Conference Agenda

Session
Session W3.5: Improving indoor environmental quality
Time:
Wednesday, 01/Sept/2021:
14:40 - 16:10

Session Chair: Natalia Giraldo Vasquez, Federal University of Santa Catarina
Session Chair: Jean-Baptiste BOUVENOT, INSA Strasbourg/ICube Laboratory
Location: Concert Hall - Kamermuziekzaal

't Zand 34, Bruges

External Resource: Click here to join the livestream. Only registered participants have received the access code for the livestream.
Presentations
14:40 - 14:58

The impact of light distribution and furniture layout on meeting light exposure objectives in an office - a simulation case study

Megan Danell1, Steffen Hartmeyer1, Lisa Petterson2, Robert Davis3, Marilyne Andersen1, Siobhan Rockcastle4

1Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland; 2SRG Partnership, Portland, OR United States of America; 3Pacific Northwest National Lab, Portland, OR, United States of America; 4University of Oregon, Eugene, OR United States of America

Aim and Approach

(max 200 words)

While the building industry is beginning to embrace the impact of light on non-visual responses that drive human health, there are limited design guidelines for how to implement effective light distribution and achieve recommended circadian light exposure. Human-centric factors that impact eye-level light exposure in a private office include the seating location, view direction, and eye level of an occupant, which interact with the light distribution, intensity, and spectrum of any given light source. This study presents a simulation-based study that compares various light distribution patterns, furniture configurations, and seating/standing positions to illustrate the impacts on human health potential from a non-visual health perspective.

The space used in our case study is an existing architectural office located in Portland, OR. Multiple luminaires and light distribution patterns are compared under different furniture layouts. The ALFA (Adaptive Lighting for Alertness) plug-in for Rhino is used to simulate Equivalent Melanopic Lux (EML) values for a series of hourly and daily time steps [4]. Electric light sources are simulated using industry-standard IES files and results are compared for both seated and standing view positions.

Scientific Innovation and Relevance

(max 200 words)

A number of simulation techniques have emerged in recent years as a means to predict how varying factors within a space affect non-visual light exposure for building occupants. These techniques include the simulation of vertical illuminance values in incremental measurements across one or more view directions [3, 4, 6]. Recent work has also compared the impact of daylight [1, 3, 4, 5] and electric lighting sources [4, 5, 6, 7] on the health of indoor occupants. A recent study compared the impact of various overhead light sources on vertical eye-level exposure and horizontal task-plane illuminance [2]. This paper builds upon these studies by comparing a broader range of occupant-centric and spatial conditions that can impact the lighting design of an office space.

The novelty of our research lies in the comparison of light distribution patterns, furniture layout, and ergonomics within a private office space. Because eye-level light exposure accounts for light reflected off of vertical surfaces, luminaires that distribute light onto vertical surfaces and are located closer to the eye are potential assets for improving healthy lighting conditions. This work has the potential to bridge research with practical lighting design recommendations to improve the health and well-being of occupants located in private offices.

Preliminary Results and Conclusions

(max 200 words)

Our results reveal a variety of intriguing outcomes. Intuitively, the combination of electric light and daylight sources systematically outperforms scenarios that rely exclusively on electric light only. While 36 out of 48 simulated scenarios achieved the minimum threshold for the WELL Building Standard of 150 melanopic lux (assuming continuous exposure between 9am and 1pm), less than half achieved the recommended 250 melanopic lux, despite achieving recommended task-plane illuminance values.

Comparing results from the various light distribution scenarios (direct vs. direct/indirect ceiling-mounted luminaires and wall-wash luminaires), the fixtures that provided a significant wall wash component achieved the highest EML values. This indicates the potential for vertical light distribution to act as a source of healthy light exposure that has not yet been thoroughly studied. A comparison of various ergonomic positions also reveals the variability in EML exposure between standing and seated positions. Further development of this proposal would contribute to the currently limited design guidelines for implementing effective lighting design to achieve circadian light exposure through the use of furniture configurations, eye level, and lighting distributions.

