Matching Items (4)
Filtering by
- Genre: Masters Thesis
Description
Machine learning (ML) and deep learning (DL) has become an intrinsic part of multiple fields. The ability to solve complex problems makes machine learning a panacea. In the last few years, there has been an explosion of data generation, which has greatly improvised machine learning models. But this comes with a cost of high computation, which invariably increases power usage and cost of the hardware. In this thesis we explore applications of ML techniques, applied to two completely different fields - arts, media and theater and urban climate research using low-cost and low-powered edge devices. The multi-modal chatbot uses different machine learning techniques: natural language processing (NLP) and computer vision (CV) to understand inputs of the user and accordingly perform in the play and interact with the audience. This system is also equipped with other interactive hardware setups like movable LED systems, together they provide an experiential theatrical play tailored to each user. I will discuss how I used edge devices to achieve this AI system which has created a new genre in theatrical play. I will then discuss MaRTiny, which is an AI-based bio-meteorological system that calculates mean radiant temperature (MRT), which is an important parameter for urban climate research. It is also equipped with a vision system that performs different machine learning tasks like pedestrian and shade detection. The entire system costs around $200 which can potentially replace the existing setup worth $20,000. I will further discuss how I overcame the inaccuracies in MRT value caused by the system, using machine learning methods. These projects although belonging to two very different fields, are implemented using edge devices and use similar ML techniques. In this thesis I will detail out different techniques that are shared between these two projects and how they can be used in several other applications using edge devices.
ContributorsKulkarni, Karthik Kashinath (Author) / Jayasuriya, Suren (Thesis advisor) / Middel, Ariane (Thesis advisor) / Yu, Hongbin (Committee member) / Arizona State University (Publisher)
Created2021
Description
The measurement of the radiation and convection that the human body experiences are important for ensuring safety in extreme heat conditions. The radiation from the surroundings on the human body is most often measured using globe or cylindrical radiometers. The large errors stemming from differences in internal and exterior temperatures and indirect estimation of convection can be resolved by simultaneously using three cylindrical radiometers (1 cm diameter, 9 cm height) with varying surface properties and internal heating. With three surface balances, the three unknowns (heat transfer coefficient, shortwave, and longwave radiation) can be solved for directly. As compared to integral radiation measurement technique, however, the bottom mounting using a wooden-dowel of the three-cylinder radiometers resulted in underestimated the total absorbed radiation. This first part of this thesis focuses on reducing the size of the three-cylinder radiometers and an alternative mounting that resolves the prior issues. In particular, the heat transfer coefficient in laminar wind tunnel with wind speed of 0.25 to 5 m/s is measured for six polished, heated cylinders with diameter of 1 cm and height of 1.5 to 9 cm mounted using a wooden dowel. For cylinders with height of 6 cm and above, the heat transfer coefficients are independent of the height and agree with the Hilpert correlation for infinitely long cylinder. Subsequently, a side-mounting for heated 6 cm tall cylinder with top and bottom metallic caps is developed and tested within the wind tunnel. The heat transfer coefficient is shown to be independent of the flow-side mounting and in agreement with the Hilpert correlation.
The second part of this thesis explores feasibility of employing the three-cylinder concept to measuring all air-flow parameters relevant to human convection including mean wind speed, turbulence intensity and length scale. Heated cylinders with same surface properties but varying diameters are fabricated. Uniformity of their exterior temperature, which is fundamental to the three-cylinder anemometer concept, is tested during operation using infrared camera. To provide a lab-based method to measure convection from the cylinders in turbulent flow, several designs of turbulence-generating fractal grids are laser-cut and introduced into the wind tunnel.
ContributorsGupta, Mahima (Author) / Rykaczewski, Konrad (Thesis advisor) / Pathikonda, Gokul (Thesis advisor) / Middel, Ariane (Committee member) / Arizona State University (Publisher)
Created2024
Description
During summer 2014, a study was conducted as part of the Landscape Architecture Foundation Case Study Investigation to analyze features of three sustainably designed landscapes. Each project was located in a southwest desert city: Civic Space Park in Phoenix, AZ, the Pete V. Domenici US Courthouse Sustainable Landscape Retrofit in Albuquerque, NM, and George "Doc" Cavalliere Park in Scottsdale, AZ. The principal components of each case study were performance benefits that quantified ongoing ecosystem services. Performance benefits were developed from data provided by the designers and collected by the research team. The functionality of environmental, social, and economic sustainable features was evaluated. In southwest desert cities achieving performance benefits such as microclimate cooling often come at the cost of water conservation. In each of these projects such tradeoffs were balanced by prioritizing the project goals and constraints.
