What is the impact of the hydronic configuration on the performance of a hybrid heat production system?

PROJECT INFORMATION

Graphic Name: What is the impact of the hydronic configuration on the performance of a hybrid heat production system?

Submitted by: Freek Van Riet

Firm Name: University of Antwerp

Other contributors or acknowledgements (optional)

What tools did you use to create the graphic?

  • Matlab

What kind of graphic is this?

Primary Inputs: The inputs required to create this extended type of Load Duration Curve (LDC) are heat loads throughout a year of an hydronic heat production system with its different components. The loads can originate from both measurements and simulations. In this simulation-based case the yearly loads of three components were taken as input: a cogeneration device (CHP), a storage tank and an auxiliary boiler.

Primary Outputs:

1) the total heat load of the building, which is the sum of the loads of the three components (CHP, tank and boiler), sorted in descending order = conventional LDC.

2) the heat loads of the three components separately, corresponding to the descending order of the total heat load.

3) the load duration curve of the CHP itself.

4) areas indicate total heat produced (or charged/discharged) by each component.

5) upper outputs are given for both the ‘expected’ behavior (i.e. what one can expect of a properly working installation) and the real ones for different design concepts.

6) in the graphs, three levels of load can be distinguished: peak, base and part load.

GRAPHIC INFORMATION

What are we looking at?

Most people who design heating systems of buildings are familiar with a ‘load duration curve’: it is a common graphical representation of the heat demand of a building. But did you know that in the same graphic much more information can be shown, to indicate the performance of the heat production system in the blink of an eye? Especially when different heat production components are combined into a ‘hybrid heat production system’, this is beneficial. In the four schemes (upper part of figure), a hybrid production system with its possible designs of hydronic configurations are shown as an example. The first, ‘expected’, explains the heat flows in general and in the three following ones particular design options for the hydronic configuration are shown. In the lower part of the figure, the load duration curves including the loads of the three components are shown, corresponding to the upper schemes. This is the part of the figure that is automatically generated and is proposed here as ‘the fingerprint of a hybrid heat production system’. The x-axis shows the duration in hours and the y-axis the loads, relative to the maximum load of the building. In the reference results (‘expected’) three levels of heat load are distinguished, according to how an in-theory-proper-working installation would perform: -the base heat load is provided by the CHP -the heat demand higher than the CHPs load (dotted line=load duration curve of CHP itself) is covered by an auxiliary boiler - at heat demands of the building lower than the heat load of the CHP, two possible situations can occur. First, the CHP is on and the heat that is not consumed by the building is stored in the tank. Second, the CHP is off and the stored heat is discharged from the tank to heat the building. If the behaviour of the three hydronic configurations are compared with this expected behaviour, some conclusions can be drawn, such as: - for the ‘parallel basic’-configuration, a part of the base load is covered by the boiler instead of the CHP: the figure clearly shows that the hydronics fail, causing the CHP to operate less hours than expected. This configuration should be excluded from the possible designs. -for the two other hydronic configurations the tank can cover partially the base load, for which the CHP can produce more heat than expected.

How did you make the graphic?

An algorithms post-processes the data of full year simulations. The total heat demand is calculated as the some of the loads of the three components, and that total demand is sorted in descending order. Then, the different loads are plotted as stacked areas.

What specific investigation questions led to the production of this graphic?

- How to evaluate hybrid heat production systems?

- How should a CHP device be integrated in the heat production system of a building?

- What is the effect of buffer tank volume on CHP operation?

- How to ensure the priority of operation of the CHP above that of an auxiliary heater?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

This type of LDC shows how a particular hydronic configuration influences the operation of a heat production system. Because it is compared with an in-theory-proper-working installation, it enables to detect faults and benefits of the configuration. By creating these figures I learned about the disadvantages of each configuration, without necessarily looking into the detailed dynamic simulation data.

What was successful and/or unique about the graphic in how it communicates information?

Load Duration Curves (LDC) are widely used to communicate about the heat demand of a building. In this extended type of LDC, the load of each single heat-producing component is added, thereby providing much more insight. Indeed, not only the heat demand of the building can be seen, also the way this heat is delivered or stored is shown. Because LDCs as such are familiar, these graphics are easy accessible for both researchers and engineers working in the field of heating.

What would you have done differently with the graphic if you had more time/fee?

