What is the impact of toplighting on daylight and energy performance in a hot climate?

PROJECT INFORMATION

Submitted by:  Elliot J Glassman, Associate

Firm Name: WSP Built Ecology

ASHRAE Climate Zone: 1A

Building/Space Type: Office

Who performed the simulation analysis?  External Energy Consultant

What tools were used for the simulation analysis? 

  • Grasshopper Honeybee

  • DIVA for Grasshopper

What tools did you use to create the graphic?

  • Excel

  • Adobe InDesign

What phase of the project was analysis conducted?  Design Development

What are the primary inputs of the analysis?  Material reflectance and thermal properties, various toplighting options, internal loads

What are the primary outputs of the analysis?  sDA, ASE, DGP, EUI

PROCESS

List the investigations questions that drove your analysis process.

Can we bring top lighting to a below grade office space while managing the climate’s high angle sun and strong solar radiation? What are the different possible design directions that would work for both daylight and energy performance? What is the impact of different alternatives on daylight, visual comfort, and energy? Are there trade-offs or synergies between daylight and energy performance?

How was simulation integrated into the overall design process?

This study was conducted on the project during early design development. The design team had wanted to bring daylight to a below grade office space through a lightwell and toplighting. The initial proposal for the toplighting was a large, centralized horizontal skylight. Our building performance team expressed concern that this would introduce a great deal of direct sun to the space because of the high sun angles in the climate, imposing unwanted solar loads and greatly increasing the risk of glare.

Simulation was brought into the process to illustrate the concerns as well as propose alternative toplighting approaches that could successfully daylight the space while reducing solar loads and glare. The goal was not to restrict the design but demonstrate a variety of alternatives that would work, giving the design team the flexibility to choose the approach that they felt was compatible with their design intent.

A reference case was created with no toplighting that would have only the lightwell to provide daylight but would perform well in terms of energy relative to the toplit alternatives. In addition to the horizontal toplight design, we proposed alternative design approaches such as a louvered shading system over the toplight to diffuse sunlight, a north facing monitor, and smaller monitors distributed across the space. Within each design direction, we look at three different parameters which varied by scheme such as the VLT of the glass for the unprotected skylight, louver angle for the shaded skylight, or the area of glass for the monitors.

How did you set up the simulation analysis and workflow?

We utilized the power of Grasshopper to perform the simulation analysis on all the alternatives. Each of the toplight schemes were preprogrammed into the script along with the parameterized variable we wanted to explore within each scheme.

The simulation was set up to run daylight, glare, and energy analysis for each alternative and output images of the results with unique file names referenced to the scheme. The simulations were connected so that the daylight analysis would inform the lighting schedule used in the energy simulation.

In traditional modelling approaches, you may need to coordinate inputs across multiple models to make sure you are getting an apples to apples comparison of alternatives. In our workflow, the geometry was parametric so we had one model where our inputs for material properties, thermal constructions, and internal loads were entered once and remained constant through each iteration. The only thing that varied was the lighting schedule which was automatically generated by the daylight performance.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

Looking at multiple alternatives can produce a great deal of outputs. We wanted to balance being able to understand the relative performance of each alternative at a glance with the ability to dig deeper into the metrics if desired.

We conceived of the result visualization as a matrix for each of the metrics. Each cell of the matrix contained the visual output of the simulation and the border of each cell was color coded based on the result to give a quick indication of the performance. Grasshopper helped us assign the colors based on gradients so that we could look at the performance of the alternatives both relative to each other and against fixed performance thresholds.

Based on the color coding, it was clear that while the unprotected toplight appeared to be well daylit, it was actually letting in too much direct sun, negatively impacting visual comfort and energy.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

The interesting conclusion illustrated by the study was that there were a number of schemes that performed better in terms of energy than the reference case without a toplight since the daylight reduced the lighting load and subsequently the cooling load as well.

This meant that there was not a trade off we needed to make between providing daylight or good energy performance. Rather, by sensibly managing the way that daylight entered the building, we could provide a well daylit and comfortable indoor environment for the occupants while reducing energy consumption.