During summer 2015, a study was conducted to characterize effects of tree species and shade structures on outdoor human thermal comfort under hot, arid conditions. Motivating the research was the hypothesis that tree species and shade structures will vary in their capacity to improve thermal comfort due to their respective abilities to attenuate solar radiation. Micrometeorological data was collected in full sun and under shade of six landscape tree species and park ramadas in Phoenix, AZ during pre-monsoon summer afternoons. The six landscape tree species included: Arizona ash (Fraxinus velutina Torr.), Mexican palo verde (Parkinsonia aculeata L.), Aleppo pine (Pinus halepensis Mill.), South American mesquite (Prosopis spp. L.), Texas live oak (Quercus virginiana for. fusiformis Mill.), and Chinese elm (Ulmus parvifolia Jacq.). Results showed that the tree species and ramadas were not similarly effective at improving thermal comfort, represented by physiologically equivalent temperature (PET). The difference between PET in full sun and under shade was greater under Fraxinus and Quercus than under Parkinsonia, Prosopis, and ramadas by 2.9-4.3 °C. Radiation was a significant driver of PET (p<0.0001, R2=0.69) and with the exception of ramadas, lower radiation corresponded with lower PET. Variations observed in this study suggest selecting trees or structures that attenuate the most solar radiation is a potential strategy for optimizing PET.
During summer 2015, a study was conducted to characterize effects of tree species and shade structures on outdoor human thermal comfort under hot, arid conditions. Motivating the research was the hypothesis that tree species and shade structures will vary in their capacity to improve thermal comfort due to their respective abilities to attenuate solar radiation. Micrometeorological data was collected in full sun and under shade of six landscape tree species and park ramadas in Phoenix, AZ during pre-monsoon summer afternoons. The six landscape tree species included: Arizona ash (Fraxinus velutina Torr.), Mexican palo verde (Parkinsonia aculeata L.), Aleppo pine (Pinus halepensis Mill.), South American mesquite (Prosopis spp. L.), Texas live oak (Quercus virginiana for. fusiformis Mill.), and Chinese elm (Ulmus parvifolia Jacq.). Results showed that the tree species and ramadas were not similarly effective at improving thermal comfort, represented by physiologically equivalent temperature (PET). The difference between PET in full sun and under shade was greater under Fraxinus and Quercus than under Parkinsonia, Prosopis, and ramadas by 2.9-4.3 °C. Radiation was a significant driver of PET (p<0.0001, R2=0.69) and with the exception of ramadas, lower radiation corresponded with lower PET. Variations observed in this study suggest selecting trees or structures that attenuate the most solar radiation is a potential strategy for optimizing PET.
ContributorsColter, Kaylee (Author) / Martin, Chris (Thesis advisor) / Coseo, Paul (Committee member) / Middel, Ariane (Committee member) / Arizona State University (Publisher)
Created2016
Description
The rapid rate of urbanization coupled with continued population growth and anthropogenic activities has resulted in a myriad of urban climate related impacts across different cities around the world. Hot-arid cities are more vulnerable to induced urban heat effects due to the intense solar radiation during most of the year, leading to increased ambient air temperature and outdoor/indoor discomfort in Phoenix, Arizona. With the fast growth of the capital city of Arizona, the automobile-dependent planning of the city contributed negatively to the outdoor thermal comfort and to the people's daily social lives. One of the biggest challenges for hot-arid cities is to mitigate against the induced urban heat increase and improve the outdoor thermal. The objective of this study is to propose a pragmatic and useful framework that would improve the outdoor thermal comfort, by being able to evaluate and select minimally invasive urban heat mitigation strategies that could be applied to the existing urban settings in the hot-arid area of Phoenix. The study started with an evaluation of existing microclimate conditions by means of multiple field observations cross a North-South oriented urban block of buildings within Arizona State University’s Downtown campus in Phoenix. The collected data was evaluated and analyzed for a better understanding of the different local climates within the study area, then used to evaluate and partially validate a computational fluid dynamics model, ENVI-Met. Furthermore, three mitigation strategies were analyzed to the Urban Canopy Layer (UCL) level, an increase in the fraction of permeable materials in the ground surface, adding different configurations of high/low Leaf Area Density (LAD) trees, and replacing the trees configurations with fabric shading. All the strategies were compared and analyzed to determine the most impactful and effective mitigation strategies. The evaluated strategies have shown a substantial cooling effect from the High LAD trees scenarios. Also, the fabric shading strategies have shown a higher cooling effect than the Low LAD trees. Integrating the trees scenarios with the fabric shading had close cooling effect results in the High LAD trees scenarios. Finally, how to integrate these successful strategies into practical situations was addressed.
ContributorsAldakheelallah, Abdullah (Author) / Reddy, T Agami (Thesis advisor) / Middel, Ariane (Committee member) / Coseo, Paul (Committee member) / Arizona State University (Publisher)
Created2020