In the end, designers or owners are interested in fuel consumption and the resulting costs, and not necessarily in heat as such. The relation between both are, of course, the efficiencies of all the components, which are not shown in these figures. Hence, extra graphics showing the efficiencies or fuel consumption would be interesting, especially to incorporate the effect of temperature levels in the installation.

What is the impact of building massing on annual energy end uses and heat gains & losses?

PROJECT INFORMATION

Graphic Name: What is the impact of building massing on annual energy end uses and heat gains & losses? 

Submitted by: Carl Sterner

Firm Name: Sefaira

Other contributors or acknowledgements (optional) 

What tools did you use to create the graphic?

  • Sefaira

  • Excel

  • Adobe Photoshop

  • SketchUp

What kind of graphic is this? Bar chart

Primary Inputs: Building geometry, and location- and building-use-specific assumptions related to building envelope (utilizing prescriptive values from ASHRAE Std. 90.1) and mechanical systems (utilizing system templates in Sefaira).

Primary Outputs:

  1. Annual energy end use breakdown, normalized per square foot (kWh/ft2/yr)

  2. Sources of heat gain and heat loss, shown as a % of total heat gains and total heat losses (rather than in BTUs)

GRAPHIC INFORMATION

What are we looking at?

The graphic is designed to assess the relative performance of the three massing options, and to provide enough information to help the design team come up with possible strategies for improvement. The bars on the left show energy end uses, normalized per square foot to facilitate quick comparison. For example, one can quickly see that the third option has the highest heating energy use per square foot among the three designs.

On the right are two bars showing annual sources of Heat Gains and Heat Losses, which shed additional light on the heating and cooling end uses. For example, the third option’s “Heat Losses” bar reveals much higher conduction losses through walls than the other options. Because the envelope properties are held constant in this analysis, this difference is due to the greater wall area (higher surface-area-to-volume ratio) of this design compared to the others.

How did you make the graphic?

  1. Created massings in SketchUp and performed energy analysis using Sefaira.

  2. Exported perspective images from SketchUp (without glazing, in order to maintain focus on massing and not facade design).

  3. Logged results in Excel, including total energy end uses and annual heat gains and heat losses.

  4. Normalized the results based on the floor area of each massing.

  5. Created graphs in Excel.

  6. Combined the graphs and model images in Photoshop, added a gray background, and adjusted text and alignment.

What specific investigation questions led to the production of this graphic?

  • Which design option performs best, and why?

  • What is driving the energy use of each design option?

  • What is driving the heating and cooling loads of each design option?

  • What strategies could improve the performance of each design option?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

This graphic is a very early-stage analysis. The most important learning is where to focus moving forward -- in this case, on plug loads, glazing U-factor, and wall R-value. The graphic is designed to generate questions and hypotheses -- for instance, “How high would the wall R-value have to be in Option 3 for it to perform as well as Option 2?” These questions are the logical next steps for subsequent analyses..

What was successful and/or unique about the graphic in how it communicates information?

Analysis alone rarely determine a design’s direction. Instead, such analyses are successful if they provide insight on why a design performs the way it does, and what can be done to improve that performance. This graphic sought to provide just the right information to inform decision-making, presented as clearly as possible. We avoided pie charts, as these are more difficult to compare. We used common-sense colors (e.g. red for heating, blue for cooling), and kept them consistent across all charts (e.g. lighting and equipment are the same in both the “end uses” and “heat gains” charts. The result is legible, aesthetically compelling, and above all useful for quickly assessing a design, understanding the factors affecting performance, and generating clear, actionable hypotheses that can be tested with further analysis.

What would you have done differently with the graphic if you had more time/fee?

Experimented more with the color palette, and with alternate compositions for heat gains and losses.

.

What is the impact of window-to-wall ratio on Energy Use Intensity (kBTU/ft2/yr)?

PROJECT INFORMATION

Graphic Name: What is the impact of window-to-wall ratio on Energy Use Intensity (kBTU/ft2/yr)?

Submitted by: Carl Sterner

Firm Name: Sefaira

Other contributors or acknowledgements (optional) 

What tools did you use to create the graphic?

  • Sefaira

  • Adobe Illustrator

  • SketchUp

What kind of graphic is this? Scatter Plot

Primary Inputs: Building geometry, location, and location- and building-use-specific assumptions related to building envelope (utilizing prescriptive values from ASHRAE Std. 90.1) and mechanical systems (utilizing system templates in Sefaira).

