spatial relationships solar radiation prediction deep neural networks Clouds
We explore the impact of using weather data from multiple locations for predicting one hour ahead solar radiation. We begin with an examination of the intuitive scenario in which transient cloud formations may migrate from one area to another, effecting solar radiation as the cloud cover moves from station to station. Our dataset, obtained from multiple weather stations across eastern Washington state, shows some support for this scenario. Given these initial findings, we proceed to examine whether solar radiation prediction can be improved by combining information across multiple weather stations. Our results, using deep neural networks, show modest improvement in hour ahead prediction when multiple stations are used as input.
Metrics
64 File views/ downloads
10 Record Views
Details
Title
SPATIAL RELATIONSHIPS FOR SOLAR RADIATION PREDICTION
Creators
Farzanuddin Syed
Contributors
Scott Wallace (Chair)
Xinghui zhao (Committee Member)
Ben McCamish (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