Research Article
Comparative Analysis of LSTM and Hybrid Attention-Based Architectures for Short-Term Electricity Consumption Forecasting: A Case Study in Togo
Issue:
Volume 14, Issue 3, September 2026
Pages:
65-82
Received:
18 June 2026
Accepted:
6 July 2026
Published:
21 August 2026
Abstract: TAccurate short-term electricity consumption forecasting is essential for operational planning, reserve allocation, and energy management in modern power systems. This study investigates the performance of recurrent and hybrid attention-based deep learning architectures for short-term electricity consumption forecasting, a sase study in Togo. The proposed framework integrates electricity consumption, meteorological, demographic, and temporal information in order to capture both intrinsic temporal dependencies and exogenous influences affecting electricity demand. Four forecasting architectures were evaluated using a weekly temporal window of 168 hours: LSTM-only, LSTM-decoder, LSTM-attention, and a hybrid LSTM--Multi-Head Attention--LSTM model. The experiments included multi-seed evaluation, ablation study, robustness analysis, and statistical comparison using the Wilcoxon signed-rank test. The results show that all models achieved extremely high forecasting accuracy, with coefficients of determination exceeding 0.9998 and MAPE values below 0.004%. The hybrid architectures slightly improved average forecasting performance, while the standalone LSTM model remained highly competitive. The robustness analysis revealed strong sensitivity to noisy inputs but moderate degradation under missing-data conditions. Statistical analysis indicated that the performance differences between architectures were not statistically significant at the 5% level. Overall, the study demonstrates that recurrent deep learning architectures provide reliable and operationally relevant solutions for electricity consumption forecasting in highly structured energy demand series.
Abstract: TAccurate short-term electricity consumption forecasting is essential for operational planning, reserve allocation, and energy management in modern power systems. This study investigates the performance of recurrent and hybrid attention-based deep learning architectures for short-term electricity consumption forecasting, a sase study in Togo. The p...
Show More
Research Article
Assessment of the Impact of Climatic Variations on the Photovoltaic Potential of Kankan (Guinea) Using NASA POWER Data and Artificial Neural Network (2000–2025)
Issue:
Volume 14, Issue 3, September 2026
Pages:
83-97
Received:
25 August 2026
Accepted:
7 September 2026
Published:
24 September 2026
DOI:
10.11648/j.sjee.20261403.12
Downloads:
Views:
Abstract: West Africa benefits from strong solar resources, yet in Guinea the influence of climate change on photovoltaic availability has received little attention. This work evaluates the effects of climate change on the photovoltaic potential of Kankan (Guinea) using daily data from the NASA POWER database for 2000-2025. The study focuses on global solar irradiation, air temperature, relative humidity, and wind speed. A semi-empirical model is used to estimate daily photovoltaic potential, and trends are analyzed with linear regression and the non-parametric Mann-Kendall test. In parallel, a multilayer perceptron artificial neural network (MPANN) is developed to reproduce the estimated photovoltaic potential from the selected climatic inputs. The analysis shows significant declines in air temperature (-0.0559°C year-1) and wind speed (-0.0077 ms-1year-1), together with a significant rise in relative humidity (+0.6249% year-1). Global solar irradiation presents no statistically significant trend over the study period. Despite these changes, the photovoltaic potential remains broadly stable, with a small and statistically non-significant decrease of -0.0152 kWh day-1year-1 (p = 0.186 > 0.05). The MPANN delivers very high predictive performance, with the coefficient of determination of 0.99, the mean absolute error (MAE) of 0.0266 kWh day-1, the root mean square error (RMSE) of 0.0478 kWh day-1, and the mean absolute percentage error (MAPE) of 0.134%. Overall, the climatic shifts observed in Kankan during the last twenty-six years have not significantly altered the region’s photovoltaic potential. This stability supports the suitability of solar energy as a sustainable option to meet Guinea’s growing energy needs. The study also highlights the usefulness of artificial intelligence methods for assessing and forecasting renewable energy resources in tropical regions.
Abstract: West Africa benefits from strong solar resources, yet in Guinea the influence of climate change on photovoltaic availability has received little attention. This work evaluates the effects of climate change on the photovoltaic potential of Kankan (Guinea) using daily data from the NASA POWER database for 2000-2025. The study focuses on global solar ...
Show More
,
Yao Bokovi
