AI‑Driven Adaptive Hybrid Forecasting of Solar PV Power via Dirichlet Process Clustering and Ensemble Regression

Hajian, Mohammad; Sami, Soha; Nazghelichi, Tayyeb

doi:10.48301/vjtve.2026.542635.1139

AI‑Driven Adaptive Hybrid Forecasting of Solar PV Power via Dirichlet Process Clustering and Ensemble Regression

Document Type : Original Article

Authors

Mohammad Hajian ¹

soha sami ²

Tayyeb Nazghelichi ¹

¹ Gorgan University of Agricultural Sciences and Natural Resources, Faculty of Water and Soil Engineering, Department of Biosystems Engineering, Gorgan, Iran.

² Mechanics of Biosystems Engineering Department, College of Aburaihan, University of Tehran, Tehran, Iran.

10.48301/vjtve.2026.542635.1139

Abstract

Accurate forecasting of photovoltaic (PV) power generation is important to enhance the efficiency and reliability of renewable energy in the present power grids. Despite significant progress in predicting PV generation, the nature of PV data remains heterogeneous, non-linear, and dynamic, making modelling and predictive ability challenging in their own right. This dissertation proposes a self-adaptive hybrid approach that offers a novel development avenue within the renewable energy forecast paradigm. It utilizes a Nonparametric Dirichlet Process Gaussian Mixture Model (DPGMM) to identify the distinct predictive capabilities of machine-learning algorithms from a validation perspective, thereby informing their employment in the hybrid method. DPGMM was able to identify homogeneous subgroups from a ~3-year PV dataset of 300 households in Sydney, Australia, allowing the development of distinct predictive models for each subgroup. From a statistical validation perspective, contrast testing was completed using a Friedman and Wilcoxon signed-rank procedure. While there exist differences between the models (p < 0.05), it was clear that Gradient Boosting Regression (GBR) outperformed the support vector regression (SVR) and random forest regression (RFR) models on all clusters. Finally, metrics of the mean absolute error (MAE), mean square error (MSE) and root mean square error (RMSE) were employed to examine and measure improvements in accuracy compared to baseline methods. The hybrid approach yielded statistically significant improvement, which included increases of 12% for MAE, 22% for MSE, 15% for RMSE compared against single-stage convolution implementations.

Keywords

Photovoltaic Power Forecasting

Hybrid AI Model

Deep Learning

Renewable Energy

Machine Learning

Cluster-Based Prediction

Subjects

Electrical Engineering

[1] El Hendouzi, A., Bourouhou, A., & Ansari, O. (2020). The Importance of Distance between Photovoltaic Power Stations for Clear Accuracy of Short-Term Photovoltaic Power Forecasting. Journal of Electrical and Computer Engineering, 2020. https://doi.org/10.1155/2020/9586707

[2] Park, T., Song, K., Jeong, J., & Kim, H. (2023). Convolutional Autoencoder-Based Anomaly Detection for Photovoltaic Power Forecasting of Virtual Power Plants. 16(14).

[3] Phan, Q. T., Wu, Y. K., Phan, Q. D., & Lo, H. Y. (2023). A Novel Forecasting Model for Solar Power Generation by a Deep Learning Framework With Data Preprocessing and Postprocessing. IEEE Transactions on Industry Applications, 59(1), 220–231. https://doi.org/10.1109/TIA.2022.3212999

[4] Huang, H., Zhu, Q. L., Zhu, X., & Zhang, J. (2023). An Adaptive, Data-Driven Stacking Ensemble Learning Framework for the Short-Term Forecasting of Renewable Energy Generation. Energies. https://doi.org/10.3390/en16041963

[5] Elsaraiti, M., & Merabet, A. (2022). Solar Power Forecasting Using Deep Learning Techniques. IEEE Access, 10, 31692–31698. https://doi.org/10.1109/ACCESS.2022.3160484

