Introduction
OceanTM, developed by Eclat Digital, is a spectral ray tracer that provides high accuracy illumination calculations. Thanks to its specific spectral features, it allows a highly physically accurate description of materials and provides physically true predictive images. Relying on a valid and strong physical approach, it guarantees a perfect match between renders and real world and can therefore be used for assessing the quantity of light received by the elements of a scene, for instance passing through glazing structures and reaching the ground of a building.
When dealing with physico-realistic rendering, namely in the case of daylighting studies [1] , lighting conditions play a major role. Thus, be able to get weather information, namely radiative observables, for a given location is necessary to simulate realistic illumination conditions.
This article deals with the use of weather data in our rendering software OceanTM. We first focus on the weather data needed, then we discuss the OceanTM Sky Importer new feature before switching to use case examples.
Weather data
It currently exists several sources providing weather data [2]. As it is mentioned in the literature, it is important to avoid weather sources using only a single year of measurements (TRY, TMY…) when dealing with quantified radiation study, since these sources are not able to well represent a typical long-term weather behavior (and so, the procedural skies already present in OceanTM are not suitable for this kind of studies).
This is why, we will focus only on data coming from EnergyPlus website (https://energyplus.net/weather), whose the principal source is the International Weather for Energy Calculations (IWEC) [3], ensuring a sufficiently long data acquisition time (~15 years). List of sites providing EnergyPlus Weather (EPW) files:
- https://www.ladybug.tools/epwmap/
- https://energyplus.net/weather (reliable for daylighting studies)
- http://climate.onebuilding.org/
This section will list the elements of importance for the creation of realistic illumination conditions into Ocean™. The format of an EPW file can be found here: EnergyPlus documentation, Auxiliary Programs. It is generally composed of a header of 8 rows.
Figure 1 - Example of the header of an EPW file.
In this header, it is important to check if the source is IWEC data (first line, position 5). This is important to ensure that data are reliable. Then, for the creation of Ocean™ environment, we will need the city name, its position (longitude and latitude) and the time zone. The three last information is needed to compute the sun position. The rest of the header is not needed.
Let us now switch to the body of the document. Each row contains a lot of information, such as the date, illuminance data, atmospheric information, radiation data… We need first to get the date and time information (year, month, day, hour, minute).
Figure 2 - Example of the body of an EPW file.
Then, we will focus only on radiation information: the DNI and DHI measurements. DNI stands for the direct normal radiation, in Wh/m², and represents the amount of solar radiation received on a surface perpendicular to the rays of the sun, during 60 minutes prior to the corresponding datetime. DHI stands for the diffuse horizontal radiation, in Wh/m², and represents the amount of sky radiation (only the diffuse part, meaning all the sky excluding the solar disk) received on a horizontal surface, during 60 minutes prior to the corresponding datetime.
In summary, what we get from the EPW file is, for each row:
- The date and time,
- The location: longitude and latitude,
- The DNI,
- The DHI.
From that, we can build realistic illumination conditions, based on the association of a direct sun model to represent the direct component of the solar radiation and the Perez [4] all sky model to capture the diffuse part.
Ocean™ Sky Importer
The new OceanTM sky importer dialog allows importing weather data into Ocean™. For more information please refer to the documentation page https://docs.eclat-digital.com/ocean2021-docs/reference/toolbox/skyconverter.html
When importing weather data, created lighting conditions will be composed of two contributions: a direct sun model and the Perez all sky model
Depending on the study you want, you can concatenate all the hourly illumination conditions for a quantitative study over a longer period (> 1 hour), or seperate illumination conditions by step of one hour. Examples are given in figure 4.
Figure 4 - Sky importer examples. Left: Corresponding result when not splitting at the import. Only one environment is in the Objects toolbox. The concatenated environment is an additive environment of sub additive environments. Right: Corresponding result when splitting at the import. A list of environments is in the Objects toolbox.