There were a number of different alternatives that performed well giving us some flexibility to choose based on aesthetic or constructability criteria.

What is the impact of orientation and other design decisions on daylight quality?

PROJECT INFORMATION

Submitted by: Melissa Kelly and Elliot Glassman

Firm Name: WSP Built Ecology

Other contributors or acknowledgements: Cortney West developed the scope and structure of the study in an earlier iteration

ASHRAE Climate Zone: 4A

Building/Space Type: Education

Who performed the simulation analysis?  External Energy Consultant

What tools were used for the simulation analysis? 

  • Rhino - DIVA

What tools did you use to create the graphic?

  • Excel

  • Adobe Indesign

What phase of the project was analysis conducted?  Post Occupancy

What are the primary inputs of the analysis?  Orientation, window configuration, shading configuration, and interior finishes

What are the primary outputs of the analysis?  Spatial daylight autonomy (sDA), annual sunlight exposure (ASE)

PROCESS

List the investigations questions that drove your analysis process.

What are the key design decisions that may help or limit achievement of the LEED v4 Daylight credit? To what extent is daylight quality determined by orientation? Are there any specific considerations for rooms based on orientation?

How was simulation integrated into the overall design process?

The simulation process began with the single goal of coming as close as possible to meeting the LEED v4 Daylight credit criteria for as many configurations of a prototype classroom as possible. Parametric variables were determined through discussion with the design team in order to map out which aspects of the design were fixed and which were flexible. We then ran cases illustrating each potential design change, and presented initial results for discussion at a team meeting. With their feedback, we then looked at a few other opportunities of interest and re-rand the simulations. While the simulations themselves could not be interactive, the setup and results allowed us to have meaningful and detailed discussions.

How did you set up the simulation analysis and workflow?

The initial setup involved building out a shoebox model from floorplans and creating an Excel spreadsheet of each design change under consideration. This spreadsheet then became an input source for a Grasshopper for Rhino script which automated the creation and running of DIVA simulations. The script automatically exported ASE and sDA results into a spreadsheet and saved image files for each case. We then post-processed the results in Excel, and created an InDesign file with linked images which could automatically update. This saved a lot of time and effort at the beginning and end of the process.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

To answer the question of which design decisions create the greatest risk (or best opportunity) for achieving the LEED v4 credit, we plotted the range of results for each parameter by orientation. This helped to illustrate that there are a few variables which are critical to achieving high quality daylighting, while other areas could be more flexible without risking daylight access.

One challenge of presenting daylighting results, particularly for the LEED v4 criteria, is that it requires the balance of two opposing metrics rather than the optimization of a single metric. To look at the results on a granular level, we initially plotted ASE and sDA individually by orientation, but it didn't tell the whole picture on daylight quality. To solve this, we created compact graphs with both sDA and ASE results for each case, with a shaded background showing the LEED credit thresholds. Having one metric where lower is better and another where higher is better could be confusing, but the shaded background and color-coding helps to frame the data.

Lastly, we compiled spatial ASE and sDA results for each case. Superimposing the floorplan onto these results helps to clearly communicate how light levels and especially glare levels (through the proxy metric of ASE) could affect students, and provides a more tangible counterpart to the metric-driven visualizations.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

The extent to which orientation determined achievability was one key takeaway from the process. We learned that interior parameters such as surface reflectivity can only go so far, but strategic, climate-oriented shading strategies are almost mandatory to achieve LEED v4 thresholds for all but north-facing classrooms. This wasn’t a revelation, but the study helped to reinforce and illustrate exactly why that’s the case and that there are real benefits beyond LEED to doing so.

What is the impact of daylighting on creating dynamic learning environments?

PROJECT INFORMATION

Submitted by: Helena Zambrano

Firm Name: Overland Partners

Other contributors or acknowledgements: LAM Partners - Architectural lighting consulants

ASHRAE Climate Zone: 2A

Building/Space Type: Education

Who performed the simulation analysis?  Architect - Design Staff

What tools were used for the simulation analysis? 