Primary Outputs: Energy Use Intensity in kBTU/sf/yr

GRAPHIC INFORMATION

What are we looking at?

This graphic shows how the optimum southern window-to-wall ratio (WWR) for a building changes as other elements of the design change--illustrating that WWR cannot be optimized in isolation, or sized based upon generic rules of thumb. Each row illustrates an optimum south WWR given a specific envelope and mechanical system design, and shows whether the calculated optimum aligns with the range described by a prescriptive rule of thumb (illustrated as a shaded bar on the sensitivity graph). The optimum WWR changes dramatically based upon the properties of the envelope and mechanical systems, ranging from 0% to 65%.

How did you make the graphic?

  1. Modeled building geometry in SketchUp

  2. Uploaded the model to the Sefaira web application

  3. Set up the various simulations (envelope, internal conditions, and HVAC parameters)

  4. For each scenario, a ran a Response Curve to find the optimum South Window-to-Wall ratio looking at the impact on EUI. Exported the resulting graphic.

  5. Used SketchUp to illustrate the optimum Window-to-Wall ratio for each scenario, and exported a high-resolution image of each.

  6. Combined the model image and Response Curve image in Illustrator, adding linework and descriptive text.

What specific investigation questions led to the production of this graphic?

  • What is the optimum south window-to-wall ratio for this design?

  • How do changes to the building envelope and mechanical systems affect the optimum?

  • How accurate are rules of thumb regarding WWR?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

The major takeaway was that WWR cannot be optimized in isolation from the rest of the design. While we knew this in theory, the magnitude of the variation was still startling. It was particularly interesting to see the difference between the option that used “Passive House windows” (U-0.15) and the full “Passive House” option that included high insulation, high air-tightness, and high-efficiency mechanical systems with energy recovery ventilation. The full Passive House design reduced heating loads to such an extent that the design did not benefit from as much solar gain, resulting in a small optimum WWR. This exercise reinforced the need for continual stimulation throughout the process to ensure the design stays on track as the design evolves.

What was successful and/or unique about the graphic in how it communicates information?

The image graphically communicates analysis results, which makes the variation in window-to-wall ratio immediately apparent. It illustrates how changes to inputs create wide variation in outputs, and that rules of thumb cannot be relied upon for optimum results.

What would you have done differently with the graphic if you had more time/fee?

The inputs for each scenario are summarized on the right-hand side of the graphic. I would have liked to better clarify inputs vs. outputs, and perhaps find a way to represent the inputs graphically through iconography or sparkline graphics, so viewers can better understand what was changing between the various scenarios.

What is the impact of glazing on energy and daylighting?

PROJECT INFORMATION

Graphic Name: What's the impact of glazing specifications on peak loads and energy consumption? 

Submitted by: RJ Hartman

Firm Name: Clark Nexsen

Other contributors or acknowledgements (optional) Ryan Johnson, Architect. Ryan was the primary contributor/modeler of the graphic

What tools did you use to create the graphic?

  • Grasshopper Diva

  • Grasshopper Human UI

What kind of graphic is this? 

Primary Inputs: Room massing, interior furniture, exterior shading, window geometry, and SHGC.

Primary Outputs: sDA, ASE, Lighting Energy, Peak Energy, and Chilled Beam Length

GRAPHIC INFORMATION

What are we looking at?

The graphic shows a dashboard like interface that shows a plan of the space with the sDA/ASE. Below the plan is an elevation of the exterior wall showing the configuration of the glazing and the amount of solar energy hitting the glass. Alongside the plan/elevation, there are several large metrics highlighting the impact of the glazing on the daylight and energy. There are indicators to the right of each metric that shows what is established as an industry standard for acceptable performance, or project specific goals for what is considered good/acceptable/poor performance.

How did you make the graphic?

Elements were first modeled as individual spaces/rooms in Rhino then brought into Grasshopper and analyzed using Diva energy and daylight analysis. Since the Diva output data is saved locally to the users' workstation, the Grasshopper script was set up to be able to run several analyses at once. After the analysis, the Diva data could be read off the workstation by Diva and processed with Grasshopper into the desired metrics. The graphic interface of Human UI, a plugin for Grasshopper, allows users to choose 2 specific options which are read by Grasshopper, processed, and graphically displayed via Human UI..

