: Deformation versus maximum von Mises stress
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
In the next step, all nine cases were plotted on a graph with deformation on the horizontal axis and
von Mises stress on the vertical axis retrieved from ANSYS. The goal is to identify cases with least
deformation and lease von Mises stress levels (Figure 13). The original shell (case 07) with deflection
less than 0.425 cm and von Mises stress levels less than 0.43 KN/cm^2 was set as the base point.
Looking at the graph, case 9 has the least deformation and von Mises stress levels, very close to the
un-perforated shell (case 0) and its maximum deflection is even less than the unperforated shell. This
can be explained by pointing to the fact that since case 9 is perforated, it weighs less thus deflects less.
Case 8 and 6 both fit in the opted range as well.
4.3. Observation of principal stress lines in different topologies
Principal stress lines from Ansys and force flow lines from Karamba were extracted in this stage to
study how they were affected by introducing apertures in the shell. Although these are local effects
which should be solved locally by providing reinforcement for instance, but studying them will be an
aid in understanding the structural performance of a perforated shell. Force flow lines or load paths
illustrate the load distribution in structures. These are “useful if one wants to strengthen a structure
with linear element such as fibers and align them with force flow lines to get the most effective input”
(Preisinger[9]). Force flow lines are not the same as principal stress lines because the latter lack the
property of constant force between adjacent lines. Looking at figure 14, tension stresses are presented
in the edge lines of the original shell. But as soon as apertures are introduced, tension stresses start to
develop around the apertures too. They affect the force flow to be concentrated around the apertures.
On one hand, these areas would need some reinforcement, on the other hand, the areas with less
concentrated lines can have reduced thickness to save on material. From another point of view, if the
first case with no aperture is considered as a base case, then the other cases showing force flow lines
can be compared to it. In this regard, case 9 with 31 small apertures represents a force flow very
similar to case 1, and can be assumed to behave close to an un-perforated shell, and thus exhibit a
better structural performance. Recording the deflection values, case 9 is representing the minimum
deflection among all. Observing a general trend, the shells with higher number of smaller apertures
perform better than the shells with fewer and larger apertures.
Case 00 Case01 Case07(original case) Case 09
Figure 14: force flow lines from Karamba and vector principal lines from Ansys
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
5. Mono-disciplinary performance assessment: Daylighting performance
5.1. Daylighting analysis
In this phase, the daylighting performance of design alternatives is assessed using DIVA for
Grasshopper. DIVA links validated performance simulation engines to Rhino, namely "Radiance" for
visualizations, "Daysim" for climate-based metrics, and "Energy Plus" for thermal loads
(Reinhart[11]). Since the original site of COOP warehouse is in Wangen bei Olten located in
Switzerland, Geneve was chosen as the closest site with available weather data; its weather files were
downloaded from DOE (U.S Department of energy) website. The assigned materials are 80%
reflectance general ceiling, 88% reflectance single pane glazing, 50% reflectance general interior
walls, 20% reflectance generic floor and 20% reflectance outside ground. These have been selected
from the default material library of DIVA. Daylighting performance of the space is assessed using a
“climate-based metric” namely Daylight Autonomy. Daylight Autonomy (DA) at a sensor is defined
as the “percentage of the occupied times of the year when the minimum illuminance requirement at
the sensor is met by daylight alone” (Reinhart, F, Mardaljevic, and Rogers[12]). In this research,
Daylight Autonomy is calculated throughout a year from 8 a.m. to 6 p.m. weekdays; and has been set
to 300 lux as the satisfying minimum daylight level for a warehouse. Therefore, if a space with 300
lux requirement had a DA of 80% for one sensor, it means that the required lighting level would be
met by daylight alone, during 80% of the year for that sensor; and artificial lighting must be used for
the remaining percentage. Since this is a warehouse categorized under industrial buildings, the task is
three dimensional rather than two dimensional such as for office paper work (Illuminating
Engineering Society of North America[8]). Therefore, vertical illuminance is important to derive, as
well as the horizontal illuminance. The horizontal DA is measured on a horizontal set of nodes which
is placed at desk level of 0.9 meters. In a similar fashion, the vertical DA is measured on nodes placed
on both sides of six vertical planes (representing hypothetical shelves) distanced eight meters apart.
Two scenarios for placing the shelves has been assumed, one in which shelves are facing north-south
and the other where shelves are facing east-west (Figure 15). According to IESNA, uniform lighting is
used more often in industrial lighting which allows for repositioning of task locations without the need
to relocate lighting (Illuminating Engineering Society of North America[8]). Therefore, uniform
lighting was assessed visually.