[6] Mofidian, R., Hassankhani, I., Jahanshahi, M., Hosseini, S. S., & Miansari, M. (2024). Cost Effective Design of a 200 kW On-grid Rooftop Photovoltaic System Using PVsyst Software in Shiraz. Journal of Engineering and Applied Research, 1(1), 13–24. https://doi.org/10.48301/jear.2024.194109

[7] Taghandiki, K. (2024). Quantum Machine Learning Unveiled: A Comprehensive Review. Journal of Engineering and Applied Research, 1(2), 29–48. https://doi.org/10.48301/jear.2024.446673.1021

[8] Wang, S., Yan, S., Li, H., Zhang, T., Jiang, W., Yang, B., Li, Q., Li, M., Zhang, N., & Wang, J. (2024). Short-term prediction of photovoltaic power based on quadratic decomposition and residual correction. Electric Power Systems Research, 236, 110968–110968. https://doi.org/10.1016/j.epsr.2024.110968

[9] Qiu, L., Ma, W., Feng, X., Dai, J., Dong, Y., Duan, J., & Chen, B. (2024). A hybrid PV cluster power prediction model using BLS with GMCC and error correction via RVM considering an improved statistical upscaling technique. Applied Energy, 359, 122719–122719. https://doi.org/10.1016/j.apenergy.2024.122719

[10] Hu, W., Zhang, X., Zhu, L., & Li, Z. (2021). Short-Term Photovoltaic Power Prediction Based on Similar Days and Improved SOA-DBN Model. IEEE Access, 9, 1958–1971. https://doi.org/10.1109/ACCESS.2020.3046754

[11] Zhu, R., Li, T., & Tang, B. (2024). Research on short-term photovoltaic power generation forecasting model based on multi-strategy improved squirrel search algorithm and support vector machine. Scientific Reports, 14(1), 14348–14348. https://doi.org/10.1038/s41598-024-65159-1

[12] Akoglu, H. (2018). User's guide to correlation coefficients. Turkish Journal of Emergency Medicine, 18(3), 91–93. https://doi.org/10.1016/j.tjem.2018.08.001 https://doi.org/10.1016/j.tjem.2018.08.001

[13] Tao, Z., Li, J., Fu, H., Kong, Y., & Fu, Y. (2021). From ensemble clustering to subspace clustering: Cluster structure encoding. IEEE Transactions on Neural Networks and Learning Systems, 34(5), 2670–2681. https://doi.org/10.1109/TNNLS.2021.3107354

[14] Persson, C., Bacher, P., Shiga, T., & Madsen, H. (2017). Multi-site solar power forecasting using gradient boosted regression trees. Solar Energy, 150, 423–436. https://doi.org/10.1016/j.solener.2017.04.066

[15] Ridgeway, C. L., & Nakagawa, S. (2017). Is Deference the Price of Being Seen as Reasonable? How Status Hierarchies Incentivize Acceptance of Low Status. Social Psychology Quarterly, 80(2), 132–152. https://doi.org/10.1177/0190272517695213

[16] Liu, D., & Sun, K. (2019). Random forest solar power forecast based on classification optimization. Energy, 187, 115940–115940. https://doi.org/10.1016/j.energy.2019.115940

[17] Borup, D., Christensen, B. J., Mühlbach, N. S., & Nielsen, M. S. (2023). Targeting predictors in random forest regression. International Journal of Forecasting, 39(2), 841–868. https://doi.org/10.1016/j.ijforecast.2022.02.010

[18] Ahmed, S. I., Bhuiyan, K. A., Rahman, I., Salehfar, H., & Selvaraj, D. F. (2024). Reliability of regression based hybrid machine learning models for the prediction of solar photovoltaics power generation. Energy Reports, 12, 5009–5023. https://doi.org/10.1016/J.EGYR.2024.10.060

[19] Awad, M., & Khanna, R. (2015). Support Vector Regression BT - Efficient Learning Machines: Theories, Concepts, and Applications for Engineers and System Designers. In M. Awad & R. Khanna (Eds.), (pp. 67–80). Apress. https://doi.org/10.1007/978-1-4302-5990-9_4