Current and future residential electricity demand using large-scale smart meter data in a changing climate
Mingyue Guo, Youngsik Choi, So‐Min Cheong, Zheng D. O’Neill · Sustainable Cities and Society · 2025
The rising demand for electricity in a changing climate places an increased strain on the grid and heightens the risk of outages. Residential buildings, which account for more than one-third of electricity consumption in the U.S., contribute significantly to this growing demand. Given the increased blackout risk from climate change and the importance of demand-side management, this study offers a comprehensive quantitative analysis of current and future county-level residential electricity demand resilience of 1.93 million consumers across 134 ZIP Code Tabulation Areas (ZCTA5) in Harris County, Texas, U.S. in the cooling season of 2022 and 2023. Nine resilience-related key performance indicators (KPIs), such as peak demand, peak time, load factor, and ramping rate, were extracted to characterize resilience. Spatial analysis revealed higher peak demands and ramping rates in northwestern and western Harris County, while areas near Downtown Houston exhibited higher load factors. Peak demands across ZCTA5 tended to occur almost simultaneously. Temporal analysis showed a 3.87% increase in county-level maximum peak demand in 2023 (peak demand: 7.51 GW) compared to 2022 (peak demand: 7.21 GW), alongside a maximum outdoor air temperature rise of 1.93°C. The most frequent time of peak demand was at 17:00 and the time of maximum ramping rates was at 11:00. The load factor and the valley-to-peak ratio displayed similar distributions in 2022 and 2023, with a median of 0.69 for both years. The maximum ramping rate in 2022 was 0.69 GW/hr, occurring at 11:00 on June 20, 2022, while in 2023, the maximum ramping rate reached 0.80 GW/hr at 12:00 on August 27, 2023. To assess the future electricity demand in Harris County, an artificial neural network (ANN) was used to forecast electricity consumption for the typical cooling season under four climate scenarios (SSP5-RCP8.5, SSP3-RCP7.0, SSP2-RCP4.5, and SSP1-RCP2.6) in years from 2020 to 2099. Under high and very high carbon emission scenarios, county-level electricity demand is projected to increase by 7.8% in the SSP3-RCP7.0 scenario and by 5.7% in the SSP5-RCP8.5 scenario by 2080-2099, compared with 2023. The most frequent peak time is expected to remain at 17:00, while the maximum ramping rates are projected to occur between 11:00 and 12:00. Distributed energy resources, thermal energy storage, battery systems, demand-side energy management, and improved passive survivability are viable ways to meet the increasing electricity demand in the future.