irradpy2 0.2.5

Multi-source solar radiation downloader and forecasting toolkit

irrapy2

irrapy2 is a lightweight Python toolkit for downloading, parsing, and processing multi-source ground-based solar radiation and meteorological datasets, combined with a simple forecasting module (RNN / LSTM / Informer).

The package provides a unified interface to several international radiation networks:

• BSRN – Baseline Surface Radiation Network
• MIDC (NREL) – Measurement & Instrumentation Data Center
• SAURAN – Southern African Universities Radiometric Network
• SRML – Solar Radiation Monitoring Laboratory (UOregon)
• SURFRAD (NOAA) – Surface Radiation Budget Network
• SOLRAD (NOAA) – Solar Radiation Network

It also includes a built-in forecasting tool for solar irradiance time series.

---

📦 Installation

pip install irrapy2

---

🚀 Quick Start

Download BSRN data

from irrapy2 import download_bsrn

download_bsrn(
    site="cab",
    start="2023-01-01",
    end="2023-01-31",
    username="your_bsrn_username",
    password="your_bsrn_password",
    save_path="bsrn_out.csv"
)

Download MIDC data

from irrapy2 import download_midc

download_midc(
    site="BMS",
    begin="20230101",
    end="20230131",
    save_path="midc_out.csv"
)

Forecast Solar Radiation

from irrapy2 import run_forecast

metrics = run_forecast(
    csv_path="bsrn_out.csv",
    model="lstm",       # rnn / lstm / informer
    seq_len=72,
    horizon=6,
    epochs=10
)

print(metrics)

---

📁 Supported Datasets

Source	Region	Module	Time Handling
BSRN	Global	download_bsrn	UTC
MIDC (NREL)	USA	download_midc	Local + DST + UTC
SAURAN	Africa	download_sauran	Local → UTC
SRML	USA	download_srml	America/Los_Angeles
SURFRAD	USA	download_surfrad	UTC
SOLRAD	USA	download_solrad	UTC + Local

---

🔧 Requirements

pandas
numpy
requests
pytz
torch
scikit-learn

---

📄 License

This project is open-source and licensed under the Apache License 2.0.

---

🙌 Contributing

Issues and pull requests are welcome.
Feel free to open a PR if you want to add more stations, enhance parsing logic, or improve the forecasting module.