Authors: Faisal Ali, Mehmood Ul Haq, Usman Aziz, Nisha Anwar, Muhammad Baqir
Overview
Compares statistical, machine learning and deep learning models for forecasting solar energy in Lahore across multiple forecast horizons, to address the weather-driven uncertainty in the city's solar generation.
The models compared include SARIMAX, XGBoost, random forest, LitTFT, BiLSTM, CNN-BiLSTM, TCN and TFT, calibrated with NASA POWER and Open-Meteo data. The work supports PV planning, energy-yield estimation and renewable-energy integration.
This paper is under review. Full details will be added once it is published.
Details
- Status
- Under review
- Submitted to
- Renewable Energy










