![](https://crypto4nerd.com/wp-content/uploads/2023/06/1OredqNX3-7JbH6F4wlhecg-1024x576.jpeg)
Since I joined the Technical University of Denmark, I’ve seen myself driven by new technologies. How to use them to solve sustainability issues the world is facing. Additionally, being part of the sustainable generation, those demonstrated on Friday thriving for bringing up the green revolution. Well, that period ended: coronavirus outbreaks had come up. Nonetheless, the Will is still growing up in me. Denmark was the perfect place to emancipate in both technology and sustainability.
Now I am wrapping up two years of my master’s with a thesis aiming to reduce, or at least assess, heat stress in southern African dwellings using Artificial Intelligence. Particularly the use of Machine Learning supervised and unsupervised learning to cover the lack of building monitoring in low-income buildings.
In Europe, in Denmark, building standards and legislation assure good confection quality. In this context, building energy simulations (BES) seems relevant to predict indoor environment quality (IEQ), such as thermal comfort and air quality. However, in Africa, in-place regulations are less detailed than in Europe. It results in poorer building confection. However, advanced tools cannot assess the risks associated with these poorly constructed buildings, and research needs to be done to help monitor and assess IEQ and settle Early Warning Systems (EWS).
Africa is projected to grow up considerably their population within the next 50 years. Current humanitarian conditions do not presage affordable and adequate IEQ for inhabitants. Nevertheless, African populations are also subjected to significant environmental changes due to climate change that predict an increase in ambient temperature. The combination of the facts above may lead to a major humanitarian crisis in the future. Therefore, it urges us to tackle this problem with academic research on IEQ in poorly constructed buildings, especially in Africa.
My master’s thesis aims to tackle this problem using state-of-the-art IEQ and Machine Learning research.
I am working on dwellings in rural areas in the Province of Limpopo and Johannesburg in South Africa.
The study aims to highlight the potential of using ML for leveraging uncertainties. More is coming during the upcoming months with some results and code.
Sources
Kapwata, T., Gebreslasie, M. T., Mathee, A., & Wright, C. Y. (2018). Current and potential future seasonal trends of indoor dwelling temperature and likely health risks in rural Southern Africa. International Journal of Environmental Research and Public Health, 15(5). https://doi.org/10.3390/IJERPH15050952
Naicker, N., Teare, J., Balakrishna, Y., Wright, C. Y., & Mathee, A. (2017). Indoor temperatures in low cost housing in Johannesburg, South Africa. International Journal of Environmental Research and Public Health, 14(11). https://doi.org/10.3390/ijerph14111410
Matthews, T. K. R., Wilby, R. L., & Murphy, C. (2017). Communicating the deadly consequences of global warming for human heat stress. Proceedings of the National Academy of Sciences of the United States of America, 114(15), 3861–3866. https://doi.org/10.1073/PNAS.1617526114
Ziervogel, G., Lennard, C., Midgley, G., New, M., Simpson, N. P., Trisos, C. H., & Zvobgo, L. (2022). Climate change in South Africa: Risks and opportunities for climate-resilient development in the IPCC Sixth Assessment WGII Report. South African Journal of Science, 118(9–10). https://doi.org/10.17159/sajs.2022/14492