Name
Abstract | ||
|---|---|---|
The popularity of rooftop solar for individual homes continues to rise rapidly. However, techniques for accurately forecasting solar generation are critical to fully exploiting the benefits of such locally-generated solar energy. In this paper, we present SolarCast, a cloud-based web service, which automatically generates models that provide customized site-specific predictions of future solar generation. SolarCast utilizes a \"black box\" approach that requires only i) a site's geographic location and ii) a minimal amount of historical generation data. Since we intend SolarCast for small rooftop deployments, it does not require detailed site- and panel-specific information, which owners may not know, but instead automatically learns these parameters for each site. We evaluate SolarCast's accuracy on a dataset consisting of 118 geographically-diverse solar deployments, and show that it learns an accurate model using much less data (~1 month) than a prior SVM-based approach, which requires ~3 months of data. SolarCast also provides a programmatic API, enabling developers to integrate its predictions directly into energy-efficiency applications. We present a case study of using SolarCast to implement one such application: a \"sunny\" load scheduler, which schedules a dryer's energy usage to maximally align with a home's solar generation. Our results indicate that a representative home is capable of reducing its grid demand up to 40% by providing a modest amount of flexibility (of ~5 hours) in the dryer's start time. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1145/2674061.2674071 | BuildSys@SenSys |
Keywords | Field | DocType |
design,experimentation,energy,consumer products,grid,measurement,electricity | Black box (phreaking),Electricity,Support vector machine,Solar energy,Real-time computing,Schedule,Engineering,Web service,Grid,Cloud computing | Conference |
Citations | PageRank | References |
10 | 1.38 | 5 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Srinivasan Iyengar | 1 | 19 | 4.89 |
| Navin Sharma | 2 | 214 | 15.64 |
| David E. Irwin | 3 | 899 | 98.12 |
| Prashant J. Shenoy | 4 | 6386 | 521.30 |
| Krithi Ramamritham | 5 | 4975 | 936.38 |