What specific investigation questions led to the production of this graphic?

  • How do we balance energy and daylighting via manipulating the amount of glass and shading?

  • How much glass is too much glass?

  • What "Window to Wall Ratio" do we want?

  • Does this external shade reduce glare?

  • What glass type should we choose?

  • What is the impact of frit on the glazing performance?

  • What is good daylighting?

  • What is bad glare?

  • What is good lighting energy?

  • What is our target energy use for the space?

  • How many chilled beams are required in this space?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

The architects were trying to design the façade of the building and wanted to know how much glass they could have in a space that provides a well daylit space while managing glare and solar gain. Using traditional outputs from daylighting and energy model tools, the team was having difficulty quickly determining what it meant to have good daylighting and what it meant to have good energy use. This is led us asking “How do we present our data/findings so that different team members (architects, electrical engineers, and mechanical engineers) can understand this data and can make design decisions from it as well.” The way that the Grasshopper + Human UI script worked, it saved an image to the server of each graphic automatically once it was displayed. These graphics were used in meetings to pull up multiple images of options side by side on the same screen like cards. This allowed for a quick literal side by side comparison of options. Ultimately these goals and being able to compare options helped decisions be made about shading, glazing configuration, and glass types for the final design of the building.

What was successful and/or unique about the graphic in how it communicates information?

Our graphic was successful in that we were able to convey a significant amount of information to our team, while maintaining a clean and easy to follow organization. It eliminated a lot of confusion and questions by the team and sped up the decision making process of the designers.

What would you have done differently with the graphic if you had more time/fee?

The plan with our graphic is to continuously update based on the needs of whatever project we are using it on. With this being said, we would have liked to have a clearer naming directory for the various options we were analyzing. In our submitted graphic, the description of the option is all the information we have about the option, which resulted in a cryptic naming convention not easily understood by the team. We would have preferred to include a description of the option within the graphic but could not come up with a good solution to this problem since all the data is being read from Diva. We potentially could try to read and write data to another place, (eg: excel) to be able to use that as a place to store data that is not stored in the Diva results. Examples of data that could be stored are a written description of the option, information about shading, orientation data, or anything that the team/project deems useful for the project. We also think that the bar chart at the bottom could be cleaner, and more useful with labeling/colors/etc but used the stock Human UI chart component.

What is the impact of climate and ubication on the energy balance of an initial model versus an optimized model of customer service agencies?

PROJECT INFORMATION

Graphic Name: What is the impact of climate and ubication on the energy balance of an initial model versus an optimized model of customer service agencies?

Submitted by: Luis Godoy

Firm Name: NA

Other contributors or acknowledgements (optional) Sebastian Romero, Andrea Lobato-Cordero, Geovanna Villacreses and Cristian Espinosa. Instituto Nacional de Eficiencia Energética y Energías Renovables.

What tools did you use to create the graphic?

  • Design Builder

  • Climate Consultant

  • Excel Adobe

  • Photoshop

  • Meteonorm

  • ArcGis

  • Power Point

What kind of graphic is this? 

Primary Inputs: Average per hour of meteorological information, geographical ubication, initial and optimized architectural models.

Primary Outputs: Overall and hourly energy balance (gains-losses) for an extreme climate conditions day.

GRAPHIC INFORMATION

What are we looking at?

This graphic shows how is the energy balance (loses and gains) established, for buildings that will work as customer service agencies, to be built in affected areas by the 2016 earthquake in Manabi, Ecuador. Firstly, a weather condition study was carried out by using data of a typical meteorological year of the cities were the building will be located (Pedernales, Manta, Flavio Alfaro, Portoviejo and Calceta). From this information, the most probable hottest day of all cities was selected (April 11th). An initial model (IM), offered by the contractor, was compared with an optimized model (OM) that uses passive strategies for design. The energy performance, from the simulation, of the building with its IM and OM was analyzed, always considering the building functioning requirement such as: space, distribution, usage, etc. The set strategies must be replicable and show energy savings in all the cities. The total energy balance showed in the donut charts indicates the characteristic thermal gains and loses throughout the building materials according to their type. Even though Manabi corresponds to the ASHRAE climatic zone known as “humid very hot” each one of the agencies shows a particular behavior in which the inner temperature increased at afternoon hours due to the typical weather conditions. As first result, the solar gains on windows and roofs are representative, this is why the usage of efficient windows and reflective paint on roof and walls, is proposed. The second result of the bar plot indicate he hourly behavior of loses-gains. The internal gains on external windows on afternoon hours can be relieved by accurately orientated vegetation. Besides, it is possible to note that, for loses, these are grater on ground floor, which is a cooling advantage. Finally, it is possible to conclude that in most cases, energy savings are found after passive strategies of design were implemented, on all four cities. Portoviejo city was a special case because heat loses were diminished.

