Deep-Learning Framework for Estimating Behind the Meter Solar Generation and Electric Vehicle Penetration Level and Time-of-Use
| dc.contributor.author | Abdalla, Mohamed | |
| dc.date.accessioned | 2021-01-20T13:30:57Z | |
| dc.date.available | 2023-01-21T05:50:04Z | |
| dc.date.issued | 2021-01-20 | |
| dc.date.submitted | 2021-01-12 | |
| dc.description.abstract | The continual increase in the adoption of rooftop solar/photovoltaic (PV) generation and electric vehicles (EVs) presents challenges, as well as opportunities, in distribution power systems. Without monitoring or control, the addition of PV generation and EV charging to distribution power systems can result in power stability, as well as power congestion issues. In this research, a deep-learning framework is presented in order to monitor and estimate the penetration level of PV generation and EV charging in distribution power systems. The proposed framework is also developed to predict the time-of-use of EV charging in order to enable scheduling for demand response programs. Additionally, the framework presented in this research is capable of estimating the generated solar power behind the meter for improving distribution system operational planning as well as power procurement plans. The framework identifies the houses that include PVs or EVs and monitor their behind the meter solar generation as well as the time-of-use of EVs, through the use of only existing smart meter data, and it can also be scaled to include other flexible appliances of interest. In order to improve the overall performance of the inference system and mitigate error propagation, the framework exploits various customized sub-models that are specifically built for each sub-target. In this research, the framework was evaluated using real smart meter data from Pecan Street Dataport and achieved a promising 93-98% F-score across all its sub-models, which proves the feasibility and scalability of our approach. | en |
| dc.identifier.uri | http://hdl.handle.net/10012/16698 | |
| dc.language.iso | en | en |
| dc.pending | false | |
| dc.publisher | University of Waterloo | en |
| dc.relation.uri | Pecan Street | en |
| dc.subject | deep-learning | en |
| dc.subject | machine-learning | en |
| dc.subject | distribution power system | en |
| dc.subject | time-of-use | en |
| dc.subject | load disaggregation | en |
| dc.subject | data-driven | en |
| dc.subject | electric vehicle | en |
| dc.subject | solar generation | en |
| dc.title | Deep-Learning Framework for Estimating Behind the Meter Solar Generation and Electric Vehicle Penetration Level and Time-of-Use | en |
| dc.type | Master Thesis | en |
| uws-etd.degree | Master of Applied Science | en |
| uws-etd.degree.department | Electrical and Computer Engineering | en |
| uws-etd.degree.discipline | Electrical and Computer Engineering | en |
| uws-etd.degree.grantor | University of Waterloo | en |
| uws-etd.embargo.terms | 2 years | en |
| uws.contributor.advisor | El-Shatshat, Ramadan | |
| uws.contributor.affiliation1 | Faculty of Engineering | en |
| uws.peerReviewStatus | Unreviewed | en |
| uws.published.city | Waterloo | en |
| uws.published.country | Canada | en |
| uws.published.province | Ontario | en |
| uws.scholarLevel | Graduate | en |
| uws.typeOfResource | Text | en |