  • Rhino DIVA

What tools did you use to create the graphic?

  • Adobe Illustrator

What phase of the project was analysis conducted?  Schematic Design, Design Development, Construction Documentation

What are the primary inputs of the analysis?  Daylight as the primary light source for the four levels of the atrium.

What are the primary outputs of the analysis?  Daylight autonomy, illuminance, Radiance visualizations, and false color imaging.

PROCESS

List the investigations questions that drove your analysis process.

 

Can we provide a daylit atrium in the center of the building to support occupant well-being without overheating the building? How do we provide appropriate daylight levels for various task areas on different levels from a single source?

How was simulation integrated into the overall design process?

 

Daylighting was identified as one of the most important design attributes to provide the necessary mental and visual stimulation and improve occupant productivity and comfort. However, increased cooling loads and glare potential were potential negative effects of a glazed atrium roof in the project’s hot and humid climate zone.

An additional challenge was to provide adequate light levels from a single source—skylights on the roof—to four receding stories, avoiding an overlit top floor and an underlit ground floor, with different task areas requiring different light levels throughout.

Exterior light scoops and interior baffles were designed in response to these challenges. The following variables determined their performance:

  • Distance of the light scoop from the top of the glass to admit appropriate light levels

  • Size of the light scoop to fully or partially cover the glass to avoid overheating

  • Angle of the light scoop to redirect sunlight

  • Rotation of the light scoop to catch light from different sky orientations, to add dynamism by reflecting sunlight of various color temperature and intensity.

The iterative process was fully integrated; the daylight simulations were performed in-house and reviewed by the lighting consultant as a form of peer review. In-house simulation was very beneficial for the project, the feedback was extremely fast, and the team could test as many iterations and variables as necessary. As well as reevaluate the design of the shading structure in terms of constructability constraints during the construction documents phase to maintain performance.

How did you set up the simulation analysis and workflow?

A Revit model was imported into Rhino. The modeler cleaned up the base geometry, converting meshes into Nurbs and window components to single surfaces. The roof, skylights, baffles and light scoops were created from scratch in Rhino and parametrized with Grasshopper.

Initially DIVA for Rhino and then DIVA for Grasshopper, with Radiance and Daysim as their calculation engines, were used to simulate daylight autonomy, an annual, climate-based metric that uses climatic data from *.epw files.

Daylight autonomy was used to see the overall performance of the design. Point-in-time illuminance was used to see the performance during specific times of the day and year, as well as to look for excess light. Radiance visualizations and false color imaging were studied to see the apparent illumination in the space and in different surfaces.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

The process was highly collaborative. Being integrated in the design team, the modeler didn’t provide a formal report of each iteration; rather the flow of information was frequent and informal using diagrams, radiance renderings, and false color daylight autonomy and illuminance results.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

The ability to quickly test design iterations very early in the design process was key in the success of the goals for the atrium. The simulation results informed the overall strategy and refined the geometry of each component of the atrium’s roof to achieve the goals of modulating the gradients of light in the baffles in a dynamic pattern and maintaining a general illumination levels of 300 lux.

The daylighting simulations also allowed the design team to reduce the amount of glazing from a fully glazed atrium roof by 92 percent, which had the double benefit of making the atrium design more economically feasible and achieve the goal of avoiding overheating.

What is the impact of automated blinds on daylight performance by facade orientation?

PROJECT INFORMATION

Submitted by:  Alen Mahic, Amir Nezamdoost, Kevin Van Den Wymelenberg

Firm Name: Energy Studies in Buildings Laboratory - University of Oregon

ASHRAE Climate Zone: 4B

Building/Space Type: Office

Who performed the simulation analysis?  University

What tools were used for the simulation analysis? 

  • OpenStudio + Eplus

  • Design Builder + Eplus

  • Radiance + custom scripting

What tools did you use to create the graphic?

  • R

  • Adobe Photoshop

What phase of the project was analysis conducted?  Design Development, Construction Documentation

What are the primary inputs of the analysis?  Daylighting model, energy model, blinds operation algorithms, and AMY/TMY weather data.

