OpenStudio Measures
The OpenStudio platform has been a cornerstone of energy modeling in both public and private practice. An energy model is a building constructed in a software program with the intention of running simulations to estimate energy consumption. The standard practice of energy modeling requires a separate model to be developed, however, the industry is moving towards a seamless integration with the Architectural and Engineering modeling workflows. For an energy model to be created there are typically three different interfaces. Using OpenStudio as an example, the three different interfaces are 3D modeling, setting inputs (variables/factors), and analyzing outputs.
3D modeling is the process of developing a mathematical coordinate-based representation of any surface of an object in 3-D using a specialized software. This process is the foundation of creating in the digital world so that it can be readily applied in the physical world. Initially this was done through feeding a software program a string or script that contained the coordinates and necessary instructions to connect those coordinates. As computers moved towards the consumer market the industry adapted many features to appeal to consumers through ease of use. Two of those features that remain key today are the mouse and graphical user interface, (GUI). A GUI is fairly straightforward in that it is a visual way of interacting with a computer using features such as, windows, icons, or menus for example. Essentially, people found it more appealing to select from a list or click a button rather than write lines of code.
This transition also occurred in energy modeling where most people think of OpenStudio as a energy modeling software program, however, OpenStudio itself is a GUI. Energy modeling is broken down into two different categories, inputs and outputs. Common practice today is to use a GUI, such as, OpenStudio to determine the inputs and an engine to perform the simulation creating the outputs. The outputs or results from the simulation can be analyzed using the input software program or an independent software program that specializes in post-processing. A typical engine is a machine with moving parts that converts power into motion, but really we are utilizing a hybrid between a physics engine and a game engine so that we “predict” the performance of a building. Think of our engine as a way to convert building operation and use into building energy consumption.
So, to summarize the common workflow that developed among energy modelers is to create a 3D model utilizing CAD (Computer Aided Drawing) software then specify inputs of the model using a GUI software like OpenStudio with the intent of investigating energy use intensity. A simulation is performed on the energy model by feeding it into an engine which creates the outputs or results that are then post-processed to interpret for analysis. While best practice work flows have emerged in our industry we still have the saying that if you give (10) energy modelers the same project you will get (10) different energy models.
Table 1 - Shows the steps taken by OpenStudio, software application, to run and analyze a simulation.
What are measures?
Okay, now lets talk measures. When I say measure I am referring to the following interpretation; To estimate or assess the extent, quality, value, or effect of something, but also, a plan or course of action taken to achieve a particular purpose. Both of these statements are true when considering OpenStudio measures. However, the technical definition as defined by OpenStudio “is a program (or ‘script, or ‘macro’, if you like) that can access and leverage the OpenStudio Model and API to create or make changes to a building energy model, as defined by an OpenStudio Model (.osm)”. API being applicable programing interface.
A measure edits the inputs of an energy model but also can edit portions that the OpenStudio interfaces lacks, such as EnergyPlus inputs (step E). An input file looks like this:
Figure 1 - An example of an OpenStudio measure.
Essentially, the input file is a collection of the settings, variables, and other applicable definition that were specified in OpenStudio. Therefore, if we know the format of the input file, i.e. where items will be specified, we can then write a measure (script) to change or modify the energy model. There are plenty of reasons to change a setting or definition in an energy model but why use measures instead of the GUI provided by OpenStudio? Measures provide efficiency over time by enabling users to rapidly change their energy model at various instances in the design process, but also, for every project after the measure was created. An example of this is the ability to rotate your model so that you can analyze the differences between say having your building orientated to true north or having it offset. However, the real power of measures comes from layering them. For instance, we can have our model start at true north then rotate in increments of (15) degrees. From here we can re-run the simulations but alter our window to wall area ratio (WWR) at each rotation interval to measure the difference between, for example, 15% and 20% WWR. Let’s say that we want to assess the potential of daylight for our site and building. We have identified two main variables that effect daylighting and the other main variable we have to consider is visible transmission or Tvis. In this scenario we want to identify not only which variable has the greatest impact on our daylighting, but also, the relationship between building orientation, WWR, and Tvis while considering site context.