Main References

(max 200 words)

[1] Acosta, I., Leslie, R., & Figueiro, M. (2017). Analysis of circadian stimulus allowed by daylighting in hospital rooms. Lighting Research & Technology, 49(1), 49–61.

[2] Jarboe, C., Snyder, J., & Figueiro, M. (2020). The effectiveness of light-emitting diode lighting for providing circadian stimulus in office spaces while minimizing energy use. Lighting Research & Technology, 52(2), 167–188.

[3] Amundadottir, M., Rockcastle, S., Sarey Khanie, M., & Andersen, M. (2017). A human-centric approach to assess daylight in buildings for non-visual health potential, visual interest and gaze behavior. Building and Environment, 113, 5–21.

[4] Saiedlue, S., Amirazar, A., Hu, J., & Place, W. (2019). Assessing Circadian Stimulus Potential of Lighting Systems in Office Buildings by Simulations. ARCC Conference Repository.

[5] Danell, M., Amundaddottir, M. L., & Rockcastle, S. (2020). Evaluating Temporal and Spatial Light Exposure Profiles for Typical Building Occupants. SimAUD Conference Proceedings.

[6] Dai, Q., Huang, Y., Hao, L., Lin, Y., & Chen, K. (2018). Spatial and spectral illumination design for energy-efficient circadian lighting. Building and Environment, 146, 216–225.

[7] Rockcastle, S., Danell M., Petterson, L., & Amundadottir, M. (2020). The Impact of Behavior on Healhty Circadian Light Exposure Under Daylight and Electric Lighting Simulations. ACEEE Conference Proceedings.



14:58 - 15:16

Evaluating the use of photobiology-driven alertness and health measures for circadian lighting design

Athina Ji-Hae Alight, J. Alstan Jakubiec

University of Toronto, John H. Daniels Faculty of Architecture, Landscape, and Design, Toronto, Canada

Aim and Approach

(max 200 words)

The aim of this project is to evaluate a novel daylighting and electric lighting design workflow that assesses a space based upon light’s impact on human photobiology-driven alertness, and health. The method, being published in a separate submission to the conference, works by translating timeseries spectrally-resolved light simulation data into photobiologically driven measures mediated by the response of intrinsically-photosensitive retinal ganglion cells (ipRGCs) in the human eye. These measures are used as input to a dynamic photobiological framework that accounts for light history, timing, spectrum, and homeostatic body rhythms. These measures are then visualised in a novel manner to communicate the impacts of lighting on space occupants. To demonstrate the value of this process, the predicted non-visual biological effects of six design variables (artificial lighting schedules, artificial light spectrum, occupant location, window spectral transmittance, surface reflectance, and two space designs) are simulated. The design variables are also applied to the frameworks suggested by Mardaljevic et al. (2013), Amundadottir et al. (2017), the WELL standard (2018), and Konis (2019) to test how these models respond to variations in spectrum and light exposure and how they differ in the resulting evaluation of architectural design.

Scientific Innovation and Relevance

(max 200 words)

Several frameworks have been developed to evaluate lighting design for non-visual biological effects, which this paper compares and evaluates. Unlike light for visual tasks, the non-visual/circadian system, is sensitive to the timing of light exposure. The circadian system is also responsive to shorter wavelengths than the visual system. Illuminance that is appropriate to perform visual tasks therefore may not be enough to entrain circadian rhythms or maintain alertness and performance. Insufficient light after waking or excess light before sleep onset disrupts circadian rhythms with harmful health consequences. Furthermore, circadian disruption is cumulative, and depends on circadian entrainment in the recent past. As a result of these complexities, our model takes the history, timing, intensity, and spectrum of light exposure into account. To the best of our knowledge, the workflow presented in this paper is the only one that predicts explicit biological effects of light and spectrum over time rather than circadian light potential. Through implementing our model, we will demonstrate the results of the model and how architectural design can directly impact an occupant’s circadian health. Our comparative analysis will showcase how predicting alertness and health measures differs from previous work and impacts the evaluation of architecture.

