Urban Energy Load Dataset of Chattogram: Peak and Off-Peak Variability Insights
Description
The dataset from the Chattogram office of the Power Grid Company of Bangladesh (PGCB) covers the time period of January 1, 2021, to June 30, 2023. It records the daily electricity demand for one of the most economically important regions of the country. To strengthen the dataset for forecasting, meteorological variables from the NASA POWER database were incorporated, which allowed for a comprehensive evaluation of the climate demand variables. The Chattogram region has emerged as one of Bangladesh’s most significant industrial and commercial hubs. Driven by rapid urbanization, expanding industrial operations, and increasing residential consumption, electricity demand in the region has been steadily rising. To ensure that this growth does not result in frequent load shedding or unstable power supply, it is essential to carefully manage demand. In support of this objective, the Chattogram office of the Power Grid Company of Bangladesh (PGCB) supplied data spanning January 1, 2021, to June 30, 2023. This dataset documents electricity consumption across three distinct time windows: day peak (9 AM-5 PM), evening peak (6 PM-9 PM) and off-peak (10 PM-8 AM). It also captures fluctuations in demand shaped by industrial activity, household usage and baseline requirements, thereby offering insights into load variations over the study period. To enrich the demand analysis, meteorological variables were incorporated into the dataset, sourced from NASA’s POWER (Prediction of Worldwide Energy Resources) database. These variables include temperature, dew point, frost point, dry-bulb and wet-bulb temperatures, earth skin temperature, maximum and minimum temperatures, specific humidity, relative humidity, and precipitation. Together, these meteorological factors strengthen the analytical framework for understanding patterns of urban energy consumption. The uniqueness of this dataset lies in its integration of meteorological conditions with daily peak and off-peak demand records. This combination provides a robust foundation for exploring the complex interplay between weather and electricity use. Furthermore, the dataset is highly suitable for developing advanced forecasting models that incorporate both linear and nonlinear components, enhancing predictive accuracy in the medium term. Such models can play a critical role in advancing energy efficiency, strengthening climate resilience and improving grid reliability within one of Bangladesh’s fastest-growing urban centers. For work done on this data, refer to: https://doi.org/10.1016/j.rineng.2025.107245
Files
Institutions
- University of Science and Technology Chittagong