What are the primary outputs of the analysis?  Blinds operation schedules, blind movement and occlusion metrics, Hourly daylighting performance, spatial Daylight Autonomy and Annual Sunlight Exposure, Lighting schedule, hourly energy analysis, Annual energy end use, Non-energy benefits of automated blinds, Blind life cycle cost analysis

PROCESS

List the investigations questions that drove your analysis process.

What are the effects of manual and automated blinds on annual and point-in-time daylight performance, and energy end use? How can we leverage the simulation data to predict occupant productivity and absenteeism? What is the payback period of automated and manual blind systems?

How was simulation integrated into the overall design process?

This study was conducted on a mid-rise large open plan office during late design development to investigate the relative differences in operation patterns and resultant annual energy and daylighting performance between manual and automated system. Designers and building owners were interested in reducing the cost of shading systems by opting for manual blinds where automation might not be necessary.

Per facade blind performance analyses showed that northwest, southwest and southeast facades would benefit from automated systems by managing direct sunlight, and reducing interior glare. The lowest ASE value is recorded in northeast zone with total 5.75% (minimum=4.91% for January) while the highest ASE percentage is reported in southwest zone with total 41.72% (maximum=68.13% on December), following by southeast (26.59%) and northwest (23.53%) zones.

These findings confirm that northwest, southwest and southeast facades do benefit from automated controls while the northeast facade showed ASE value, which is low enough where manual blinds would be sufficient to control direct sunlight and minimize interior glare. However, the monthly ASE values reveal that there is still an excess of direct sunlight during summer months (12.68%), and there is an equivalent annual average number of blinds movements between the northeast (2.76) and northwest facades (2.72).

Moreover, we found that the payback period of automated blind system is lower (9.5 years) when compared to the fully manual alternative (16 years).

How did you set up the simulation analysis and workflow?

For annual climate-based daylight simulation, detailed digital interior models and relevant massing of surrounding buildings and site context were generated in SketchUp. When the model was completed and verified for accuracy against floor plans; it was exported to the RADIANCE daylight simulation engine. Rcontrib ambient calculations were used in RADIANCE to compute light source contributions. Materials and blinds were assigned to digital models based on the information from construction specifications using CIBSE surfaces and reflectance charts and bi-directional scattering and distribution functions (BSDF).

The Energy Management System (EMS) in EnergyPlus was used to create blind occlusion schedules. In order to calculate indoor horizontal illuminance for daylight sensing lighting control and creating the lighting/dimming fractional schedule, an overlaying procedure between EnergyPlus thermal zones and RADIANCE daylight zones was needed. We started with (1) EnergyPlus model including all the thermal zones, (2) a script-based analysis grid was generated based on four daylight zones in RADIANCE, (3) EnergyPlus thermal zones overlaid onto the RADIANCE analysis grid in order to isolate the illuminance grid points that fall inside the EnergyPlus thermal zone. These can be averaged, or a single point specified to generate the fractional lighting schedule, and (4) reconciling zone overlays in R-studio. This is a critical step to ensure accurate energy simulation of the interactive effects of daylight sensing electric lighting control.

Shade schedules were generated according to the combination of exterior illuminance data obtained from RADIANCE, plus horizontal profile angle from EnergyPlus.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

For blinds visualization we have annual results for 691 windows of the building, and needed a way to better understand and troubleshoot this data. R's plotting functions allowed us to visualize individual window surfaces and their blinds positions in 3D space. This method proved as effective and efficient throughout the simulation phase. Consequently, it also became an effective way to communicate the data and our findings to the clients by showing it in a similar manner as it is observed on actual building facades. Lessons learned:

  • Different ways of making simulation data spatial and easier to understand and communicate.

  • Better understand blinds as they change over time, and the algorithms used to simulate their behavior as part of the design process.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

The primary takeaway from this investigation confirmed our initial assumptions that the north-facing facades would require mitigation of direct sunlight, particularly in the summer and shoulder seasons where both sunrise and sunset hours introduce high potential of visual glare and heat gain discomfort. The daylighting analysis also provided data to inform the interior organization of partitions and work stations to minimize visual glare.

