Citation

El Ghoul, Islam Nacer Eddine and Beddar, Antar and Hadjrioua, Farid and Azzouz, Abdelbasset and Tiang, Jun Jiat (2026) Deep Learning Benchmarks for Multi-Step Photovoltaic Power Forecasting: Comparative Assessment of GRU, LSTM, and BiLSTM Architectures. Energies, 19 (19). p. 4594. ISSN 1996-1073

Text energies-19-04594.pdf - Published Version Restricted to Repository staff only Download (892kB)

Abstract

Accurate photovoltaic (PV) power forecasting is essential for renewable energy integration, dynamic reserve allocation, and generation scheduling. Unpredicted generation ramps induce substantial voltage and frequency deviations on grid-connected distribution networks. This paper provides an objective benchmark among three deep learning architectures: Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM). Telemetry spanning one year from an operational 175 kW plant in Reggane, Algeria, is evaluated across 72 h horizons (H=288 steps at 15 min resolution) using a Direct Multi-Input Multi-Output framework under strict chronological splitting. The models are systematically evaluated across 1848 rolling 72 h forecast windows using a controlled benchmark protocol under daylight-only conditions (GHI>5 W/m2 ) against a Diurnal Smart Persistence baseline. Accuracy is assessed via daylight capacity-normalized MAE (nMAE ), nRMSE , Diebold–Mariano tests (h=72 lags), and ramp metrics across 34,840 events. The results demonstrate that deep architectures significantly outperform persistence (

Item Type:	Article
Uncontrolled Keywords:	photovoltaic power forecasting, direct MIMO, bidirectional LSTM (BiLSTM), diurnal smart persistence, Diebold–Mariano test, ramp event detection, daylight capacity normalization, multi-step time series, recurrent neural networks
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	30 Oct 2026 02:35
Last Modified:	30 Oct 2026 02:35
URII:	http://shdl.mmu.edu.my/id/eprint/17027

Downloads

Edit (login required)