Having the DA for all the sensors, there are two possible outcomes. One can calculate the arithmetic
average of DA based on the number of the sensors to get Mean Daylight Autonomy (MDA). The
shortcoming of MDA is that some nodes with a high DA may cover for other sensors with a low DA
and still provide a high mean value. Therefore, MDA does not describe whether a room is half dark
and half over lit, or fully day lit. The second possible outcome is to calculate the percentage of the
space with a DA larger than 50%. To calculate this, one may count the number of sensors that indicate
50% or more value, related to the total number of sensors. This means that if a sensor does not receive
the target threshold (300 lux) during at least half of the year, then it is not counted.
5.2. Daylighting results: Horizontal and vertical illuminance
Daylight autonomy (DA) was assessed on an unobstructed horizontal plane (Figure 15a). Looking at
DA in horizontal plane, it is noticeable that all cases present under lit areas in the corners. These dark
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
corners exist in greater regions in case 1 and 4 where apertures are closer to the center, and less
present in cases where the pattern of the aperture placement is roughly circular and the number of
apertures is numerous with smaller radius, such as 6 and 9. (Figure 16)
Figure 15a: Horizontal (top) and vertical (bottom) DA. Figure 15b: DA>50% comparison
Case 1 Case 2 Case 3
Case 4 (rhombic
pattern)
Case 5 (circular
pattern)
Case 6
Case 7 (original
shell)
Case 8 Case 9
Figure 16: Daylight autonomy representation on horizontal plane (case 1 to 9 from left to right)
DA was also assessed on two sides of a set of parallel shelves facing north-south, and a set of same
shelves facing east-west (Figure 15a). For vertical double sided planes, a mean value [(DA
north
+ DA
south
)/2] for both results was calculated. Considering this, case 5 provides a higher DA than the original
shell; case 3 and 6 have similar performance compared to the original shell, and cases 1, 2, 4 and 9
have lower performance (Figure 15b). From another point of view, there is not a considerable
difference in performance between placing the hypothetical shelving facing north-south or east-west,
except case 3 and 8. In this study, the vertical DA on north-south facing shelves has been chosen for
further analysis.
6. Multi-disciplinary performance assessment
In this phase, all cases are evaluated in terms of their structural and daylighting performance
simultaneously. Deflection has been chosen as the criteria in structural evaluation, while DA has been
chosen in the daylighting discipline. For this purpose, first a pass/fail criteria has been set for each
index, therefore some cases which do not meet the base criteria in any one of the discipline will be
eliminated. The pass-fail limit was set to having a DA of at least 60%, which eliminates case 1 and 4;
and having a deflection less than 1 cm, which again eliminates case 1. Second, for the remaining
cases, the performance values were normalized and plotted as single-attribute value curves (v
1
for
structural performance and v
2
for daylighting performance). And then averaged and ranked as a Multi
Attribute Value (V) function with equal weights. Finally the linear additive multi-attribute function
V= w
1
v
1
+ w
2
v
2
to present a graph of v
1
versus v
2
with lines of constant V.
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
Figure 17: ranked based on deflection Figure 18: ranked by DA>50%
Figure 19: ranked by averaging deflection and DA with equal weights Figure 20: Pareto front plot
The information that simple ranking represents is different from that which a Pareto front represents
along the lines of constant V. As an instance, by looking at the Multi-Attribute Value graph (Figure
19), case 2 has the lowest V; this is in line with the Pareto front plot (Figure 20) which shows case 2
close to the line of constant V=0.1 with a low overall value. Case 9 and case 3 seem intermediate
alternatives by looking at the ranking graph (Figure 19), however they appear to be two extreme
alternatives with value of zero on one attribute and value of one in the other by looking at the Pareto
front plot. It is worth noting that they are close to the line of constant V = 0.6 and still high performing
overall, but with a high trade off ratio. Overall, this comparative study demonstrates that simple
ranking or weighted ranking based on performance values is not a comprehensive method to compare
different topologies. Pareto fronts have the ability to represent data in a more comprehensive way.
7. Conclusion
This research studied a perforated shell designed by Heinz Isler, and assessed its structural and
daylighting performance simultaneously. The performance in each discipline was compared both by
ranking and by studying Pareto fronts. With equal weights, the original shell had a Value of 0.6 and
the best shell based on normalized mean performance got a Value slightly higher than 6. Then it
studied other possible topologies that could have been designed with higher performance in one
discipline or both. It was observed that placing the apertures farther from edge lines enhanced the
structural performance, but increased interior dark corners in regard of daylighting performance. It
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions
was also observed that shells with higher number of smaller apertures perform better than the shells
with fewer and larger apertures. Perhaps form-exploration within a larger population of design
alternatives could expand the observations and lead to unpredicted topologies with higher
performances.