Preliminary Results and Conclusions

(max 200 words)

For the six design parameters and using our new model, the following alertness and health photobiological measures are calculated: subjective alertness measured by the Karolina Sleepiness Scale, mean reaction time, attention lapses, and performance at rote tasks, the amount of circadian phase shift per day, the time of peak melatonin concentration, and the percent of total melatonin suppression per day due to acute light exposure.

The comparative analysis with previous models is based on two aspects: (1) comparisons of quantitative lighting units and (2) assessing differences in design evaluation outcomes as positive or negative. A compilation of the circadian calculation and evaluation methods used in existing photobiological lighting design evaluative frameworks (Mardaljevic et al. 2013, Amundadottir 2017, WELL Standard 2018, Konis 2019) are compared in this manner.. Each framework’s responsiveness to lighting types (daylight, electric, or hybrid), evaluation timeframe (instantaneous, daily, or annual), and lighting design interventions is also compared and discussed.

Main References

(max 200 words)

Amundadottir, M. L., Rockcastle, S., Khanie, M. S., & Andersen, M. (2017). A human-centric approach to assess daylight in buildings for non-visual health potential, visual interest and gaze behavior. Building and Environment, 113, 5-21.

Circadian lighting design. (2018). https://standard.wellcertified.com/light/circadian-lighting-design.

International WELL Building Institute, WELL Building Standard Circadian Lighting Design Feature. https://standard.wellcertified.com/light/circadian-lighting-design Last accessed 6/3/2020.

Konis, K. (2019). A circadian design assist tool to evaluate daylight access in buildings for human biological lighting needs. Solar Energy, 191, 449-458.

Mardaljevic, J., Andersen, M., Roy, N., & Christoffersen, J. (2013). A framework for predicting the non-visual effects of daylight–Part II: The simulation model. Lighting Research & Technology, 46(4), 388-406.

Solemma.com. ALFA. https://solemma.com/Alfa.html.

Tekieh, T., Lockey, S. W., Robinson, P. A., McCloskey, S., Zobaer, M. S., & Postnova, S. (2020). Modelling melanopsin-mediated effects of light on circadian phase, melatonin suppression and subjective sleepiness.



15:16 - 15:34

Integrated analysis of daylight and solar access building requirements and performance in urban environments in Estonia

Francesco De Luca, Abel Sepúlveda

Tallinn University of Technology, Tallinn, Estonia

Aim and Approach

(max 200 words)

Daylight and solar access are essential aspects of the indoor environmental quality of buildings. Adequate quantity of daylight helps to perform tasks with ease and its distribution increases architectural quality [1]. Appropriate direct solar access helps the entrainment of the circadian rhythm and the improvement of physiological and psychological well-being of occupants [2]. Thus, in most countries, regulations prescribe minimum quantities of daylight and direct solar access [3, 4].

The new EU standard Daylight in Buildings [5], to be acquired in Estonia, for daylight requires a minimum Daylight Factor (DF) of 0.7% on 95% of the simulation plane and of 2.2% on 50% of the plane closer the window. For solar access in dwellings it requires a minimum of 1.5 hours of exposure to sunlight calculated during one day between February 1st and March 21st.

The present study investigates optimal dwelling room parameters for 1) the fulfillment of both DF and sunlight requirements of the new EU standard in urban environments, where sunlight provision is most critical, and 2) adequate daylight availability. The aim is to help local authorities in the acquisition of the new EU standard and to provide designers with guidelines for the fulfillment of both requirements.

Scientific Innovation and Relevance

(max 200 words)

The new EU standard requires different daylight but same sunlight quantity for the different countries. There are no studies about the relation between the two requirements in Estonia. Additionally, recent studies proved the scarce reliability of the Daylight Factor metric in predicting daylight availability in the country [6,7].