How did you make the graphic?

Firstly, with the coordinates of the five cities were the agencies will be place, through Meteonorm, the .epw (energy plus weather) files were generated, for the typical meteorological years (TPM) that will be used for weather study and energy simulation. With this data, the agencies were ubicated on ArcGis to confirm whether the cities are located on the same climate zone. Simultaneously, the weather study was carried out with Climate Consultant for each city. The day that showed the greater gains on the building was selected. The input data entered on Design Builder included: climate conditions, agencies characteristics IM and OM proposed. With the energy simulation, the hourly results of gains and loses were obtained, in .csv format. The analysis and calculations obtained were plotted as donut and bar charts by using Excel. The showed graphs were edited on Adobe Photoshop for a better display. After that, the final display was diagramed with the analisis content.

What specific investigation questions led to the production of this graphic?

- Which are the main energy gain and loss, as product of climate conditions, for a building model of customer service that will be built on affected areas by the earthquake in Ecuador? - How great is the energy saving that can be obtained by optimizing the initial model with the implementation of passive strategies for design, on the hottest day of the year? - How is the energy balance established and what aspects must be considered to optmize the IM?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

In Ecuador and other countries, it is proposed the construction of buildings with similar facilities and services to satisfy the customer demands. Besides, the buildings must be replicable in different locations. These buildings can be constituted by companies such as: banks, restaurant franchises, payment agencies, customer services, etc. Nevertheless, the influence of the energy and thermal behavior of the building is not considered. This is why it is important that prior the construction, efficient passive strategies on the design are analyzed and proposed, in order to obtain energy savings. The lessons learned were: though a climatic zone can be normalized (for instance, humid very hot), the characteristic weather of the area can affect differently each zone even if they are near to each other. The energy balance is a simple and useful analysis that allows to detect and prioritize sensitive zones of the building to energy gain or lose.

What was successful and/or unique about the graphic in how it communicates information?

From the geographical location, the climate conditions and architectural characteristics of the building, it is possible to obtain the hourly and total energy balance, to detect energy saving opportunities in each city. In this case, it was observed that windows and roof are sensitive zones for heat gain, for this reason, their intervention must be prioritized to obtain greater energy savings. By using window and wall protection and specific paints (that do not increase considerably the final budget), effective energy savings can be obtained in all the agencies in the studied cities.

What would you have done differently with the graphic if you had more time/fee?

Firstly, we will try to find a correlation between the geographical location of every building and the energy consumption, considering the meters above sea level in which the building is located. Also, we would like to try or to combine other passive strategies for instance: external ventilation and/or ventilation on floor over terrain. Furthermore, we will monitor the energy consumptions of the building before, during and after its construction, to corroborate implemented the design strategies.

What percent of the day does the public space get exposed to direct sunlight?

PROJECT INFORMATION

Graphic Name: What percent of the day does the public space get exposed to direct sunlight?

Submitted by: Elizabeth de Regt

Firm Name: ZGF Architects LLP

Other contributors or acknowledgements (optional)

What tools did you use to create the graphic?

  • Excel

  • Grasshopper Diva

What kind of graphic is this?

Primary Inputs: Building massing, topography, weather data

Primary Outputs: Annual Shadow/Daylight data

GRAPHIC INFORMATION

What are we looking at?

This graphic displays a heat map of direct daylight, similar to overlaying a year's worth of shadow studies on one image. It is a simple method of showing which areas get more and less daylight throughout the year. Similar methods can be used to chop the year into seasons, time of day, or other more detailed analyses as needed.

How did you make the graphic?