Additionally, this investigation allowed us to make relative performance comparisons among a variety of blinds operation algorithms to further understand the accuracy and reliability of specific simulation methods of complicated human behavior. This ultimately resulted in some of the more interesting visualizations showing the "haphazard" manual blinds algorithms contrasted against the uniform automated blinds algorithms. And on that note, we were also able to develop some interesting ways to process and visualize daylighting, energy, and blinds operation data that we can also utilize in future investigations.

What is the impact of daylight study on building layout and envelope system?

PROJECT INFORMATION

Submitted by: Lei Fu - Senior Associate

Firm Name: KGM Architectural Lighting

Other contributors or acknowledgements: Francis Mekhail / Roxanna Bailey

ASHRAE Climate Zone: 3A

Building/Space Type: Office

Who performed the simulation analysis?  Lighting Designer

What tools were used for the simulation analysis? 

  • Rhino - DIVA

  • Grasshopper DIVA

What tools did you use to create the graphic?

  • Adobe Illustrator

  • Adobe InDesign

  • Adobe Photoshop

What phase of the project was analysis conducted?  Schematic Design, Design Development, Construction Documentation

What are the primary inputs of the analysis?  Daylight Autonomy and Maximum Daylight Autonomy

What are the primary outputs of the analysis?  Percentage of floor area that meets target illuminance with original design, compared with percentage of floor area that meets target illuminance with improved design

PROCESS

List the investigations questions that drove your analysis process.

Our project is located in a high density neighborhood and the main entrance is from the street to the northwest. The main facade is solid with random cuts, while the back facade, facing southeast, is floor to ceiling glazing. As a result, our primary challenge were: How can we reduce the glare from the southeast facade without cutting too much useful daylight? How can we utilize daylighting simulations to optimize the balance between quantity and quality of natural light? Which facade strategies are most effective in decreasing glare?

How was simulation integrated into the overall design process?

We performed radiance HDR rendered videos, daylight autonomy (DA) studies and maximum daylight autonomy (DA.max) studies. The DA and DA.max studies are the primary evaluation matrix we used to determine the lighting quantity and quality of the space. We determined that spaces with high DA.max values are over lighted and are not appropriate to be used as workspace, while spaces with low DA values are under lighted and require electric lighting to be used as workspace. We presented our findings to the architect, who then reconceptualized some aspects the southeast facade based on our daylighting analysis.

How did you set up the simulation analysis and workflow?

The simulation was set up through a series of steps starting with Revit. The Revit model, provided by the client, was exported to Rhino for analysis and simulation. Revit layers were cleaned up in Rhino making the application of materials through DIVA a more seamless process. The specific epw (weather data) was downloaded and applied to the "application" portion of the simulation. Nodes were then placed (distance of sampling node: .76, distance between calculation nodes: .45) on each floor. Materials were applied according to the architects direction and, finally, metrics were defined. Prior to the simulation, grasshopper was used to investigate the sun path over the site. We initiated a series of shadow studies and developed a facade strategy for testing. Metrics: Daylight Grid-Based > Climate-Based > Metric: Daylight Autonomy, Occupancy Schedule: 8 to 6 (office space), Target Illuminance: 300 Lux. We tested a series of facade strategies until we established target metrics that were satisfied.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

Output focused on a series of graphic representations paired with diagrams. We first defined DA and DA(max). We communicated the significance and impact on the exterior facade strategy and interior programming. Through the use of diagrammatic sections we were able to explain our approach and walked the design team through the output of each facade strategy. We relied heavy on the Adobe suite (Illustrator, Photoshop and InDesign) to create a clear narrative for presentation.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

The simulation allowed us to determine which areas of the building should be reorganized programmatically based on daylight. It also enabled us to quantify the cost/benefit of proposed facade strategies on the quality of natural light in the space. With the information from our simulation, the architect reprogrammed the space to include more amenities on the southeast side of the building, where glare is most prominent. The architect is exploring facade strategies to help mitigate the glare we observed in our studies.

