Energy atlas : machine-learning-based mapping and analysis for sustainable energy and urban systems
Abstract/Contents
- Abstract
- Climate change poses increasingly severe threats to the sustainable development of the world. However, vulnerabilities of both human population and the infrastructure they rely on, as well as the abilities to mitigate and adapt to climate change, vary drastically across countries, regions, cities, and even neighborhoods. Their geospatial distributions and temporal variations are critical information for decision making to support sustainability, equity, and climate resilience but are largely unavailable today. In this dissertation, I addressed this gap by developing a set of machine-learning-based approaches towards the construction of "Energy Atlas"—a comprehensive and dynamic geospatial map overlay for distributed renewable energy, infrastructures, and population characteristics. Such approaches to collect granular spatial and temporal information are scalable, non-intrusive, and reliant only on widely-available geospatial data (e.g., remote sensing images, street views) as inputs. In particular, they enable the construction of a nationwide spatiotemporal solar PV installation dataset covering the contiguous US, the fine-grained distribution grid mapping in both the US and Africa, as well as the learning of urban neighborhood characteristics from imagery and textual data. These machine-learning-based approaches drive a closed-loop solution towards customized and evidence-informed decision making for sustainability, resilience, and equity. It begins from observing how our world looks like, to understanding how factors are correlated, and finally to informing what can be done. Specifically, based on "Energy Atlas", I identified certain types of solar energy incentives that are particularly relevant to solar adoption in low-income communities—from both correlational and causal perspectives. Based on "Energy Atlas", I also proposed a cost allocation scheme to make burying wildfire-prone power lines—an effective but costly wildfire mitigation approach—equitably affordable to communities at all income levels. The new methods, new data, and new insights gained in this dissertation can enable other researchers, policymakers, and various stakeholders to further develop econometric and engineering models, conduct in-depth analysis, and inform decision making for developing sustainable, equitable, and climate-resilient energy systems, infrastructures, and communities.
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 | 2022; ©2022 |
| Publication date | 2022; 2022 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Wang, Zhecheng |
|---|---|
| Degree supervisor | Majumdar, Arunava |
| Degree supervisor | Rajagopal, Ram |
| Thesis advisor | Majumdar, Arunava |
| Thesis advisor | Rajagopal, Ram |
| Thesis advisor | Kiremidjian, Anne S. (Anne Setian) |
| Degree committee member | Kiremidjian, Anne S. (Anne Setian) |
| Associated with | Stanford University, Civil & Environmental Engineering Department |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Zhecheng Wang. |
|---|---|
| Note | Submitted to the Civil & Environmental Engineering Department. |
| Thesis | Thesis Ph.D. Stanford University 2022. |
| Location | https://purl.stanford.edu/gb334vm4252 |
Access conditions
- Copyright
- © 2022 by Zhecheng Wang
- 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...