Workflow when using measures
Table 2 -
The table 2, above, shows how measures can be layered to investigate different design decisions impact on the energy use of a building. Changing the building’s orientation allows us to investigate the solar heat gain or exposure our building will receive throughout the year which will effect our heating and cooling loads. In addition, we can ‘test’ different window fenestrations and glazing properties that will assist our building in reducing our cooling or heating loads. Between these (3) measures we are trying to establish a floor and ceiling in regards to energy performance. As in which ‘values’ for our inputs allow us to minimize the amount of cooling and heating required (ceiling), but also, which ‘values’ will allow us to construct a code minimum building (floor). Once we have established our floor and ceiling values we can investigate what is the optimal ‘values’ for our inputs as it relates to design, for instance, spatial organization/relationships or intended use, however, a more common and powerful factor that is used to determine optimal values is economics. Another approach is to have ‘hard values’ which can be based on code standards where, for instance, our window to wall area ratio cannot exceed 15%. Hard values can also be derived from site context or building typologies, but regardless of how they are determined they should be used with caution.
Now, let’s say we have been asked by a client to investigate daylighting either as a primary lighting strategy or perhaps the client wants LEED accreditation and you have to investigate if pursuing LEED credits for daylighting is a viable design option. We can use our measures to investigate which design choices will have the most impact on daylighting but also which values or inputs will be optimal. Optimal being what is the lowest window to wall ratio and transmissibility we can have at ‘X’ orientation while trying to maintain a high solar heat gain coefficient.
For this example let’s pursue LEED accreditation and see how well our building performs using LEED metrics.
Table 3 - LEED V4 daylighting standard.
Some basic information about our building:
Figure 2 - Screen capture of building.
Our building is a DMV for a county that will remain nameless. Now before we actually run any simulation or determine inputs I want to do some simple math. With the current data available to use we can create a ball-park estimation of our building performance as it relates to daylighting.
Table 4 - Area breakdown using different standards.
• Area – total sum of the building floor area
• LEED v4 Area – total sum of the regularly occupied spaces floor area
• IECC 2018 Area – the total sum of area identified as daylight harvesting
In order to achieve credits for LEED daylighting, option 1, we must ensure that 55% of our floor area (regularly occupied spaces) is daylit to a standard of sDA 300/50%. If we calculate 55% of our eligible LEED v4 area we can see that 4,136.63 feet squared must meet or exceed a sDA 300/50%. Using our IECC daylight harvesting area we can see that code acknowledges 1,177 feet squared is actively being daylit. While I consider this a crude metric for measuring daylight it is useful here because it relies on glazing location and infers how much building area has access to daylight. Therefore, only 17.8% of our building floor area has access to daylight. We can break this down further by identifying the area to façade orientation to infer how effective our daylight sources could be. Referencing the table below, we can see that 30.5% of our daylighting area is located on the north side of the building and while there is daylight it is not enough to meet the sDA 300/50% threshold.
Table 5 - The building's access to daylight as it relates to floor area.
If your building has low access to daylight it is recommended that you do not pursue LEED daylighting credits, however, if you choose to pursue them it should be established upfront that significant investment in daylighting sources and strategy are needed. Our example score comes in at 17.8% daylighting area, but also, 12.38% of effective daylighting area.
Figure 3 - IECC 2018 daylight harvesting area for side lighting applications.
Figure 4 - Buildings daylight harvesting area according to IECC 2018 standard.
Floors, Ceilings, and Goals
While I would not recommend pursuing LEED daylighting credits for this building let’s pretend that it is a requirement for this project. The table below shows the runs we can perform to analyze our building’s daylighting performance to figure out what it will take for us to achieve daylighting credits.
To start analyzing our building’s daylight performance we have to establish a baseline. Until now we have been assessing our building’s potential for daylight giving us a ballpark estimate. By establishing a baseline we can know where we stand to see how far we have to go. For instance, if we increased our WWR on the south façade as well as adding windows to the east and west facades how will we know how effective those decisions will be. Because daylighting is defined in a range we will have several scenarios to present, therefore, we need to be able to frame our analysis in a IF THEN statement to assist with design decisions.