This graphic was created by manipulating the output data created while using Diva to run annual daylight analyses. Excel was used to divide the number of hours of direct daylight at each grid location by the number of hours of direct daylight received by a flat surface in this location. This information as then fed back into Diva to produce the graphic display of the site.

What specific investigation questions led to the production of this graphic?

What areas of the site are too dark? Which areas get lots of sunlight? Should the massing be adjusted to change these results? Are certain exterior programmatic pieces more suited to some areas than others?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

While laying out building massing and exterior program elements, it became necessary to analyze the brightest and darkest areas of the site. While a series of typical shadow studies displaying shading at various times of day and year would give an idea of the lighting levels, this graphic combines that information into a single visual. This allows a viewer to immediately understand the levels of sunlight throughout the site. It also allows designers to manipulate the massing in order to shift the results according to design constraints.

What was successful and/or unique about the graphic in how it communicates information?

This graphic is able to combine a year's worth of exterior daylighting data into a single graphic. It is easily decipherable by those outside the construction industry and can be used throughout the design process to inform massing decisions.

What would you have done differently with the graphic if you had more time/fee?

It would be useful to make this a live graphic, such that any changes to the massing would be immediately reflected in the graphic. Daylight simulation constraints, as well as Diva for Grasshoppers interface, make this difficult to achieve.

What is the impact of orientation and WWR on the energy use?

PROJECT INFORMATION

Graphic Name:  What is the impact of orientation and WWR on the energy use?

Submitted by:

Firm Name:

Other contributors or acknowledgements (optional)

What tools did you use to create the graphic?

  • Sefaira

  • Excel

What kind of graphic is this?

Primary Inputs: Energy use comparison of higher glazing area with 40% WWR

Primary Outputs: Percentage increase in energy use

GRAPHIC INFORMATION

What are we looking at?

The graphic compares 2 orientations with different WWRs for the building. The baseline for both orientations is 40% WWR. The graphic compares the effect of increasing the WWR between 50% to 60% for all the facades. Given the higher glazing area in the 'triangular section' Facade D moves the needle on energy use in Orientation 1. However reorienting the building may help in reducing energy use while maintaining the same glazing area in the 'triangular section'.

How did you make the graphic?

Exported all the cases to excel and post processed it

What specific investigation questions led to the production of this graphic?

1) What is the effect of the orientation, given the higher glazing area n certain facades?

2) What is the effect of reducing the glazing area?

3) What is the effect of reorienting the building

4) What is the effect of various combinations of WWR with different orientations?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

Optimizing the facade, interior layout, circulation and other aspects of the architectural design would be affected by the orientation and the WWR.

What was successful and/or unique about the graphic in how it communicates information?

Stacked graph with 2 different scenarios quickly says which is the ' problematic facade in each of the cases.

What would you have done differently with the graphic if you had more time/fee?

Daylight or glare anaysis

What is the impact of the most important design parameters on multiple outputs: energy demand, thermal comfort and daylight availability?

PROJECT INFORMATION

Graphic Name: What is the impact of the most important design parameters on multiple outputs: energy demand, thermal comfort and daylight availability?

Submitted by: Torben Østergård

Firm Name: MOE Consulting Engineers

Other contributors or acknowledgements (optional) Aalborg University

What tools did you use to create the graphic?

  • D3

  • JS

What kind of graphic is this? 

Primary Inputs: The design space is represented by ten uniformly distributed inputs (sort right to left by overall sensitivity)

Primary Outputs: Energy demand, over temperature (unmet hours), and daylight factor (in teaching rooms)

GRAPHIC INFORMATION

What are we looking at?

The graphics displays the interactive parallel coordinate plot (PCP), which contains the results from 5.000 simulations of a 15.000 m² educational institution. The first 10 coordinates represent uniformly distributed inputs, which are sorted (right to left) by their combined sensitivity on the three outputs represented by the three rightmost coordinates. The outputs are constrained by three criteria: energy demand maximum of 41 kWh/m², overtemperature maximum of 0 kWh/m², and daylight factor of minimum 2%. Minimum and maximum limits are shown above the PCP. Moreover, the bar plots made using “real-time” sensitivity analysis (SA) indicate how much the different parameters have been affected by the current user-defined criteria. Here, the applied constraints have affected (room) reflectance and solar panels area the most. On contrary, the three input distributions to the far left show no or very small changes, which indicate they have no or insignificant influence on meeting the current constraints. For each coordinate, a histogram shows the distribution meeting the user-defined requirements. For instance, most solutions are found for the largest values for room reflectance and solar panels, whereas a SHGC of 0.42 is slightly favorable. The parallel coordinate plot, histograms, bar plots, and limits are updated interactively when adding or changing the user-defined constraints for both input and output coordinates. Thereby, the solution space and consequences are immediately highlighted and displayed when exploring the design space using this interactive graphics.