For future studies, glare control is important to be assessed; in this comparative study though, it was
assumed that glare is present in all cases regardless of the aperture placement and that by providing a
certain method for glare control, the lighting level of all cases will decrease a certain amount. In terms
of structural performance, the designed apertures are induced to the shells and affect the force flow in
each shell. Although the force flow in different designed topologies is presented, it would be
interesting to see a resulting topology that reflects the flow of forces. This topology can vary based on
different support conditions, or by defining a target volume or by constraining the maximum
deflection. Some future research questions are possible, but not limited to: how much can we carve
out within a form found shell having assigned a specific material and still be within the structural
limits? How can we locate the apertures in spots that offer greater benefits in terms of daylighting and
structure? How can controlling the glare problem reduce the daylighting levels in a space?
Acknowledgement
I would like to thank Professor Harry Giles, Professor Peter von Buelow and Professor Mojtaba
Navvab for their guidance and support.
References
[1] Abel, John F., and John C. Chilton. “Heinz Isler - 50 Years of ‘New Shapes for Shells’: Preface.” Journal of
the International Association for Shell and Spatial Structures 52 (2011): 131–134. Print.
[2] Adriaenssens, Sigrid et al. Shell Structures for Architecture: Form Finding and Optimization. N.p., 2014.
Web. 10 Jan. 2015.
[3] Bertin, Trever. “Evaluating the Use of Particle-Spring Systems in the Conceptual Design of Grid Shell
Structures by.” MIT, 2013. Print.
[4] Chilton, John. Heinz Isler. London: Thomas Telford, 2000. Web.
[5] Chilton, John. “Heinz Isler : Shells for Two Churches.” Journal of the International Association for Shell and
Spatial Structures 52 (2011): 173–183. Print.
[6] Chilton, John. “Heinz Isler’s Infinite Spectrum Form-Finding.” Architectural Design. Vol. 80. N.p., 2010. 64–
71. Print.
[7] Emami, Niloufar, Anahita Khodadadi, and Peter von Buelow. “Design of Shading Screen Inspired by Persian
Geometric Patterns : An Integrated Structural and Daylighting Performance Evaluation.” IASS. Brasilia,
Brazil: International Associateion for Shell and Spatial Structures, 2014. Print.
[8] Illuminating Engineering Society of North America. The IESNA Lighting Handbook. 9th ed. NewYork: N.p.,
2000. Web. 8 Apr. 2014.
[9] Preisinger, Clemens. “Parametric Structural Modeling.” (2014): 1–104. Print.
[10] Ramm, Ekkehard. “Heinz Isler Shells – The Priority of Form.” Journal of the International Association for
Shell and Spatial Structures 52.3 (2011): 143–154. Print.
[11] Reinhart, Christoph. Daylighting Handbook. Volume I, Fundamentals,... N.p., 2014. Web. 16 Mar. 2015.
[12] Reinhart, F, Christoph., John. Mardaljevic, and Zack. Rogers. “Dynamic Daylight Performance Metrics for
Sustainable Building Design.” L E U K O S 3.01 (2006): 7–31. Web.
[13] Segal, Edward M. “Final Report : 2008 SOM Structural Engineering Travel Fellowship Table of Contents.”
May (2010): n. pag. Print.
... Perforated shell structures lie somewhere in between, providing an opportunity to study how the placement of their openings affects daylighting and energy performance while also considering their structural performance. Some case study-based research along those lines has been conducted on Heinz Isler's perforated COOP shell [35] and on Candela's perforated umbrella shell [36]. This paper builds on case study simulations of Heinz Isler's bubble shell as well as a preliminary investigation of perforated shell structures [37]. ...
Article
In this work, a computational interdisciplinary design approach is used to integrate assessment of the structural and environmental performance of perforated concrete shell structures. Design parameters that co-exist in these disciplines and relevant performance criteria are identified. Questions result: how do these design parameters affect performance? What is the tradeoff between performance in each? Computer-aided design tools are used for form generation and performance assessment. Statistical analyses are used to study the sensitivity of performance to each parameter. Finally, the perforation ratio is found to be the most significant parameter affecting both disciplines; a value ≤ 10% to 20% is recommended for shell structures when translucent glazing is installed without shades in a Boston-type climate.
... Past researchers have employed computational design strategies to optimize shell structures regarding their structural performance (Pugnale and Sassone 15]). Assessing the structural and daylighting performance of perforated shells simultaneously has also been conducted on a case study (Emami[4]). However, past research has not demonstrated a comparison between the performances of the two ends of the continuum of shell structures, nor the topologies generated in between. ...
