From all angles : new perspectives in urban energy modeling
Abstract/Contents
- Abstract
- As we venture deeper into the 21st century, the conversation around urban sustainability gains unparalleled urgency. Cities, long considered the cradles of innovation and economic growth, also bear an undeniable environmental cost, contributing significantly to resource depletion and climate perturbations. In this vein, buildings emerge as central actors, accounting for a substantial fraction of urban energy consumption. The complexities governing building energy use are multifaceted, involving variables from design elements to microclimate influences. \textbf{The primary goal of this work is to deepen our comprehension of urban energy ecosystems by leveraging advanced computational methods and diverse data sources, thus addressing limitations in the current paradigms of Urban Building Energy Modeling (UBEM).} The first approach outlined in this thesis utilizes high resolution satellite imagery in an exploration of its potential benefit to energy modeling. I introduce a novel data processing pipeline in a study of New York City, which synthesizes contextual urban information with satellite imagery. By using model algorithms of computer vision, I am able to distill information from the high resolutions such that significant contextual features may be extracted. I find that high resolution satellite imagery seems to be incredibly promising, with nearly the same efficacy in energy prediction as a model generated from a curated data set of valuable building features. In this same work, the potential pitfalls of pure computer vision are explored, which lay the groundwork for subsequent sections. This thesis explores the applications of new data sources in urban energy modeling to assess their benefits and spatial challenges with their implementation. One such example is climate reanalysis data, which provides a composition of high resolution climate modeling and measured historical data. This data can offer valuable insights into the climate-related factors affecting building energy consumption, such as temperature, humidity, and solar radiation. The incorporation of climate reanalysis into energy modeling may enrich the granularity and accuracy of building energy models, allowing them to better capture the complex interplay of factors affecting building energy consumption. Despite their relevance, advanced climate models are among one of many data sources which have often been overlooked in mainstream energy modeling and their relative benefits to the modeling community are largely unknown. At least partially to blame for their lack of inclusion in modern analysis are technical challenges associated with assimilating these vast and varied data streams into existing workflows. Here, machine learning with its capacity to handle high-dimensional data and model non-linear relationships, presents itself as a promising tool. This thesis complements the incorporation of new data streams by introducing novel machine learning techniques to distill information from these data sources and describe their likely sources of benefit. As we delve into this intricate terrain of data and algorithms, we hope to extend the boundaries of our current knowledge and provide meaningful contributions to the field of urban energy modeling. The work in this dissertation represents a step forward in this ongoing journey to enhance our understanding of urban energy dynamics and inform efforts towards urban sustainability.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2023; ©2023 |
| Publication date | 2023; 2023 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Dougherty, Thomas Ryun |
|---|---|
| Degree supervisor | Jain, Rishee |
| Thesis advisor | Jain, Rishee |
| Thesis advisor | Armeni, Iro |
| Thesis advisor | Rajagopal, Ram |
| Degree committee member | Armeni, Iro |
| Degree committee member | Rajagopal, Ram |
| Associated with | Stanford University, School of Engineering |
| Associated with | Stanford University, Civil & Environmental Engineering Department |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Thomas Ryun Dougherty. |
|---|---|
| Note | Submitted to the Civil & Environmental Engineering Department. |
| Thesis | Thesis Ph.D. Stanford University 2023. |
| Location | https://purl.stanford.edu/ch520pd8064 |
Access conditions
- Copyright
- © 2023 by Thomas Ryun Dougherty
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...