What is the impact of skylight geometry on museum daylighting?

PROJECT INFORMATION

Submitted by: Morgan Maiolie

Firm Name: Waggoner and Ball

ASHRAE Climate Zone: 2A

Building/Space Type: Education, Museum

Who performed the simulation analysis?  Architect - Design Staff

What tools were used for the simulation analysis? 

  • Ecotect

What tools did you use to create the graphic?

  • Excel

  • Adobe Illustrator

  • Adobe InDesign

  • Adobe Photoshop

What phase of the project was analysis conducted?  Conceptual Design, Schematic Design

What are the primary inputs of the analysis?  Material surface reflectance, time of day, date, location

What are the primary outputs of the analysis?  Daylight levels in footcandles

PROCESS

List the investigations questions that drove your analysis process.

Can we daylight a museum with sensitive pen & ink exhibits? Can we daylight a first floor from a skylit third floor? How does our solar resource on site drive programming decisions? Can we use daylight to add dramatic and / or dynamic lighting to the museum experience? Can we integrate views to the west and maintain control of daylight levels? What is the experiential effect of providing surface brightness with daylight in the space?

How was simulation integrated into the overall design process?

We used an iterative process of daylight analysis to test myriad design questions.
One question was whether we could get useful daylight on three floors of the project. We were renovating a 3-story brick building where windows were feasible only on one side and we were curious about whether we could wash the windowless back wall with daylight from the skylit floors above. Daylight modelling showed this required a larger floor area loss than the project could support, so we focused our design on bright, daylit spaces on the top floors. The two lower floors were improved by an atrium after a separate series of daylighting analyses. Another early design question was whether we could hit an 18 fc / 4000 fc / yr target for sensitive pen and ink works. Analysis showed we could, but wouldn't like the aesthetic and financial result, which changed our space programming for those works.                                            
Moving through schematic design, we used daylight analysis to determine the shape and orientation of skylight monitors, as well as the window coverings and material surface brightness of top floor museum spaces. We decided on different daylighting strategies for the two top floor rooms based on their differing geometries / programs with a light scoop and sawtooth roof represented.                                                                                                                  
The daylight modeling process was integrated with the design process in that it was carried out by a designer on the team and presented with each formal design idea. We presented analysis to the client when we discussed design options.

How did you set up the simulation analysis and workflow?

We created a model from scratch in Ecotect and used Radiance to analyze it. We then exported graphic results and added information for presentation with Illustrator and Indesign.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

We found it wasn';t intuitive to read the typical red-to-blue outputs from Ecotect and vizualized daylight intensity with a white-to-yellow-to gray palette instead. This helped our client and design team better understand the experiential difference between bright and dim areas. We also relied heavily on perspective outputs paired with daylight analysis grids to provide an experiential understanding of the different design decisions (while always mentioning that the way these show on screen isn't a direct relationship with reality). Finally, we took daylight readings at museums the design team and client had visited and compared these with our results to give them a real-world baseline.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

We really didn't know how we could use daylight within the space - perhaps because of our relative newness to daylight design - so analyzing questions like whether an opening between a daylit space and the floor below could wash a wall sufficiently was really useful. Quatifying our daylight limits for museum exhibits was invaluable because the museum had only used electric lighting in their exhibits before. Attaching numbers to the different works (pen&ink versus oil paintings versus sculpture) helped us all understand programming for these works and their ability to connect with daylight.

What is the impact of facade geometry on space daylight distribution and heating/ cooling load?

PROJECT INFORMATION

Submitted by: Xiaofei Shen

Firm Name: AECOM

Other contributors or acknowledgements: Aman Singhvi, Jason Vollen

ASHRAE Climate Zone: 4A

Building/Space Type: Office

Who performed the simulation analysis?  Architect - Internal Sustainability Personnel

What tools were used for the simulation analysis? 

  • Grasshopper Ladybug

  • Grasshopper Honeybee

  • Rhino-DIVA

What tools did you use to create the graphic?

