Identifying Residential Energy Efficiency Opportunities with Temperature and Energy Use DataEPA Grant Number: F13A20106
Title: Identifying Residential Energy Efficiency Opportunities with Temperature and Energy Use Data
Investigators: Sheikh, Imran Anees
Institution: University of California - Berkeley
EPA Project Officer: Lee, Sonja
Project Period: August 21, 2014 through August 21, 2016
Project Amount: $84,000
RFA: STAR Graduate Fellowships (2013) RFA Text | Recipients Lists
Research Category: Academic Fellowships , Fellowship - Other Engineering
About 40 percent of energy used in the United States is used in buildings, and a wide body of evidence suggests there are both large opportunities and significant barriers to making buildings more energy efficient. Combining data from smart meters, weather stations, and internet-connected thermostats will allow a better understanding of how energy is used in buildings. Using these emerging data sources to develop and apply algorithms can give insight into the likely efficiency opportunities that exist for specific customers.
Energy use in homes will be characterized by analyzing patterns of temperature and energy use and creating regression-based metrics that relate to how buildings use energy. These metrics will, in theory, relate to physical properties of buildings such as insulation, solar gain, or air conditioner efficiency. Looking at distributions of these metrics for buildings of similar size and in a similar climate should make it possible to identify particularly inefficient buildings. An understanding of the distribution of these metrics will allow quantification of the savings potential that would result from improving the efficiency of buildings in the tail of the distribution.
The goals of this research are to develop methods that make it possible to use emerging data sources to identify residences with a high potential to save energy, quantify the expected savings and assess the types of energy efficiency projects that are most likely needed. With these aggressive goals in mind, it is important to recognize that a wide variety of factors (that are not directly observed) contribute to energy use, including behavior, appliance types, design and building systems. Using only smart meter and temperature data it will be impossible to give perfectly accurate, specific recommendations of actions that should be taken. However, the results of this research will allow utilities and their partners to begin to create more effective energy efficiency programs by identifying and engaging those consumers where large energy savings are likely. Analysis of these emerging data sources at scale could create a foundation for new residential efficiency business models that remove some of the barriers to investment in energy efficiency projects.
Potential to Further Environmental/Human Health Protection
This research could lead to improved energy efficiency at scale in the residential sector and therefore decrease demand for electricity and the associated emissions that come from electricity generation. Combustion for power generation leads to emissions of a wide range of pollutants that cause health problems, and since major emission sources from electricity generation are often in economically distressed areas, reducing electricity generation through improved efficiency is particularly important to those communities.