Thesis
Deep Learning Based Solar Radiation Prediction Models for the Pacific Northwest
Washington State University
Master of Science (MS), Washington State University
2023
DOI:
https://doi.org/10.7273/000005043
Abstract
Over the past few decades, as a consequence of the depleting fossil fuels and increasing environmental pollution, renewable energy resources have acquired remarkable attention. Though renewable energy resources are available in various forms, solar energy undoubtedly tops the list in terms of convenience, reliability, and recently, affordability. As of 2023, the Guardian, a world news publication, reports that it is now cheaper in 99% of the United States to build new solar or wind generation capability than to continue to run existing coal-fired power plants. However, whereas the US has the domestic capability to produce significant amounts of coal and natural gas, and thus provide generation capacity that meets consumer demand, solar and wind generation capacity is subject to ever-changing weather patterns. Thus, introducing these renewable resources adds the complexity of determining for any given period of time, how much generation capacity is available, and whether that capacity will meet demand so that additional generation resources (coal, natural gas, hydro-energy) can be brought online. In our work, we have developed a number of deep-learning models to predict solar generation capacity in the near future term, so that appropriate actions can be taken well ahead of time to balance this power generation and use in real time. These models are targeted to the Pacific Northwest region of the United States of America, where the availability of sunlight is relatively unpredictable due to year-round rainfall and frequent wildfires. The proposed models, especially the seasonal and the monthly model, show a significant improvement in short-term solar radiation forecasting over other related works, and existing standard machine learning and recurrent neural network models, thus proving their efficacy.
Metrics
1 File views/ downloads
79 Record Views
Details
- Title
- Deep Learning Based Solar Radiation Prediction Models for the Pacific Northwest
- Creators
- Pinaki Prasad Guha Neogi
- Contributors
- Scott Wallace (Advisor)Xinghui Zhao (Committee Member)Anna Wisniewska (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- School of Engineering and Computer Science (VANC)
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Number of pages
- 103
- Identifiers
- 99901019939001842
- Language
- English
- Resource Type
- Thesis