Efficient Deep Learning for Short-Term Solar Irradiance Time Series Forecasting: A Benchmark Study in Ho Chi Minh City

TL;DR

This study benchmarks ten deep learning models for short-term solar irradiance forecasting in Ho Chi Minh City, identifying the Transformer as the most accurate and demonstrating effective model compression techniques.

Contribution

It provides a comprehensive comparison of deep learning architectures for GHI forecasting and introduces a knowledge distillation approach to reduce model size while maintaining accuracy.

Findings

Transformer achieves R^2 of 0.9696 in GHI forecasting.

Transformers focus on recent atmospheric data, Mamba captures 24-hour periodicity.

Knowledge distillation reduces Transformer size by 23.5% and error by MAE of 23.78 W/m^2.

Abstract

Reliable forecasting of Global Horizontal Irradiance (GHI) is essential for mitigating the variability of solar energy in power grids. This study presents a comprehensive benchmark of ten deep learning architectures for short-term (1-hour ahead) GHI time series forecasting in Ho Chi Minh City, leveraging high-resolution NSRDB satellite data (2011-2020) to compare established baselines (e.g. LSTM, TCN) against emerging state-of-the-art architectures, including Transformer, Informer, iTransformer, TSMixer, and Mamba. Experimental results identify the Transformer as the superior architecture, achieving the highest predictive accuracy with an R^2 of 0.9696. The study further utilizes SHAP analysis to contrast the temporal reasoning of these architectures, revealing that Transformers exhibit a strong "recency bias" focused on immediate atmospheric conditions, whereas Mamba explicitly leverages 24-hour periodic dependencies to inform predictions. Furthermore, we demonstrate that Knowledge Distillation can compress the high-performance Transformer by 23.5% while surprisingly reducing error (MAE: 23.78 W/m^2), offering a proven pathway for deploying sophisticated, low-latency forecasting on resource-constrained edge devices.