A data-driven toolkit for rural residential energy efficiency evaluations

Abstract

Residential buildings have historically consumed a significant portion of energy and electricity in the U.S. Thus, to meet residential energy consumption and greenhouse gas emission reduction goals, the current state of the residential building stock’s energy consumption and the impact of various energy efficiency efforts on its performance, both now and in the future, need to be quantified and understood. Rural housing, or housing that lies outside metropolitan areas, is a specifically less studied section of the U.S. housing stock and, due to its unique characteristics, brings new challenges to research and implementation of energy efficiency efforts. The residential housing stock in rural areas of the U.S. is unique in that it is generally more likely to be older and less energy-efficient than those homes built in more urban areas. Moreover, rural household are often disproportionately burdened with energy costs. Thus, in an effort to improve the energy efficiency of such residential buildings, with benefits including an overall reduction in emissions resulting from building operations, and reduction in costs to the building occupants, this study aims to understand the energy performance characteristics of residential buildings in rural areas. The overarching goal of this work is hence to help better define which buildings are in greatest need of energy efficiency retrofits and help to establish a framework by which to target such buildings for performance improvements through efforts such as utility-supported rebate programs. In this research, through collaboration with the communities of Bloomfield, and Cedar Falls, Iowa, multi-year periods of building energy use metered data and building characteristic data (from energy audits and tax assessors data) is used to assess the energy performance of residential buildings in these rural areas. Using hierarchical time-series clustering, high intensity energy users in each community are identified amongst homes with similar sizes and subsequently characterized. In Bloomfield (N=320), Homes classified as high intensity electricity (N=17) and natural gas users (N=23), were on average 44% and 30% more energy intensive, respectively, when compared to their counterpart average intensity consumers. The high intensity electricity (N=100) and natural gas users (N=42) in Cedar Falls (N=661), were on average 36% and 104% more energy intensive than those homes classified as average intensity users for each energy type. Then, the performance of the proposed clustering framework in terms of identifying the appropriate households is evaluated and compared against commonly used energy efficiency metrics and program qualification guidelines. The findings of these efforts support that the proposed rural energy assessment framework can be used to design policies and programs that benefit both individual households most in need, while also addressing absolute levels of consumption in a community now and moving forward. Since the energy performance of residential buildings is closely correlated with occupants’ behavior and their schedules, accurate occupancy presence and activity profiles are also important in evaluating and predicting energy performance of residential buildings. Therefore, another set of efforts in this study are focused on identifying household and housing characteristics that affect occupancy schedules based on 10 years of American Time-Use Survey (ATUS) data (2010-2019). Then, the most significant identified variables, household size and weekday type, are used to create representative schedules for unique user types using a first-order inhomogeneous Markov chain technique. The generated stochastic occupancy schedules are beneficial for future building performance modelling or analysis efforts in both rural and metropolitan areas. One such application, is to quantify the impact of changes in occupancy on energy consumption in rural residential buildings, which is the focus of the latter part of this study. Electricity consumption data from the 2020 COVID-19 pandemic period is compared against a baseline period (2010 to 2016) to evaluate the impact of unprecedented changes in occupancy on energy consumption in rural residential buildings. These comparisons suggested that 74% of single-family owner-occupied detached homes in Cedar Falls, Iowa, witnessed an increase in their electricity consumption in 2020, resulting in a 39% increase in electricity consumption community-wide. Such large-scale changes can have profound impacts on the grid as well as the households’ ability to pay for utility bills and thus, need to be carefully monitored moving forward. This is especially important as some of the behavioral changes from the COVID-19 lockdown, such as remote working, continue to be adopted by many households in the post-pandemic era.