Monday Study – The Real Electricity Use During The Pandemic
Multiscale effects masked the impact of the COVID-19 pandemic on electricity demand in the United States
Casey D.Burleyson, AowabinRahman, Jennie S.Rice, Amanda D.Smith, NathalieVoisin, December 2021 (Science Direct)
1-COVID-19 modified when, where, and how we use electricity.
2-The impact of COVID-19 varied across spatiotemporal scales.
3-There are multiple offsetting effects that can mask the impact of COVID-19 on electricity demand.
Shelter-in-place orders and business closures related to COVID-19 changed the hourly profile of electricity demand and created an unprecedented source of uncertainty for the grid. The potential for continued shifts in electricity profiles has implications for electricity sector investment and operating decisions that maintain reserve margins and provide grid reliability. This study reveals that understanding this uncertainty requires an understanding of the underlying drivers at the customer-class scale. This paper utilizes three datasets to compare the impacts of COVID-19 on electricity consumption across a range of spatiotemporal and customer scales. At the utility/customer-class scale, COVID-19-induced shutdowns in the spring of 2020 shifted weekday residential load profiles to resemble weekend profiles from previous years. Total commercial loads declined, but the commercial diurnal load profile was unchanged. With only total loads available at the balancing authority scale, the apparent impact of COVID-19 was smaller during the summer due in part to phased re-opening and spatial variability in re-opening, but there were still clear variations once total loads were broken down zonally. Monthly data at the state scale showed an increase in state-level residential electricity sales, a decrease in commercial sales, and a small net decrease in total sales in most states from April-August 2020. Analyses that focus on total load or a single scale may miss important changes that become apparent when the load is broken down regionally or by customer class.
Understanding and predicting the diurnal profile of electricity demand (also called load) is critical for planning economically efficient electricity grid operations, designing rates, and evaluating infrastructure and market policies , , , . Grid operators rely on load forecasts that are highly dependent on the day of week and time of year, holidays, and medium-range weather forecasts . While most of the load forecast uncertainty at these time scales is typically driven by uncertainty in weather forecasting , , behavioral and economic shifts during the COVID-19 pandemic and their potential to persist beyond the current crisis are introducing new uncertainties .
Shelter-in-place measures to limit the viral spread of COVID-19 during the spring of 2020 disrupted the ways in which people work, learn, and socialize. These disruptions modified and may continue to modify when, where, and how we use electricity. For example, several large companies in the United States (U.S.) and United Kingdom have announced plans to allow for widespread permanent teleworking even after employees are allowed back into the office , , . A recent Pew Research Center survey found that most of those working from home were not doing so before the pandemic and the majority of the teleworkers surveyed would prefer to keep working from home after the pandemic , .
Teleworking and stay-at-home orders resulting from COVID-19 caused increases in residential electricity demand and decreases in commercial and industrial demand , , , , . In the U.S., the Energy Information Administration (EIA) reported that nationwide total energy consumption in April 2020 dropped to its lowest level for any single month since September 1989 . The changes in electricity consumption varied by sector. Residential electricity sales were 6% higher in April 2020 than in any April of the five previous years while commercial and industrial sales decreased by 11% and 9%, respectively, compared with April 2019 . Overall, electricity demand and prices dropped markedly in multiple electricity markets in the U.S. . In an analysis of integrated electricity market data across the U.S., Ruan et al. found reductions relative to a backcast estimation (no COVID-19 scenario) of 6.4% – 10.2% in April and 4.4% – 10.7% in May . They correlated reductions in electricity consumption in populous cities with reduced commercial activity and an increase in the stay-at-home population .