The innovation of the study lies in the integrated analysis of building performance for daylight and solar access at northern latitudes in urban environments. Additionally, it contributes to the assessment of the reliability of the DF requirements through climate based daylight simulations using the metrics of the LM-83-12 method Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE) [8]. The first predicts reliably daylight availability, the latter potential visual discomfort.

The integrated analysis of DF and solar access, DF and sDA, ASE and solar access permits to assess the efficacy of the requirements of the new EU standard for Estonia and to develop guidelines for the design in urban environments.

For the study a parametric model is realized that generate variations of a sidelit room orientation, window size, shading (upper floor balcony) and surrounding environment. The parametric model automates the calculation of solar access and the different daylight simulations at each variation.

Preliminary Results and Conclusions

(max 200 words)

Simulations for 144 room variations, 8 orientations, 9 window sizes and 2 shading states are performed for March 21st without surrounding buildings and in three urban environments in Tallinn.

Without surrounding buildings, DF, that doesn’t depend on orientation, is fulfilled by 66.7% of 18 variations, solar access, sDA and ASE by 62.5%, 49.3% and 56.9% of 144 variations respectively. DF and sunlight together, and sDA and ASE together, as required by the EU standard and by the LM-83-12 method, are fulfilled by 27.8% and 13.9% of all variations respectively.

In the three urban environments, DF is simulated also for the 8 orientations due to the different external obstructions. On average, solar access, DF, sDA and ASE are fulfilled by 51.8%, 20.3%, 17.8% and 68.5%, DF and sunlight, and sDA and ASE together are fulfilled by 14.4% and 3.7% of all variations respectively.

Preliminary results show the different fulfillment of DF and sunlight, the influence of the urban environment and the difficulty to achieve the two pair of performance together, and the daylight overestimation by the DF requirement in Estonia. The paper will present detailed simulation results and parameters necessary to fulfil the different metrics singularly and in pairs as required.

Main References

(max 200 words)

1 - Reinhart, C.F. 2014. Daylighting Handbook I. Fundamentals. Designing with the Sun. Building Technology Press, Cambridge (USA).

2 – Lockley, S.W. 2009. Circadian rhythms: influence of light in humans. In: Squire LR (ed), Encyclopedia of Neuroscience (Vol. 2), Cambridge, MA, USA: Academic Press, 971–988.

3 – Dogan, T. and Park, Y.C. 2019. A critical review of daylighting metrics for residential architecture and a new metric for cold and temperate climates. Lighting Research & Technology, 51, 206–230.

4 – Darula, S., Christoffersen, J. and Malikova, M. 2015. Sunlight and insolation of building interiors. Energy Procedia, 78, 1245–1250.

5 - European Commission 2018. EN 17037:2018 Daylight in Buildings.

6 - De Luca, F., Kiil, M., Simson, R., Kurnitski, J. and Murula, R. 2019. Evaluating daylight factor standard through climate based daylight simulations and overheating regulations in Estonia. Proceedings of 16th IBPSA International Conference and Exhibition (BS2019), 3968-3975.

7 – Sepúlveda, A., De Luca, F., Thalfeldt, M. and Kurnitski, J. 2020. Analyzing the fulfillment of daylight and overheating requirements in residential and office buildings in Estonia. Building and Environment, 180, 107036.

8 - Illuminating Engineering Society 2013. IES LM-83-12 Approved Method: IES Spatial Daylight Autonomy (sDA) and Annual Sunlight Exposure (ASE).



