Date of Award

7-7-2020

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Civil and Environmental Engineering

First Advisor

Rupp Carriveau

Second Advisor

Lindsay Miller

Keywords

Electricity Demand, Energy Modeling, Energy Systems Transition, Load Curves, Systems Planning

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Electricity systems are undergoing significant changes. Demands are shifting in magnitude and temporal distribution due to developing policies and technologies such as electric vehicles, heat pumps, embedded generation and energy storage, while an increasingly renewable supply is intermittent and less flexible. As such, there is currently great uncertainty in the industry and future business pathways may vary significantly from the current paradigm. This research focused on developing a set of models which can be used by utility companies to leverage their smart meter data and gain insights into possible future impacts and opportunities. The thesis presents a series of novel models, developed and implemented with data provided from a utility in Southern Ontario. First, a regression model was developed to leverage the full value of utility smart meter data by disaggregating residential and commercial sector demands into base, heating and cooling end uses. The use of a variable temperature changepoint only marginally improved prediction accuracy, but significantly shifted disaggregation results, particularly at hourly resolution. This model was also applied for weather normalization, assessment of technology change and projection under different climate scenarios. A second model used this and additional data from literature to project long term utility level average and peak seasonal load curves. A dynamic interface with parameterized controls allowed real-time visualization of technology and policy impacts on the demand curve. A set of eight literature-based scenarios were also projected to demonstrate the extreme range of impacts predicted by different literature. These led to the conclusion that unmanaged technology penetration can lead to significant challenges such as increased peaks, large ramp rates and lower utilization. An analysis was then performed at finer geographic resolution, investigating impacts on representative distribution system transformers. First, the current variation in local technology penetration was examined, showing a significantly skewed distribution with many transformers having up to ten times the average rates. Clustering was then used to identify a set of eight diverse, representative transformer load profiles. Future scenarios were modeled, demonstrating that the impacts of technology and optimal mitigation techniques vary significantly between regions of the distribution system. Finally, the dynamic utility load curve model was also updated to project demands for the representative transformer groups identified. This allows users to simultaneously assess local impacts and mitigation strategies, as well as aggregate effects on the overall system demands. Together these works combine to provide a valuable toolset and significant insight into potential system impacts.

Share

COinS