Optimizing Monthly Solar PV Tilt Angles and Energy Yield Across Global Climate Zones

This dataset supports the study “Optimizing Monthly Solar PV Tilt Angles and Energy Yield Across Global Climate Zones: A Hybrid Machine Learning and PVLib Approach.” It provides simulation-derived optimal monthly tilt angles and machine learning–ready input features for photovoltaic (PV) performance optimization across 17 globally distributed cities representing tropical, subtropical, temperate, and cold/subpolar climate regions. The dataset combines: Typical Meteorological Year (TMY) climate data from PVGIS PV energy output simulations using PVLib (tilt range 0°–70°) Derived optimal monthly tilt angles based on maximum simulated energy yield Machine learning features, including irradiance components, atmospheric variables, geographic coordinates, and one-hot encoded climate zone identifiers The dataset enables: Monthly and seasonal tilt optimization for rooftop or ground-mounted PV systems Climate-sensitive PV yield analysis across regions and latitudes Training and benchmarking of supervised ML models for tilt prediction Comparative study of empirical, seasonal, and ML-derived tilt strategies The file all_cities_ml_ready_onehot.csv contains ~67,000 samples. Each row represents a unique city-month observation with associated environmental variables, simulation-derived energy outputs for multiple tilt angles, and the corresponding optimal tilt label. This facilitates both regression (predict tilt angle) and classification (tilt category) tasks. This dataset is released under a Creative Commons Attribution 4.0 (CC BY 4.0) license to support transparency and reproducibility in solar PV research. Users may share, adapt, and build upon the dataset with proper citation.