15:34 - 15:52

Machine learning techniques for the daylight and electric lighting performance predictions

Chantal Basurto, Oliver Paul, Jérôme H. Kämpf

Idiap Research Institute, 1920 Martigny, Switzerland

Aim and Approach

(max 200 words)

Despite the technological advance in the field of energy efficient buildings, the achievement of adequate lighting interior environments it is still a tight corner spot. The latter, due to the complex interplay occurring at different levels of the building performance, involving energy related and occupant’s comfort issues. Such as, achieving a right balance between an increased daylight penetration for the reduction of heating and lighting loads in winter, while minimizing the risk of glare for the occupants. Providing an adequate solar protection while achieving a sufficient daylight provision at task area is a similar quest for summer time. Therefore, in order to undertake the intrinsically linked energy efficiency and occupant’s comfort goals, recent research endeavors involve an integrated assessment of daylight, electric lighting, blinds and lighting controls. Nowadays, such evaluations are mostly performed with the use of computer simulations, which, due to the complexity of the issue, are still highly demanding in terms of computing time and performance capabilities; besides of the human-hours invested on the interaction with distinctive tools and interfaces. In order to improve the response time of daylight and electric lighting performance-predictions, machine learning techniques based on existing daylighting evaluation methods, are employed using surrogate models.

Scientific Innovation and Relevance

(max 200 words)

In order to achieve an optimal control of blinds and electric lighting, a predictor model is employed to evaluate the impact of a blind’s position choice on the work-plane illuminance and of glare in the occupant’s eye. Including the predictor model in a Model Predictive Control (MPC) is the ultimate goal, aiming to obtain a quasi-real-time optimization of the building parameters, to provide visual comfort to the user with less electric lighting. Ubiquity is the main feature of this work, since the predictor model is derived from year-round simulations generated by the RADIANCE based matrix multiplication methods, where all possible blinds positions and weather conditions are considered. Simulation cost is another relevant feature of this work, since, due to the longer time that RADIANCE simulations require to complete, the predictor model is rather based on a statistical surrogate model realized with an Artificial Neural Network (ANN). A database is first produced and employed for the training of the surrogate model. Its input parameters are the weather data (direct and diffuse irradiance), sun and blinds position and electric lighting intensity, while the output data are key performance indicators (average work-plane illuminance and DGP glare index), for specific users.

Preliminary Results and Conclusions

(max 200 words)

The method is applied to an office building located in Martigny, Switzerland, where two specific rooms are modeled and used as a case-study to demonstrate the model’s predictor capabilities. The two models were created using Sketchup, while their material properties (reflectance, transmittance and other physical parameters) were measured using a Minolta Chromameter (CR-200b) and Gloss-meter (GM-060). The models were then calibrated according to on-site illuminance measurements obtained during the summer 2020, where the accuracy was reported as below 20%, providing the target precision for our surrogate model. A nearby weather station is used to gather direct and diffuse irradiance on hourly basis for the whole year under study. The vast amount of data obtained from the year-round RADIANCE simulations was determinant for defining the function between the inputs and outputs. Different ANNs (FFN, LSTM, GRU and CNN) are tested and compared to provide satisfactory precision for both illuminance and DGP, at a negligible simulation cost due to the data obtained beforehand from the RADIANCE simulations. The electric lighting contributions to the illuminance on the work-plane are computed separately, and the potential glare from the luminaires neglected. The developed surrogate model is finally validated against actual RADIANCE simulations and real monitoring.

Main References

(max 200 words)

Ayoub, M., 2020. A review on machine learning algorithms to predict daylighting inside buildings. Solar Energy 202, 249–275. https://doi.org/10.1016/j.solener.2020.03.104

McNeil, A., 2013. The Five-Phase Method for Simulating Complex Fenestration with Radiance (Tutorial). Lawrence Berkeley National Laboratory, Berkeley, CA.

McNeil, A., 2012. A validation of the RADIANCE Three-Phase Simulation Method for Modeling Annual Daylight Performance of Optically Complex Fenestration Systems. Journal of Building Performance Simulation 1–14.

McNeil, A., 2010. The three-phase method for simulating Complex Fenestration with Radiance. Lawrence Berkeley National Laboratory, Berkeley, CA.

Nault, E., Moonen, P., Rey, E., Andersen, M., n.d. Predictive models for assessing the passive solar and daylight potential of neighborhood designs: A comparative proof-of-concept study. Building and Environment 116, 1–16. http://dx-doi.org/10.1016/j.buildenv.2017.01.018

Ward, G., 1985. RADIANCE lighting simulation software.

