Date of Award
Doctor of Philosophy (PhD)
Engineering and Public Policy
Inês M.L. Azevedo
This dissertation is comprised of four studies that examine issues where submetered device-level energy use data can be used to inform energy efficiency policy and investment decision making in residential buildings. In addition to identifying applications and developing the methods for incorporating these data in engineering and economic analyses, the nontechnical aspects of these issues are also considered as implementation of these solutions depends on more than their technical feasibility. The first study frames the existing methods being used today to build a device-level accounting of where energy is used in the residential sector in the US. Three methods are identified and categorized: direct metering methods, non-intrusive load monitoring, and statistical methods. Two of the most prominent studies are compared, and a method is proposed by which the Department of Energy (DOE) and others could easily and cost-effectively incorporate existing submetered data into their estimates of device-level energy consumption in the US. In the second study, energy audit and survey records are used to model 106 homes in the DOE’s EnergyPlus building simulation software. Simulation results are compared to submetered data from the audited homes to provide a device-level measure of the accuracy of EnergyPlus models in a large sample of homes. Results show the models do not accurately or consistently estimate occupied home energy use due to factors such as occupant behaviors and appliance stocks which are not well-captured in traditional audit reports. These results provide context for the growing use of EnergyPlus models for homes, and highlight that care is needed to ensure that decision-makers are aware of its limitations. The third study uses device-level data to assess the technical and economic feasibility of distributing direct current (DC) power in homes with solar photovoltaic (PV) arrays. Monte Carlo simulation is used to estimate the costs and benefits of this intervention while accounting for uncertainty in the engineering, economic, and other parameters. Results show significant energy savings potential, but at present DC equipment prices these savings are not cost-effective. In addition to quantifying energy and cost savings, a number of major nontechnical barriers to implementing DC distribution in homes are identified. In the final study, an expert elicitation is conducted with 17 experts on DC systems to better understand these barriers. Results show that the two biggest barriers to adoption are industry professionals unfamiliar with DC and small markets for DC devices and components. To address these, experts proposed developing training programs for engineers and electricians, and developing pilot projects to prove the benefits of DC in niche applications where DC power distribution holds a clear advantage over alternating current (AC). Experts also identified lasting and inherent benefits of DC that make these systems better suited to serve future building loads.
Glasgo, Brock, "Device-Level Data Analytics to Guide Policy" (2017). Dissertations. 896.