  • Adobe Illustrator

What phase of the project was analysis conducted?  Schematic Design

What are the primary inputs of the analysis?  Window Geometry, Shading Geometry

What are the primary outputs of the analysis?  Daylight Distribution, Solar Radiation, Heating and Cooling Load

PROCESS

List the investigations questions that drove your analysis process.

What is optimum Window/ Wall Ratio for Daylighting? What is the reduction in annual daylight based on external shading systems? How fenestration and shading design holistically affect the building performance? How to obtain the most optimal facade trade-offs for balanced daylight and energy?

How was simulation integrated into the overall design process?

The study was conducted when the design team reached out for quantifiable design guidelines to inform the facade conceptualization process during the schematic design phase. The aim was to achieve best performance of the space with respect to daylight and energy.

The number of design iterations are massive due to multiple variables involved. More than 200 solutions were generated with various window widths, height and location as well as shading types (horizontal or vertical) and shading depth. Moreover, the performance related objectives, including minimizing glare area, minimizing under-daylit area, minimizing solar irradiation on the glazing, and minimizing heating/ cooling loads are all conflicting to each other due to their mutually exclusive dependencies.

An automatic Multi-Objective Optimization was utilized for rapid exploration of options and optimization at early stage of design for quick decision making.

How did you set up the simulation analysis and workflow?

A typical south-faced bay space in an office building located in DC with a total floor area of 2,500 sf was modeled using Rhinoceros 3D modeling software and Grasshopper plugin. The pre-determined variables, such as windows and shadings, were parametrically constructed. Parameters controlling window dimensions, type of shading and shading depth were setup to vary within predetermined limits set based on design constraints.

The initial iteration was connected to DIVA daylight simulation software. Sensors were placed at a 12-inch square grid. Two daylight simulations were carried out to get Illuminance levels at 9 am and at 3 pm at the design day (Sep 21st). Colored Illuminance maps and numeric values of lux levels on each floor were outputted. The iteration was also connected to DIVA thermal analysis software. Room boundaries were set up as typical climate zone 4A materials. Heating and cooling loads related to the facade elements were simulated and recorded. 

Colibri for Grasshopper was selected to automatically trigger the simulations and save both inputs and outputs for this experiment. The recorded values were connected to a customized script in Grasshopper for further result sorting and finding.

How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?

The building performance simulation is traditionally time-consuming and the report is usually hard to understand with redundant information showed. To visualize such a complex process of multi-objective optimization, variable clusters as well as energy data was an equally challenging task as well to represent in one or more graphics. The resulting visual simplified the raw data in Adobe Illustrator with minimized information presented but in a more graphic way, such as variables with 3D geometries and objectives with 3D bar charts. This visualization made it considerably easier for clients and designers to understand the results as compared to just stats and numbers.

Additionally, multiple recommendations were provided with their performance values listed through a filtering and sorting process. Preferred options were selected based on constraints and bios after discussion with the designers. A more detailed 3D section was also provided for a closer look of the potential development of the selected solution.

Most importantly, what did you learn from the investigation?  How did simulation and its outputs influence the design of the project?

This evolutionary trade-off analysis tool is the best way to provide performance feedback for design decision support in a quicker and comparatively more accurate way, especially at the beginning of the conceptual stage when there are multiple iterations generated from the parametric models. By managing the Window/Wall Ratio, shading dimensions and wall construction types, the rich-data multi-task methodology were utilized for better performance on daylighting, thermal comfort, energy load and cost, making the optimized facade result more reliable.

The user-friendly graphics made the design decision making more efficient. Multiple stakeholders could get involved into the discussion when the complex data was well visualized.

Findings from this analysis: the optimized modular design of the facade with Window/ Wall Ratio 50% and 6’ horizontal shadings properly mediates summer solar exposure but still ensures daylight penetrating reducing both lighting load and heating/cooling load. Glare control and reduced contrast ratio improves the visual comfort. The facade is further integrated with building systems. For example, outdoor air is pre-cooled and pre-warmed through the localized DOAS units and Heat Exchangers embedded in the modules.