Wienold, J., Christoffersen, J., 2005. Towards a new daylight glare rating. Presented at the LUX Europa: Lighting for Humans, Berlin, Germany.

Zhaoyang, L., Cheng, S., Qi, D., 2020. A daylight-linked shading strategy for automated blinds based on model-based control and Radial Basis Function (RBF) optimization. Building and Environment 177. https://doi.org/10.1016/j.buildenv.2020.106854



15:52 - 16:10

Luminance distributions in consultancy: Simulations or measurements?

Thijs Willem Kruisselbrink

Peutz BV, Netherlands, The

Aim and Approach

(max 200 words)

The luminance distribution is suitable tool to assess the lit environment in a holisitic manner. The luminance distribution is either simulated or measured. Both approaches have different strenghts and weaknesses. This work presents an analysis of the strenghts and weaknesses of the two approaches in order to find the most suitable approach for specific cases in consultancy.

Scientific Innovation and Relevance

(max 200 words)

The lit environment is an intangible, but rather relevant component of the built environment, impacting performance, comfort, health and well-being. Research has shown that the lit environment has a multidimensional character. Consequently, the lit environment cannot be fully grasped by singular metrics that are often utilized inconsultancy such as the daylight factor (DF). Preferably, multiple metrics, associated to e.g. amount,distribution or directionality of light, are utilized todescribe the lit environment. However, despite significant effort, the research community has not found asatisfactory and holistic metric to capture the lit environment as a whole.

Alternatively, the luminance distribution can be a suitable means to describe the lit environment, as it contains information on the majority of relevant metrics. Nevertheless, practical implementation of the luminance distribution, simulation or measurement, in consultancy is rather limited. In addition to the advantagesof the luminance distribution, multiple limitations areassociated to its use, both for simulations and mea-surements.

Preliminary Results and Conclusions

(max 200 words)

Simulations or measurements of the luminance distributions are suitable for different consultancy cases. Simulations are more suitable to find a optimal solution for the lit environment while measurements are able to assess a problem associated to the lit environment.

Main References

(max 200 words)

Inanici, M. (2006, 6). Evaluation of high dynamicrange photography as a luminance data acquisitionsystem.Lighting Research and Technology 38(2),123–134.

Kruisselbrink, T. (2020, 10).Practical and continu-ous luminance distribution measurements for light-ing quality. Ph. D. thesis, Eindhoven University ofTechnology, Eindhoven.

Kruisselbrink, T., R. Dangol, and A. Rosemann(2018, 6).Photometric measurements of light-ing quality: An overview.Building and Environ-ment 138, 42–52.

Ochoa, C. E., M. B. Aries, and J. L. Hensen (2012, 7).State of the art in lighting simulation for buildingscience: a literature review.Journal of BuildingPerformance Simulation 5(4), 209–233.

Pierson, C., M. Bodart, J. Wienold, and A. Ja-cobs (2017). Luminance maps from High DynamicRange imaging : calibrations and adjustments forvisual comfort assessment. InLux Europa, Ljubl-jana, Slovenia, pp. 147–151.

Reinhard, E., G. Ward, S. Pattanaik, and P. Debevec(2006, 8).High Dynamic Range Imaging: Ac-quisition, Display, and Image-Based Lighting (TheMorgan Kaufmann Series in Computer Graphics).San Fransisco: Morgan Kaufmann Publishers Inc.

Reinhart, C. and O. Walkenhorst (2001). Valida-tion of dynamic RADIANCE-based daylight simu-lations for a test office with external blinds.Energyand Buildings 33(7), 683–697.

Van Den Wymelenberg, K. (2012).Evaluating Hu-man Visual Preference and Performance in an Of-fice Environment Using Luminance-based Metrics.Ph. D. thesis, University of Washington.