Let me now discuss in detail the rotation angle option. This option allows to modify the azimuth position of the Sun in the sky. Indeed, sometimes when dealing with large 3D model of building which is not correctly oriented compared to the geographic East, it is easier to modify the Sun position than to modify the orientation of the 3D model.
Let us, for the explanation, consider a building aligned with the X axis of the 3D software.
In reality, the building will be built and oriented toward the northeast (the red square is in reality at an angle of 35° from the East axis).
In Rhinoceros, for instance, the X axis is equivalent to the East axis of Ocean™ environment orientation, and the Y axis is equivalent to the North axis of Ocean™ environment orientation. A natural solution will be to rotate the entire 3D model to orient it correctly. However, when dealing with customer 3D model, it is common to have to modify or add elements to the CAD during the project. If we rotate the orientation of the model on our side, it will no longer have the same reference point as the client’s model, and the integration of new elements will be an ordeal (really).
The other solution is not to change the orientation of the 3D model, but to change the position of the Sun accordingly. Thus, in our case, modifying the sun position is equivalent to rotate the initial coordinate frame, as described in the sketch.
This is equivalent to rotating the initial reference frame (X, Y) by an angle of -35° (trigonometric convention). This gives a new coordinate frame (X’, Y’) which matches the real orientation of the building. -35 is the value the user must enter in the rotation parameter, at the import.
In terms of Sun positions, let us for the demonstration consider a morning sun, with an azimuthal position of , meaning that the Sun is perfectly align with the East axis (see sketch below on the left). In the case of our 3D model not correctly oriented, we have rotated the coordinate frame of -35° (see sketch below on the right). As we can see, in both cases, the sun is aligned with the right bottom corner of the building, so the lighting conditions will be realistic with regard to the orientation of the considered building.
Use case examples
1. Validation on simple scene
To validate the sky importer and capabilities of OceanTM to accurately model sky luminance distribution, a benchmarking case study was performed considering existing validated software: EnergyPlus [5] and Radiance [6]. The yearly irradiances (radiant flux per unit area received by an object in [W/m²]) over the 5 different faces of a cube (see Figure 6) are calculated with the three softwares. The faces’ cube are simulated with a basic Lambertian material. The lighting conditions of a full year were simulated (in the case of OceanTM, the lighting conditions were generated with the sky importer). A direct sun model to represent the direct contribution of the solar radiation (the sun) and the Perez [4] all sky model to capture the diffuse part (the atmosphere, clouds etc) were used. The IWEC weather file of Brussels available EnegyPlus website [5] was used to describe the weather conditions. The results of the comparison are given in Figure 5: whatever the considered face, the irradiances computed with the three software are consistent. The deviation can be fairly attributed to the diffuse sky model which appears to be slightly different between the three software, especially between EnergyPlus and the other 2. Other discrepancies could come from resolution used to sample the Perez sky in Radiance and OceanTM. Nevertheless, this validation case consolidates the accuracy of the OceanTM algorithm.
Figure 6 - Cube geometry
Figure 5 -Comparison of the yearly irradiance over the faces of a cube, calculated with the three software (EnergyPlus [5], Radiance [6] and OceanTM.)
2. Yearly irradiance study on simple scene
Now that we have demonstrated the correct functioning of the sky importer, we will illustrate a yearly study using the same simple scene as in IV.1. Indeed, as it will be discussed in the following of the article, questions arise when it comes to design a building and its facades (use of solar shading systems, orientations of the building…) in order to control and optimize the quantity of light entering the building.
In this small example, we want to deal with yearly study on a simple scene, to evaluate the exposition of the facades to the sun. The idea here is to use the sky importer to create the ~4500 lighting conditions (1 year of lighting by step of 1 hour), and to proceed to the simulations to retrieve the quantity of light reaching the several elements of the scene, i.e in this case the 5 faces of the cube.
Once the weather data imported into Ocean™, we use the project manager of Ocean™ to automate the preparation and the execution of the batch of the renders (https://eclat-digital.com/ocean2021-docs/reference/toolbox/projectmanager.html?highlight=render%20project).
