Advanced Deep Learning Approach for Solar Radiation Prediction in Medan: CNN-LSTM Hybrid Model
DOI:
https://doi.org/10.15294/edukom.v13i1.42730Keywords:
CNN-LSTM Hybrid Model, Deep Learning, Solar Radiation PredictionAbstract
The availability of solar energy in Indonesia is hindered by small-scale variations in solar radiation that occur in tropical climates. For maintaining grid stability and for ensuring the efficient operation of solar power systems, accurate short-term prediction is necessary. In this study, a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model is developed in order to predict minute-scale solar radiation in Medan City. The model makes use of minute-resolution meteorological data (including temperature, humidity, dew point, and wind speed), combining CNNs for the extraction of spatial features with LSTMs for temporal learning. Four different architectural setups were evaluated, and Model 4 — consisting of four LSTM layers with 50, 100, 150, and 200 units, respectively — achieved the best results: an RMSE of 39.88 W/m², a MAE of 26.89 W/m², and an R² of 0.955, which represents a 20.82% improvement on the baseline models. The feature importance analysis found that temperature, dew point, and relative humidity were the most significant predictive factors. The model is able to capture both the daily cycles and the rapid fluctuations typical of tropical climates, thus providing accurate predictions that are important for the stability of the electricity grid and for the optimisation of solar power plants. The hybrid architecture that has been developed provides a transferable framework which can be applied to other tropical cities in Indonesia and thus contributes to the country's energy transition objectives and to the development of sustainable renewable energy. This research promotes advances in deep learning methods for forecasting renewable energy while tackling the real-world problems associated with integrating solar energy under changing atmospheric conditions.
References
Abdulkadir Tasdelen, B. S. (2021). A hybrid CNN-LSTM model for pre-miRNA classification. IEEE Access, 1-9.
Asmaa Halbouni, T. S. (2022). CNN-LSTM: Hybrid Deep Neural Network for Network Intrusion Detection System. IEEE Access, 99837-99849.
Awan Nahel Mahmood, M. N. (2025). Enhancing QoE Prediction for Video Streaming Using a Hybrid CNN-LSTM Model. IEEE Access, 174105-174119.
Daram, M. R. (Muhammad Rezza., M. Ismalil Yusuf, Redi Ratiandi YAcoub). Prediksi Radiasi surya Menggunakan Metode Long Short-Term Memory. Journal of Computer Science and Applied Informations, 6 (1)(Journal of Computer Science and Applied Informations), 33-44.
Felipe P. Marinho, P. A. (2022). Short-Term Solar Irradiance Forecasting Using CNN-1D, LSTM, and CNN-LSTM Deep Neural Networks: A Case Study With the Folsom (USA) Dataset . Journal of Solar Energy Engineering, 145(ASME Digital Collection), 1-11.
Hadi Suyono, H. S. (2018). Prediction of Solar Radiation Intensity using Extreme Learning Machine. Indonesia Journal of Electrical Engineering and Computer Science, 691-698.
K. P. Rasheed Abdul Haq, V. P. (2022). Water Quality Prediction for Smart Aquaculture Using Hybrid Deep Learning Models. IEEE Access, 60078 - 60098.
Kaleem Ullah, M. A. (2024). Short-Term Load Forecasting: A Comprehensive Review and Simulation Study With CNN-LSTM Hybrids Approach. IEEE Access, 1-24.
Meenal, R. M. (2021). Weather prediction using random forest machine learning model. International Journal of Engineering and Computer Science, 22 (2)(International Journal of Engineering and Computer Science), 1208-1215.
Mishra, D. P. (2023). Global solar radiation forecast using an ensemble learning approach. . International Journal of Power Electronics and Drive Systems, 14 (1)(International Journal of Power Electronics and Drive Systems), 496-505.
Muhammad Arif Budiyanto, M. H. (2020). Physical reviews of solar radiation models for estimating global solar radiation in Indonesia. International Conference on Power and Energy Systems Engineering (CPESE 2020),, 6(Elsevier), 1206-1211.
Musaed Alhussein, K. A. (2020). Hybrid CNN-LSTM Model for Short-Term Individual Household Load Forecasting. IEEE Access, 180544-180557.
Onur Polat, A. A. (2025). Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection. IEEE Access, 102109-102132 .
Phalaagae, P., & Adamu Murtala Zungeru, A. Y. (2025). A Hybrid CNN-LSTM Model With Attention Mechanism for Improved Intrusion Detection in Wireless IoT Sensor Networks. IEEE Xplore, 57322-57341.
Priya Singh, M. J.-M. (2023). Harnessing a Hybrid CNN-LSTM Model for Portfolio Performance: A Case Study on Stock Selection and Optimization. IEEE Access, 104000-104015.
Radha Mahendran, S. D. (2025). Real-Time Analysis of Biomedical Signals in IoT-Enabled Healthcare Systems Using a Hybrid CNN-LSTM Model. IEEE , 1-12.
Ravinesh C. Deo, R. H.-M. (2022). Forecasting solar photosynthetic photon flux density under cloud cover effects: novel predictive model using convolutional neural network integrated with long short-term memory network. Stochastic Environmental Research and Risk Assessment, 3183–3220.
Sahbi Boubaker, M. B. (2021). Deep Neural Networks for Predicting Solar Radiation at Hail Region, Saudi Arabia. IEEE, 36719-36729.
Tianzhe Bao, S. A.-Q. (2020). A CNN-LSTM Hybrid Model for Wrist Kinematics Estimation Using Surface Electromyography. IEE Transactions on Intrumentation and Measurement, 1-9.
Tuyen Nguyen-Duc, N. N.-V.-N.-D.-T. (2026). A Hybrid Spatio-Temporal Deep Learning Framework for Short-Term Solar Power Forecasting Using Satellite Imagery Data: A Case Study in Hanoi, Vietnam. IET Renewable Power Generation, 1-18.
Xiuhong Chen, X. H. (2020). Intra day Forecast of Ground Horizontal Irradiance Using Long Short Term Memory Network. Journal of the Meteorological Society of Japan, 945-957.
Yeshanew Ale Wubet, K.-Y. L. (2022). Voice Conversion Based Augmentation and a Hybrid CNN-LSTM Model for Improving Speaker-Independent Keyword Recognition on Limited Datasets. IEEE Access, 89170-89180 .