.

How did you make the graphic?

The graphics displays the results from Monte Carlo based building simulations. These simulations were made using Excel and the Danish normative software Be15. Excel is used to describe the parameter variations and run the “black box” Be15 calculations. The results are aggregated in Excel and uploaded to the PCP using a comma separated file. As part of my industrial PhD, we extended the features of a parallel coordinate plot available from D3.js. We added the histogram functionality and I developed sensitivity methods to rank inputs according to multiple outputs and to indicate how much each parameter is affected by user-defined constraints. The visualization is available at https://buildingdesign.moe.dk/phd/ibpsa.html. The site also allows the user to construct rapid metamodels based on neural networks. Thereby, millions of new predictions can be made in a few seconds. These will be displayed in another PCP below the original.

What specific investigation questions led to the production of this graphic?

In general: How can we explore a multi-dimensional design space in a multi-actor decision-making process? How can we help decision-makers navigate the PCP and highlight the most influential parameters and the most favorable input spans? How can we meet the computational challenges of Monte Carlo simulations? (leading to the metamodeling feature) How can we answer “what-if” questions Case specific for the shown example: Can we avoid photovoltaics and still meet the requirements? What is the appropriate window-to-wall-ratio which leads to the largest possible solutions space? How should the louvres (side fins) be design while balancing daylight availability with energy demand and thermal comfort? Is venting necessary and if so, how much air flow should it provide? Should the contractors aim for the lowest level of airtightness (infiltration) or focus on insulation levels of the facade and windows?

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

This type of interactive graphic is essential when dealing with multidimensional data such as Monte Carlo building simulations, which are growing in popularity. Indeed, a Monte Carlo based framework fits very well with the large uncertainties and variability, which are dominant i building design. The plot can be used recurrently in the iterative design process and help identify the solutions that meet the requirements of all decision-makers.

What was successful and/or unique about the graphic in how it communicates information?

The extension of the PCP with histograms, sensitivity analysis and metamodeling strengths the already popular PCP and makes it easier for the decision-makers to investigate an enormous multi-dimensional design space and observe the consequences of different decisions.

What would you have done differently with the graphic if you had more time/fee?

Primarily, I would like to develop and test another method for real-time SA based on variance-decomposition, which may be better than the current one based on the two-sample Smirnov test. In addition, I would like to extent and improve the metamodeling features (using larger neural networks) to improve accuracy of those. In general, I would like an experienced front-end develop to enhance user-flow and design. Finally, there are a few bugs that must be dealt with.

What is the impact of various retrofit strategies on energy saving?

PROJECT INFORMATION

Submitted by: Biwen Sun, Yingjia Wang

Other contributors or acknowledgements: Made Bagus Yudha, Nima Maghzi

ASHRAE Climate Zone: 5A

Building/Space Type: Mixed Use (Office, Lab, Classroom)

Who performed the simulation analysis?  University Student

What tools were used for the simulation analysis? 

  • eQuest

  • Climate Consultant

  • Grasshopper Ladybug

  • Grasshopper Honeybee

  • Computational Fluid Dynamics (CFD)

What tools did you use to create the graphic?

  • Excel

  • Adobe Photoshop

What phase of the project was analysis conducted?  Post Occupancy

What are the primary inputs of the analysis?  Equipment loads

What are the primary outputs of the analysis?  Electricity consumption, gas consumption

PROCESS

List the investigations questions that drove your analysis process.

How to build an energy simulation model close to the actual energy consumption of the existing building? Which zone in the building cost most energy? How can we improve the daylight environment in the building? How can we improve the ventilation? What are the possible strategies to improve the performance of the current building? How much energy can be saved possibly?

How was simulation integrated into the overall design process?