Conference Paper
Full-text available
This research assesses structural and daylighting performance of perforated shell structures. By employing computational design tools and performance assessment methodologies, an array of generated topologies of perforated shell structures spanning the two extremes is studied. These generated topologies are coupled with structural and daylighting performance criteria to allow a performance-oriented exploration of the design space. ParaGen is used to automate the cycle of generation and evaluation. ParaGen is a method that uses a genetic algorithm (GA) to search for well performing geometric solutions to architectural engineering problems. By using ParaGen, the quantitative performance results are stored in a SQL database, accessible through an online website. The significance of this study is twofold: first, it studies a spectrum of generated forms of well-established structural typologies with perforations; and second, it couples structural and daylighting performance with geometry towards computational performance-based design and search of well performing alternatives. The results of the study contribute to the area of computational design, multi-objective design exploration as well as shell structures.
Article
Full-text available
Heinz Isler as the most famous contemporary shell designer has widely employed physical pre-modelling techniques for construction of many concrete shell structures. Through the physical approach to optimal form finding, Isler accomplished shell structures with robust performance. It would be interesting and beneficial to re-assess Isler’s shells, hence, this article attempts to study the structural performance of eight notable shells of Isler. Through reverse engineering and by the assistance of Rhino, MATLAB and Grasshopper, the precise geometry of Isler’s selected shells were modelled for the finite element analysis under their self-weight. The structural analysis was performed, with the parallel use of finite element software SAP2000 and Abaqus. The identical results of the two packages, further confirmed the accuracy of the analysis. The essential properties of various forms of the shells and their differences in behaviour were pinpointed and discussed within the calculations and the results were compared with the data of the genuine published references on Isler’s works. The internal forces, the amount of von Mises stresses, support reactions and the buckling loads of the shells are explored. The analyses revealed that, despite of their major membrane action, all the shells had negligible amount of bending moments, especially near the supports. However, in general, all the shells exhibited an appropriate performance under the applied actions. But, at the same time, they exhibited different buckling behaviour as a probable source of instability in them.
Article
Full-text available
The Swiss engineer Heinz Isler is considered to be the last of the great shell builders of the 20th century. He is set apart by the fact that his shells were constructed over a period spanning more than 50 years, from the first in 1954, to the last four constructed under licence by Willi Bösiger AG and completed in 2008/9. Although most were built in his native Switzerland, there are also examples in southern Germany, France, the United Kingdom and even in Saudi Arabia. He is best known for his innovative form-finding methods – by expansion, inflation and hanging of thin membranes – which resulted in shells such as the Wyss Garden Centre, Solothurn (1962); COOP Warehouse Wangen bei Olten (1960) and Deitingen Süd Service Station (1968), respectively, and led to him being acclaimed by David Billington as a structural artist. This special issue of the Journal of the IASS marks (slightly belatedly) the 50th anniversary of Heinz Isler’s presentation of his paper “New Shapes for Shells” [1] at the first Congress of the, then, International Association for Shell Structures, held in Madrid from the 16th to 20th September 1959, at which he first introduced his methods to the IASS.
Conference Paper
Full-text available
Shading screens have been used as daylight control systems, which also play a role as design elements of transparent facades. A façade’s configuration can be an explicit representation of its functions. There are multiple functions within a building and sometimes a dominant function imposes one configuration to the whole system; in this case, a replicated geometry that is used for the shading screen. In contrast, by grading the screen’s geometry, there is a response to each individual programmatic function according to the interior space of the building.We propose a functionally graded shading system that responses to different programmatic building functions. In this study, some geometric patterns, widely used in Persian historical ornamentations, have been chosen as the underlying geometry for shading screens. Geometric ornamental patterns are based on mathematical concepts, which are implemented on regular shapes of a certain arrangement. In general, such configurations have three geometric characteristics that have raised the tendency of employing the geometric patterns in ornaments of shading screens, facades, floor finishing and windows of both historical and contemporary architecture. First, these patterns can be fitted on different surfaces through geometrical concepts such as propagation, curtailment and scaling. Second, most of these types of patterns are self-similar configurations that are roughly similar to one part of themselves. This characteristic assists in making use of the properties of fractal geometry in parametric design of the patterns and further modification of an arrangement and density of a typical configuration. Third, some different patterns are generally based upon the same underlying rules and can be generated using almost the same geometrical processing techniques. Considering the self-similarity characteristic of the patterns, some shading screens are designed in this study with some Persian geometric patterns and then they are evaluated regarding their daylighting and structural performance. In the daylighting performance evaluation phase, we look at the year around daylighting levels and light distribution of a regular office space, while the specified shading screen is installed in front of a transparent full floor to ceiling glass facade. It is assumed that by using the shading screen, the required lighting levels are being maintained, while the heat gain is being reduced. In addition, the designed patterns play a role as an architectural feature of the building. In the structural performance evaluation phase, the shading screens which are considered as exterior self-supporting systems are analyzed under their self-weight and wind loads. The goal is to reach a minimum weight for the screens that are performing structurally. Ultimately, the presented shading screens are based on Persian geometric ornamental patterns and arranged according to the lighting and structural requirements. Results are indicating a multi-disciplinary approach in the design of the shading screens, which can be employed in creating similar prototypes with different climatic and loading requirements.
