Hydropower Generation and Weather-Based Energy Dataset from Karnaphuli Power Station

This dataset is a processed and integrated dataset constructed using publicly available power generation data from the Bangladesh Power Development Board (BPDB) and satellite-based meteorological data from NASA IMERG.

The dataset combines hydropower generation records from the Karnaphuli Hydro Power Plant with corresponding environmental and weather parameters, including temperature, relative humidity, precipitation, solar radiation, surface pressure, and wind speed.

Data preprocessing, cleaning, and feature engineering were performed to align and integrate multiple data sources into a structured format suitable for analysis and machine learning applications.

This dataset was developed and utilized in a published research study examining the relationship between environmental factors and hydropower generation.

Applications:- Energy forecasting- Machine learning modeling- Renewable energy analysis- Climate impact studies

Dataset Details:- Samples: 1759- Features: 11- Format: CSV (tabular time-series)- Time span: January 2021 to November 2025

Feature Description:- Date: Observation date- Day Peak (MW): Daytime peak power generation- Evening Peak (MW): Evening peak power generation- Temperature (°C): Daily average air temperature- Relative Humidity (%): Atmospheric moisture level- Precipitation (mm): Daily rainfall amount- Solar Radiation (W/m²): Surface solar irradiance- Wind Speed (m/s): Wind speed at 10 meters- Surface Pressure (kPa): Atmospheric pressure- LHF: Latent heat flux related feature (derived)- Additional engineered features: Lag and interaction-based variables used for modeling

Sources:- Bangladesh Power Development Board (BPDB)- NASA IMERG satellite data

Related Publication:Daily Hydropower Generation Forecasting Scheme Using Satellite Climate Data, SHAP, and LightGBM Technique, IEEE, 2026.DOI: 10.1109/IATMSI68868.2026.11466157