HST–MB–CREH: A Hybrid Spatio-Temporal Transformer with Multi-Branch CNN/RNN for Rare-Event-Aware PV Power Forecasting

Taganova G. Tussupov J. Abdildayeva A. Kaldarova M. Kazi A. Simpson R.C. Zakirova A. Nurbekov B.
February 2026 Multidisciplinary Digital Publishing Institute (MDPI)

Algorithms
2026 #19 Issue 2

We propose the Hybrid Spatio-Temporal Transformer with Multi-Branch CNN/RNN and Extreme-Event Head (HST–MB–CREH), a hybrid spatio-temporal deep learning architecture for joint short-term photovoltaic (PV) power forecasting and the detection of rare extreme events, to support the reliable operation of renewable-rich power systems. The model combines a spatio-temporal transformer encoder with three convolutional neural network (CNN)/recurrent neural network (RNN) branches (CNN → long short-term memory (LSTM), LSTM → gated recurrent unit (GRU), CNN → GRU) and a dense pathway for tabular meteorological and calendar features. A multitask output head simultaneously performs the regression of PV power and binary classification of extremes defined above the 95th percentile. We evaluate HST–MB–CREH on the publicly available Renewable Power Generation and Weather Conditions dataset with hourly resolutions from 2017 to 2022, using a 5-fold TimeSeriesSplit protocol to avoid temporal leakage and to cover multiple seasons. Compared with tree ensembles (RandomForest, XGBoost), recurrent baselines (Stacked GRU, LSTM), and advanced hybrid/transformer models (Hybrid Multi-Branch CNN–LSTM/GRU with Dense Path and Extreme-Event Head (HMB–CLED) and Spatio-Temporal Multitask Transformer with Extreme-Event Head (STM–EEH)), the proposed architecture achieves the best overall trade-off between accuracy and rare-event sensitivity, with normalized performance of RMSE_z = 0.2159 ± 0.0167, MAE_z = 0.1100 ± 0.0085, mean absolute percentage error (MAPE) = 9.17 ± 0.45%, R² = 0.9534 ± 0.0072, and AUC_ext = 0.9851 ± 0.0051 across folds. Knowledge extraction is supported via attention-based analysis and permutation feature importance, which highlight the dominant role of global horizontal irradiance, diurnal harmonics, and solar geometry features. The results indicate that hybrid spatio-temporal multitask architectures can substantially improve both the forecast accuracy and robustness to extremes, making HST–MB–CREH a promising building block for intelligent decision-support tools in smart grids with a high share of PV generation.

deep learning , extreme event detection , Hybrid Spatio-Temporal Transformer , multi-branch CNN/RNN , PV power forecasting , transformer encoder

Text of the article Перейти на текст статьи

Faculty of Information Technology, L. N. Gumilyov Eurasian National University, Astana, 010000, Kazakhstan
Higher School of Information Technology and Engineering, Astana International University, Astana, 010000, Kazakhstan
Department of Smart Technologies in Engineering, International Engineering-Technological University, Almaty, 050000, Kazakhstan
Techno Women, Non-Profit Organization, Astana, 010000, Kazakhstan
GovTech & Al Foundation Public Foundation, Astana, 010000, Kazakhstan
Freelance Engineering Ltd., Leeds, Garforth, LS25 1NB, United Kingdom
Mechanics and Mathematics Faculty, L. N. Gumilyov Eurasian National University, Astana, 010000, Kazakhstan

Faculty of Information Technology
Higher School of Information Technology and Engineering
Department of Smart Technologies in Engineering
Techno Women
GovTech & Al Foundation Public Foundation
Freelance Engineering Ltd.
Mechanics and Mathematics Faculty

10 лет помогаем публиковать статьи Международный издатель

Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026