Article
Full-text available
The objective of this document is to promote the use of dynamic daylight performance measures for sustainable building design. The paper initially explores the shortcomings of conventional, static daylight performance metrics which concentrate on individual sky conditions, such as the common daylight factor. It then provides a review of previously suggested dynamic daylight performance metrics, discussing the capability of these metrics to lead to superior daylighting designs and their accessibility to non-simulation experts. Several example offices are examined to demonstrate the benefit of basing design decisions on dynamic performance metrics as opposed to the daylight factor. L?objectif visé par ce document est de promouvoir l'utilisation de mesures de la performance sous éclairage naturel dynamique aux fins de la conception de bâtiments durables. Comme point de départ, ce document explore les lacunes des paramètres de mesure conventionnels de la performance sous éclairage naturel statique, lesquels visaient les conditions de ciel prises individuellement, comme le coefficient d'éclairage diurne courant. On fournit ensuite un examen des paramètres de mesure de la performance sous éclairage naturel dynamique qui avaient été auparavant suggérés, et traite de la capacité de ces paramètres de mesure à générer des conceptions d'éclairage naturel supérieures et de leur accessibilité aux non-spécialistes en matière de simulation. Plusieurs types de bureaux sont examinés comme exemples de l'avantage découlant de décisions de conception qui soient fondées sur des mesures de performance dynamiques au lieu du coefficient d'éclairage diurne. RES
Article
This paper describes the concept, design and realisation of two churches where Heinz Isler (1926-2009) was responsible for the structural design and in one case also for the architectural concept. One example, the Heilig Geist Kirche (Holy Spirit Church) Lommiswil, near Solothurn, Switzerland (1967) includes, unusually for Isler, a "hypar" shell whilst the other, the "Steinkirche" of the Evangelische Kirchgemeinde in Cazis, Switzerland (1996), is derived directly from given natural forms.
Article
The Swiss engineer Heinz Isler (1926-2009) was convinced that "formfinding is the most important factor in shell design" [7]. The present contribution starts with an appreciation of his very first lecture at the founding colloquium of IASS in 1959; then it mentions the follow-up presentation 20 years later after an extremely active and successful period. His three main formfinding methods are discussed. Isler's aversion against computer models is briefly mentioned. Finally a short remark on the personality of Heinz Isler is given.
Book
Bringing together experts from research and practice, Shell Structures for Architecture: Form Finding and Optimization presents contemporary design methods for shell and gridshell structures, covering form-finding and structural optimization techniques. It introduces architecture and engineering practitioners and students to structural shells and provides computational techniques to develop complex curved structural surfaces, in the form of mathematics, computer algorithms, and design case studies.
Article
Heinz Isler (1926-2009), the Swiss designer renowned for his shell structures, was extraordinary for his innovative and exacting work. He directly produced physical models by hand in order to not only create design prototypes, but also to generate scaled-up measurements for construction. John Chilton describes how Isler successfully applied the principle of the inverted catenary arch, which was first pioneered by Robert Hooke in Sir Christopher Wren's St Paul's Cathedral in the 17th century, to thin membrane structures in three dimensions. Copyright © 2010 John Wiley & Sons, Ltd.
Evaluating the Use of Particle-Spring Systems in the Conceptual Design of Grid Shell Structures by
  • Trever Bertin
Bertin, Trever. "Evaluating the Use of Particle-Spring Systems in the Conceptual Design of Grid Shell Structures by." MIT, 2013. Print.
Parametric Structural Modeling
  • Clemens Preisinger
Preisinger, Clemens. "Parametric Structural Modeling." (2014): 1-104. Print.
Daylighting Handbook. Volume I, Fundamentals
  • Christoph Reinhart
Reinhart, Christoph. Daylighting Handbook. Volume I, Fundamentals,... N.p., 2014. Web. 16 Mar. 2015.