• If we increase the WWR on the south façade and add windows to the east and west façade then the daylight performance will increase from ##% to ##%.
Due to our initial daylighting area being so low we need to identify the performance of each change in the daylighting strategies from our baseline. The statement above contains three changes to the building’s daylighting strategies which can be broken down into the following:
If we increase the WWR on the south façade then the daylight performance will increase from ##% to ##%.
• If we add windows to the east façade then the daylight performance will increase from ##% to ##%.
• If we add windows to the west façade then the daylight performance will increase from ##% to ##%.
The optimal settings for your energy model depends on your goal. In this example our goal is to design a building that meets LEED V4 daylighting criteria. Therefore, I like to ‘grade’ the glazing which usually means separating the glazing into two main categories, performance and aesthetics. The performance is defined as glazing that interacts with a ‘regularly occupied space’ and aesthetics contains glazing that interacts with non ‘regularly occupied space’. This distinction is specifically for LEED evaluation and is not to say that the aesthetic glazing does not provide daylight or a purpose but rather it has no impact on the LEED performance criteria. Therefore, in this context, it has no value. Within the performance category the glazing is separated into (3) categories of failing, potential, and passing. In my opinion this helps convey the performance aspects of the glazing for each façade which can inform design decisions regarding where to expand or contract the WWR.
Referencing the table below, I like to start around the 50% mark for Tvis unless I have been informed otherwise by the client or site conditions. We will keep the basic (4) variables of rotating our model in (15) degree intervals because the Architectural team is still trying to figure out the spatial arrangement or preferred relationships and we don’t want to limit the design yet. Keeping our Tvis value consistent while manipulating our WWR by 10% allows us to compare how much area we fail to meet, barely meet, or exceed our sDA of 300/50 at 40%, 55%, or 75% (see table 6).
Table 6 -
As I mentioned earlier you can frame the results as a if then statement, for example, if the WWR for the south façade is at least 23% with a VT of 35% then the spaces that utilize the daylighting will meet an sDA 300/50% for at least 55% of the area. However, it could also be written another way, such as, if the WWR for the south façade is at least 15% with a VT of 45% then the spaces that utilize the daylighting will meet an sDA 300/50% for at least 55% of the area. In both statements the spaces along the south façade are considered passing, so why present multiple options?
The first statement uses a higher WWR while minimizing the glazing VT which generally means a higher U-value while the second statement is the inverse. Presenting both options provides the opportunity for the design team to determine which route they can pursue. We present this information to the design team (Architecture) and also at the meeting is the Engineering team. They inform us that they will be adhering to ASHRAE 90.1 – 2016 standards and that it would be ideal if the WWR could be between 10.1% to 20.0%(shown below).
Table 7 - ASHRAE 90.1 - 2016 standard.
Table G3.4-5 Performance Rating Method Building Envelope Requirements for Climate Zone 5
Based on that information we have new limitations for our daylighting strategies. Up to this point we have limited our efforts to the south façade but now we will have to expand our investigation to include the east and west façade, but also, we may have to consider skylights. This is because our WWR ratio is now limited to 10.1 to 20% as well as locking in our Max U to .57 and SHGC to .39. Luckily, none of our façade’s WWR exceeds 20% which means we are below our ceiling, see table X below.
Table 8 - Building facade area and allowance of area according to ASHRAE standard 90.1 - 2016.
Referencing the tables (##) above we can infer how much wiggle room we have to increase our daylighting performance using sidelighting as the primary daylighting strategy. It should be noted that our highest performing façade, south, is almost maxed out with a WWR of 17.41%. So any notable performance increase would have to come from either the East or West facades, however, to achieve that performance output we would have to rotate our building. Therefore, I would inform either the project Architect or Engineering that if we only use sidelighting for daylighting and don’t rotate the building to improve either the west or east façade’s performance then the building will not meet LEED criteria. This allows for a constructive dialog because while it presents a limitation we have to discuss a path forward. For instance, if we don’t rotate the building and still want to pursue LEED daylighting credits then we have to invest in additional daylighting strategies. Also, we know that the daylight strategy to pursue next is skylights rather than clerestories due to the limitations on our WWR. When considering skylights we have the following possibilities:
Table 9 - ASHRAE standard 90.1 - 2016 skylight fenestration requirements.
