Facade design: The benefits of early-stage energy modelling

by arslan_ahmed | March 20, 2023 10:00 am

[1]

 By Jonathan Graham

In the early stages of design, it is useful to understand how facade design decisions will influence the energy performance of a building. Assumptions about window-to-wall ratio and wall thickness, no matter how preliminary, can limit a project’s ability to achieve certifications and/or comply with the performance path of the National Energy Code of Canada for Buildings (NECB). Indeed, as a project matures, it becomes increasingly difficult, as well as costly, to make effective changes to the performance. In response to this need, KPMB LAB, the research and innovation group at KPMB Architects, has developed a basic energy modelling tool for internal use on projects. This article discusses how the tool works, how it relates to more comprehensive energy modelling software, and how it can be useful to an architect during concept and schematic design phases.

Metrics such as thermal energy demand intensity (TEDI) and peak heating load are strongly correlated to the thermal performance of facades. Given these metrics are part of the performance criteria of green building standards and certifications (e.g. Toronto Green Standards [TGS], BC Energy Step Code, Canada Green Building Council [CAGBC] Zero Carbon, Passive House), it is crucial architects are able to calculate these metrics with respect to the contemplated facade. An inadequate facade can be difficult to remedy past schematic design phase, as it may require a thicker wall or lower glazing ratio.

Energy modelling softwares can calculate TEDI and hourly heating loads for every zone in a building. These models consider hundreds of factors affecting the energy balance of a building, including internal gains, solar gains, thermal storage, air leakage, and mechanical ventilation.¹ The trade-off for this comprehensiveness is the complexity of the software itself. The breadth of inputs can be overwhelming to a non-expert, and some of those inputs are bound to be unresolved at the early stages of design (e.g. the type of HVAC system).

At the opposite end of the spectrum, practitioners can use simple hand calculations to estimate TEDI and heating loads. For example, an approximate building energy balance can be modelled using steady-state heat transfer calculations.² Steady-state calculations represent heat flow through a system, assuming the system has reached equilibrium. This is in contrast to transient calculations, which represent heat flow as a time-varying phenomenon. Fourier’s Law of 1D conductive heat transfer is an example of steady-state heat transfer. The benefit of this approach is fewer technical inputs need to be known. The drawback is the results are less precise and less accurate.

KPMB LAB, the research group at KPMB Architects, has developed a tool that sits between these two opposites. The tool is a simplified energy model written in the programming language, Python. It is designed to quantify TEDI and heating load as functions of facade performance. With this narrow scope, the model excels at providing detailed insights while only requiring inputs that are obtainable during the concept and schematic design phases.

A purpose-built model

The model considers a single room, ventilated at a constant rate, with one external wall and one internal heat source. No computer-aided design (CAD) model is required, instead, the user specifies the room geometry through an Excel spreadsheet. The user provides thermal properties (U-values) for the external wall and its windows, a solar heat gain coefficient (SHGC), a solar transmittance (Tsol), window dimensions, and the orientation of the room relative to true North. The internal heat source represents a terminal device, such as a radiant panel or radiant floor, as identified by the user. The user also provides an EnergyPlus Weather File (EPW), which serves as the weather boundary condition. With these inputs, the model calculates the amount of heat energy needed from the terminal device to maintain an Operative Temperature setpoint (e.g. 21 C [69.8 F]) at the centre of the room. Effectively, it is measuring the heat required to compensate for a cold facade. The result is the heating load of the room, as influenced by the facade design. By repeating this calculation for each hour in the EPW file, the model can also calculate the TEDI.

[2]

The simplicity of the model’s inputs is a result of the underlying calculation methodologies. (Figure 1) Most of the equations used in the Python script are calculated by hand. For example, the inside surface temperature of the facade is calculated using Fourier’s law of 1D steady-state conduction. The effect of solar radiation on the window surface temperature is approximated using SHGC and Tsol.⁴ View factors, which are necessary for computing the mean radiant temperature (MRT), are calculated using geometric solid angle formulas.⁵ Ventilation heat losses are calculated using the familiar formula, Q = mcΔT, which only requires knowledge of the ventilation rate, inside-outside temperature difference, and the volumetric heat capacity of air. The advantage of implementing these basic methods in a computer script (opposed to performing the calculations by hand or in Excel), is a user can efficiently analyze multiple facade options, at an hourly timestep, with minimal inputs.

[3]

From outputs to insights

With some data processing and visualization, the model’s raw outputs are transformed into useful insights. Comparisons of the energy performance achieved by various facade options are especially useful during the early stages of design. For example, a comparison of monthly heating demand can show exactly when a higher performance facade is expected to generate the most energy savings. Figure 2 compares a low, average, and high-performance facade applied to a multi-unit residential building in Toronto. It is clear the high-performance facade would be most impactful during the winter.