The energy simulation is to evaluate the current energy consumption of the building. The result can be used as the reference for the future renovation. The target is intended to cut down about 30% of total energy consumption. To achieve that, several strategies are simulated in the software to improve ventilation, indoor lighting environment, and building envelope performance. Based on the baseline model, improvement strategies are applied as EEM in eQuest, to show how much energy each strategy would possibly save. All the analysis and strategies are presented to the Facilities and Services Department of the University. They gave opinions on whether those strategies are feasible. Then, improvements are modified based on their advice.

How did you set up the simulation analysis and workflow?

The thermal load and occupancy schedule are generated into graphics through Grasshopper Ladybug and Honeybee. CFD is also a good tool to show the ventilation result in graphics. Other data like total electricity and energy consumption can be transferred into bar charts by using excel. The graphics clearly show the problem area and how much the performance would be possibly improved with the proposed strategies. With the graphics, the audience could understand the process easily and get a direct image of the outcome after renovation.

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

The thermal load and occupancy schedule are generated into graphics through Grasshopper Ladybug and Honeybee. CFD is also a good tool to show the ventilation result in graphics. Other data like total electricity and energy consumption can be transferred into bar charts by using excel.

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

The investigation is a great opportunity to know the complexity of the building and understand the depth of involvement required to make accurate building model. Also this process allows us to apply our current knowledge to test the feasibility of design strategies.

What's the impact of glazing specifications on peak loads and energy consumption? 

PROJECT INFORMATION

Graphic Name: What's the impact of glazing specifications on peak loads and energy consumption? 

Submitted by: Rufei Wang

Firm Name: Harvard Graduate School of Design

Other contributors or acknowledgements (optional) Huishan He, Holly W. Samuelson, Panagiotis Michalatos

What tools did you use to create the graphic?

  • Grasshopper Ladybug
  • Grasshopper Honeybee
  • Adobe InDesign
  • Grasshopper - WinEnergy

What kind of graphic is this? Scatter Plot

Primary Inputs: Glazing U-value, Glazing SHGC

Primary Outputs: Peak load in W/m2 and energy use in kWh/m2

GRAPHIC INFORMATION

What are we looking at?

The graph provides energy performance results for the full range of possible window properties at one time for a given climate, building type and orientation. Vertical and horizontal axes plot U-value and SHGC respectively, the most important window properties in energy performance. Gradient colors represent normalized total heating and cooling energy, and each plot represents one window type. Users can plot window products, using SHGC and U values on the graph to compare their respective performances.

The colorstep value is 2 kBtu/ft2 [6.3 kWh/m2], meanings that window products falling in the same color area produce similar on-site normalized total cooling and heating results (i.e. within 2 kBtu/ft2 [6.3 kWh/m2]).  Note: the heating and cooling values were produced from a full energy model of a perimeter zone.  Thus, the values take into account the window, internal gains, and other comprehensive energy model inputs.

How did you make the graphic?

The graph is the visualization from our proposed window selection tool. We’ve created 14 Grasshopper components for parametric simulation and data visualization, the code is written in C# using Visual Studio Platform. The workflow is as follows: 1. Conduct parametric simulation using self developed components and Grasshopper – Ladybug & Honeybee components. 2. Post process the data using self developed components. 3. Make screenshot from the tool visualization and further annotated the graph in Indesign.

What specific investigation questions led to the production of this graphic?

Choosing from the multitude of available fenestration products remains challenging. The existing window selection methods have limitations. We are proposing a new parametric workflow and visualization to better integrate with the early-design process.

How does this graphic fit into the larger design investigations and what did you learn from producing the graphic?

The graphic is part of our research for proposing a new workflow for window selection. We’ve researched the shortcomings of existing methods for selecting windows. We believe that the ability to visualize the impact of full range of window properties could be a beneficial aid to the early design process. A parametric simulation-based methodology was proposed. During the research, we’ve focused on exploring methods to simplify window system modelling and to increase simulation speed while maintaining desired accuracy.

What was successful and/or unique about the graphic in how it communicates information?

The graph allows the users to view the energy results of the entire set of window choices (assembly U-factor and solar heat gain coefficient) and immediately recognize trends in their energy performance. Users can also plot window products of interest on the graph to compare their respective performances.

What would you have done differently with the graphic if you had more time/fee?

The tool visualization legend is automatically determined from the maximum result value and the minimum result value. The visualization does not provide the option to fix the legend ranges. It can create confusion when two graphs are put together for comparison.