With this information we can simply rinse and repeat the process describe above for sidelighting but now with skylights. The measure we use instead of affecting WWR, wall assembly, and visual transmission will need to focus on our roof. It’s still the same format and structure but instead of affecting or changing the inputs of the wall our variable is now the roof.
Conclusion
Measures are a method by which designers can perform parametric analysis on buildings in a timeframe that is accelerated so that information can be processed in the early stages of design. The ideal goal of this process is to more readily allow architects and engineers to design high performance buildings to fulfill their clients demands, occupant comfort, and reduce the project’s impact on the environment when built, but also, as it operates. Parametric, as defined by mathematics, defines a group of quantities as function of one or more independent variables called parameters. For our purposes we can consider parametric to be a set of facts or a fixed limit that establishes or limits how something can or must happen or be done. In the example above we allowed code to establish our limits by creating a floor (minimum) and ceiling (maximum) for each variable that impacted our building performance. Through this method we develop the following workflow:
• Define goal of the project, usually the standard being applied.
• Identify variables or inputs that will directly impact your goals.
• Of those variables associate them with other aspects of building performance and design to determine the effect on the rest of the design.
• Determine the floor and ceiling for your inputs.
• Perform simulations and analysis to define the range of performance for your building.
• Perform simulations and analysis at different intervals between floor and ceiling inputs.
• Use analysis to inform design decisions.
• Rinse and repeat until you arrive at construction documentation by which you will need to perform one last simulation and analysis for compliance.
The workflow above is intended for local use instead of the cloud meaning that simulations are performed as well as hosted by your computer. If you are performing somewhere between (5) to (15) simulations over the design process then the local workflow method can meet the Architect’s or Engineers needs for performance design. Cloud computing does not change the workflow describe above, but rather, takes it to the logical conclusion of computing or consider every single ‘what if’. For instance, an Insight simulation which considers every possible what if for a building’s design results an average of (250) simulations. Because Insight runs so many ‘what if’ simulations it allows the software program to present the analysis to directly assist Architects and Engineers with design decisions. The main methodology utilized by Insight is to present analysis in the form of widgets that have slider bars. Architect, Engineer, or clients can interact with the widget by moving a slider to change the input value of the variable. The software program then updates the analysis, building EUI, based on the new value which visually shows the effect of design decisions. This not only helps explain results to Architects and Engineers, but also, helps them convey the importance of some design decisions to their clients.
OpenStudio has their own version of cloud simulation which conveniently is called the Parametric Analysis Tool. This tool allows you to layer the workflow above by allowing users to set-up multiple simulation runs and change the inputs of variables to test ‘what if’ scenarios through applying measures. The notable difference between the Parametric Analysis Tool and Insight is that you define the number of simulations to run, whereas, Insight will always run every possible simulation. This is a notable difference because it affects the cost of cloud simulations. Cloud simulation cost is determined by the amount of computing power it will take to complete the given task, i.e., the time it will take to complete the simulation(s). Therefore, it is up to you to determine which methodology is correct for your project.
This is part two of a series that is based on parametric analysis and modeling. If you found the information in this blog helpful please let us know so we can continue to create content around this topic.
Dylan Agnes
After earning a Bachelor of Science in Architecture from the University of Idaho, Moscow, Dylan studied the science and engineering of building design, completing a Master's in Architecture with an emphasis in urban planning and net-zero/energy efficiency building design. As a student he worked at the Integrated Design Lab and gained hands-on experience in the practice of Integrated Design. As an IDL Research assistant, Dylan worked with both the architectural and engineering side of integrated design, providing a broader opportunity to cross over fields of study. Since graduation, Dylan has been working as a Research Scientist at the IDL and has been working on a wide range of projects from Energy Modeling to Daylighting Design.
References