A comparison of the hourly heating load can show how the building would respond to an extreme weather event, and if certain facades preclude the use of certain mechanical equipment (e.g. low temperature hot water systems). Figure 3 shows how the heating load on a radiant ceiling panel would respond to a cold snap in Toronto. The peak loads for the low performance and high-performance facades—1000 W and 600 W respectively—were determined to be within the capacity of the considered equipment.

[4]

Most importantly, a comparison of TEDI can help an architect understand which facades put the building on track to achieve green building certifications, such as TGS or Passive House. Figure 4 (page 3) compares three facade options across four compass orientations (North, East, South, West) and two glazing ratios (40 per cent and 80 per cent). Only the high performance, 40 per cent glazing options came close to the Passive House standard. Only the low performance, 80 per cent glazing options fared worse than TGS’s minimum requirement for residential buildings.

The Python script automatically generates these comparisons and accompanying charts, making it convenient to assess multiple facade options simultaneously. The script can also create more “visual” outputs, such as a spatial heat map of the operative temperatures in a room. These types of outputs can be useful for explaining how the energy required to maintain thermal comfort varies with facade performance. The example on page 3 (Figure 5) considers a radiant ceiling panel installed above a window of varying thermal performance. It is clear the mechanical system has to work harder to compensate for the low-performance window when compared to the high-performance window.

[5]

Accepting uncertainty

All predictive models come with limitations, and it is important to consider these in relation to the intended use. The model described in this article hinges on several assumptions about the physical world. It assumes no thermal storage is occurring in the building materials, and, consequently, exaggerates the heating load. Moreover, the model does not consider transmitted solar radiation; only the fraction absorbed by the glazing. Internal heat gains, from people, lights, and equipment, are also missing. These exclusions lead the model to overestimate heating demand. Most notably, the model does not include heat losses through the ground and roof, as it  only models a room on the middle floor of a building.

[6]

Given these limitations, the model is not suitable for proving the overall design of a building which meets a certain energy target or complies with a certain standard. For example, the model would not be able to prove a design which complies with Part 8 of the NECB. This is a task for comprehensive whole-building energy modelling software. Rather than competing for this scope, the model described here introduces energy modelling to the earliest stages of a project, where typically none would occur. During these stages, the “overall design” may be no more than a napkin sketch; therefore, the model’s simple inputs are its strength. With some basic geometries, thermal properties, and a weather file, the architect can compare, in relative terms, how various facade designs affect heating load/demand (e.g. option A is 50 per cent better than option B). Further, the model can give insight as to whether a design puts the building within the ballpark of a TEDI or peak heating load target. These insights are delivered rapidly by the Python script, which means an architect can iterate the facade to converge on an energy target. The benefit of performing these studies early is that costly changes to the wall thickness, glazing ratio, etc., can be avoided later in the design process.

The concept of using simulation to shape early-stage design is not new, nor is it unique to this example. For more than five years, the industry has seen the emergence of many cloud-based tools for assessing energy, daylight, and wind. Some of these are built on top of well-known simulation engines, while others are built from scratch. Both share an objective of increasing the role of simulation in building design by making the process quicker and more accessible to non-experts. While these tools have yet to become the norm in everyday practice, they promise to boost design efficiency in the future.

Notes

¹ DOE. (2021). Input Output Reference. In DOE, EnergyPlus Version 9.6.0 Documentation. U.S. Department of Energy.

² Feist, W., Bastian, Z., Ebel, W., Gollwitzer, E., Grove-Smith, Jessica, . . . Steiger, J. (2015). “Passive House Planning Package Version 9.” Darmstadt: Passive House Institute.

³ Graham, J., Turnbull, G., Constable, D., Reimer, M., & Vanwyck, J. (2022). “Modelling Perimeter Heating Demand: A Function of Occupant Thermal Comfort.” Facade Tectonics World Congress. Los Angeles: Facade Tectonics Institute. Retrieved from www.facadetectonics.org/papers/modelling-perimeter-thermal-energy[7].

⁴ Huizenga, C., Zhang, H., Mattelaer, P., Yu, T., Arens, E., & Lyons, P. (2005). “Window Performance for Human Thermal Comfort – Final Report.” Berkeley: Center for the Built Environment.

⁵ Tredre, B. E. (1964). “Assessment of mean radiant temperature in indoor environments.” British Journal of Industrial Medicine, 22, 58-66.

[8]Author

After graduating from Toronto Metropolitan University with a master’s degree in building science, Jonathan Graham joined KPMB Architects with a mission to help architects prioritize energy efficiency, carbon reductions, and occupant comfort in their design decisions. As an analyst with KPMB LAB, the research and innovation group at the firm, he specializes in using building performance simulation to test and rationalize sustainable strategies. He has conducted varied research on sustainable solutions for the architectural design process, and co-authored peer-reviewed journal publications about urban microclimate, solar photovoltaics, and thermal comfort modelling.
A certified Passive House designer, Graham leverages his expertise in building science and programming to create custom tools tailored to each design question.

Source URL: https://www.constructioncanada.net/facade-design-the-benefits-of-early-stage-energy-modelling/

---