The obtain results are shown in Figure 7. It represents the irradiance of each faces as a function of the time. As we could expect, whatever the considered face, the irradiance globally increases during summer, and decreases during winter.
Figure 7 - Yearly irradiances by step of 1 hour. Top left: East face. Top right: North face. Bottom left: West face. Bottom right: South face.
As previously mentioned, we carried out the study on a really simple scene, but this kind of studies can be done on more complex scene, and for several parameters: orientation and inclination of the facades (could be windows, solar panels etc), the location of the building and its surroundings (shadowing effects..) and so on.
3. Building simulations
In this section, we propose a simple case of building simulations with OceanTM. During the construction of a building, there are often questions about glazings, especially for energy reasons, but also about the installation of solar cells (see section IV.2). Of course, it is of paramount importance to be able to simulate the building in its future environment, and to be able to simulate, over a year for example, its illumination.
For the demonstration, we consider a complete 3D building with its internal and external environment, at two different locations: Jeddah in Saudi Arabia and Brussels in Belgium. Once the 3D imported into OceanTM, dedicated optical properties are assigned to each element of the scene. These optical properties can be fully spectral. For instance, for a glazing panel, we have the possibility to use as inputs reflectance and transmittance spectra for several angle of incidence, giving a very accurate description of the material.
We simulate the lighting conditions for the two locations for the 21st of June of 2020. The two corresponding environments are created and imported with the OceanTM Sky Importer. The obtained simulation results are shown in Figure 8 and Figure 9 for the outside and inside irradiance respectively. Indeed, in addition to aesthetic renders, OceanTM can provide details daylight metrics. So, irradiance maps of the outside of the building and of one office inside the building were calculated. As we see, depending on the location, and so depending on the sun position in the sky, the distribution and the quantity of the light received by the objects in the scene are different. In this example, for Jeddah, the sun is very high in altitude (87°, almost at zenith). The sun does not penetrate a lot into the office due to the sun position, but also due to all the balconies that induce shadowing effects. These shadowing effects are visible on the outside irradiance map, the building facades does not receive direct lighting. In contrary, for Brussels, the sun being less high in altitude, more light can reach the facades and so more light enters the office.
Figure 8 - Irradiance for the outside of the building. From top to bottom: Brussels and Jeddah. The color scale is in W/m².
Figure 9 - Irradiance for the inside of the building. From top to bottom: Brussels and Jeddah. The color scale is in W/m².
Beyond the sky limits
The new OceanTM sky importer permits to import weather data. It currently supports EPW files.
This feature allows you to import weather data and create realistic illumination conditions for a more predictive result. This will help you to deal with quantification studies such as daylighting studies on buildings but also to proceed to the quantification of the irradiance inside a building (e. g. in a greenhouse), to evaluate Photovoltaic panels illumination, to check for solar concentration.
Another possibility is to proceed to sunrise to sunset animation for widely detailed 3D scenes, for instance for evaluating the daily distribution of light over crops in a greenhouse.
References
N. Jakica, «State-of-the-art review of solar design tools and methods for assessing daylighting and solar potential for building-integrated photovoltaics,» Renewable and Sustainable Energy Reviews, vol. 81, n° %110.1016, 2017.
D. B. C. f. ASHRAE. [En ligne]. Available: https://energyplus.net/assets/nrel_custom/weather/whichweatherdatashouldyouuseforenergysimulations.pdf.
ASHRAE, «International Weather for Energy Calculations (IWEC Weather Files) Users Manual and CD-ROM,» 2001.
R. Perez, R. Seals et J. Michalsky, «An all-weather model for sky luminance distribution,» Solar Energy, vol. 50, pp. 235-245, 1993.
«EnergyPlus Software,» [En ligne]. Available: https://energyplus.net/. [Accès le 1 December 2020].
«Radiance software,» [En ligne]. Available: https://radiance-online.org/. [Accès le 1 December 2020].
Responses