There were also shifts in the diurnal profile of electricity consumption associated with COVID-19. A blog post from the New York Independent System Operator (NYISO) reported later-than-normal morning peaks, higher midday residential energy use, and temporal patterns similar to “a widespread snow day” . A comparison of electricity demand profiles across Europe during the second week of April 2020 showed that total weekday electricity consumption was considerably reduced in countries that instituted COVID-19 containment measures and that the weekday hourly profiles resembled pre-pandemic weekend profiles . COVID-19-induced deviations from normal diurnal patterns of electricity consumption in the residential and commercial sectors appears to have contributed to several notable spikes in the error of day-ahead forecasts that underpin the electricity market in the U.S. (Fig. 1). While some of the U.S. reserve regions examined had minimal changes in the distribution of their errors, several had significant spikes or increases in the variability of their forecast error after the onset of COVID-19-induced shutdowns in March of 2020 (e.g., the Carolinas and the Midwest).
The potential for increased uncertainty due to long-term changes induced by COVID-19 will create challenges in predicting future hourly load profiles. Accurate hourly load forecasts are critical for both short-term operations and longer-term planning. These projections need to capture sectoral and geographical differences, as a decrease in overall electricity consumption but overall higher diurnal load profile uncertainty can affect regional production costs, ancillary service needs and investment costs (e.g., ramping and storage), and market prices differently , . For example, the effect of the pandemic on electricity demand has been shown to differ among residential, commercial, and industrial sectors ,  and the effect of the pandemic on electricity generation fuel mixes has been shown to differ among the three U.S. Regional Transmission Organizations (RTOs; ). Future load profiles will need to reflect societal changes that were not present in pre-COVID-19 operational and long-term studies. Prior research on the energy and climate impacts of teleworking has focused primarily on changes in vehicle distance traveled, with overall energy use being a secondary focus . While some pre-COVID-19 studies have addressed the impacts of teleworking on total energy use in the residential and commercial sectors , , their results were provided at annual time scales as opposed to hourly.
This paper addresses a need for new research investigating the impacts of COVID-19 on hourly electricity load profiles by customer class within a given region of the electrical grid, complemented with insights on how these hourly profile changes vary across spatiotemporal scales. The novelty of this work is that we analyzed three independent datasets of observed electricity consumption differing by spatiotemporal scale and level of customer data aggregation to explore how these differences affected our ability to detect changes in electricity demand due to COVID-19. Our work expands upon the growing literature on impacts of COVID-19 on electricity consumption, most of which is focused on a single scale or sector. The electricity sector in the U.S. is composed of a series of nested actors (e.g., customers, utilities, balancing authorities [BAs], and RTOs), each of whose decisions affect outcomes at both larger and smaller spatial scales. For example, local utilities make decisions regarding rates and infrastructure buildout while RTOs lead long-term resource adequacy planning efforts and BAs are responsible for real-time balancing of supply and demand. We will show that COVID-19 impacted the electric sector in a variety of ways and that the apparent nature of those impacts varied depending on spatiotemporal and sectoral scales. We highlight several types of offsetting effects that acted to mute the COVID-19 signal. Our analysis suggests that modeling diurnal load patterns by customer class is necessary not only to understand recent changes in the total load profile due to COVID-19, but also to improve future load forecasts to account for the persistence of COVID-19-induced changes in patterns of electricity consumption.
COVID-19 induced substantial changes in when and where we use electricity and created an unprecedented source of uncertainty for the electric grid. The primary outcome of our analysis is to show that the apparent impact of COVID-19 varied depending on the spatiotemporal scale and level of aggregation of data being examined. At every scale COVID-19 impacts first appeared in the spring of 2020 and persisted at least through the fall of 2020. We identified three factors that acted to mask the impact of COVID-19:
1. Across all three scales we examined, the impact of COVID-19 was dampened because changes in residential loads and non-residential loads had a similar magnitude but moved in opposite directions.
2. During the summer of 2020 total loads became more strongly dependent on diurnal temperature variations and less dependent on inhabitance schedules. This acted to reduce the impact of more people staying home during the day – a signal that was clearly evident in the spring and then re-emerged in the fall of 2020.
3. Phased reopening and spatial variability in reopening during the summer of 2020 masked the COVID-19 signal when total load data were analyzed at the balancing authority scale. Breaking the data down into smaller zones within the balancing authority showed that the impact of COVID-19 persisted through the summer and into the fall of 2020…