Random Forest Regression with Lag Features for Power Output Prediction of Automatic Street Lighting System Based on Fresnel Solar Concentrator
Abstract
The variability of DC Street Lighting (PJU) output power based on Fresnel Solar Concentrator due to weather fluctuations presents a challenge in solar energy management. This study develops a prediction model using Random Forest Regression (RFR) with IoT sensor time-series lag features and BMKG data, utilizing 44 predictor variables from 405 records collected over 27 days. Hyperparameter optimization via 5-fold Grid Search Cross-Validation yielded n_estimators=200 and max_depth=15. Evaluation results achieved R²=0.9850 and MAPE=8.29%, satisfying all success criteria. Panel power (P_Panel) was the most dominant predictor with importance score 0.1197. The model successfully predicted output power for 7 days ahead with an average energy estimate of 250 Wh per day, proving that lag features integration into RFR is effective for proactive PJU energy management.
Downloads
References
M. A. Kurniawan et al., “Analisis Perancangan Penerangan Jalan Umum Tenaga Surya ( PJU-TS ) Akses Desa Wisata Mantar Kabupaten Sumbawa Barat,” vol. 6, pp. 2951–2956, 2023.
M. S. Alim, S. Thamrin, and R. L. W, “Pemanfaatan Pembangkit Listrik Tenaga Surya sebagai Alternatif Ketahanan Energi Nasional Masa Depan,” vol. 4, no. 3, pp. 2427–2435, 2023.
M. A. Saputra, M. F. Azis, and E. Aditia, “No Title”.
J. Gaboitaolelwe, “Machine Learning Based Solar Photovoltaic Power Forecasting : A Review and Comparison,” IEEE Access, vol. 11, no. March, pp. 40820–40845, 2023, doi: 10.1109/ACCESS.2023.3270041.
P. Setiawati et al., “PREDICTING SOLAR POWER GENERATION : A MACHINE LEARNING,” vol. 21, no. 1, pp. 34–43, 2025, doi: 10.33480/pilar.v21i1.6126.
A. Bouakkaz, A. Lahsasna, A. J. Gil, M. Luigi, and R. Jiménez-castaneda, “Evaluating and analyzing the performance of PV power output forecasting using different models of machine-learning techniques considering prediction accuracy,” vol. 14, no. 1, pp. 158–167, 2025.
A. J. Lari, A. P. Sanfilippo, D. Bachour, and D. Perez-astudillo, “Using Machine Learning Algorithms to Forecast Solar Energy Power Output,” pp. 1–22, 2025.
C. G. Villegas-mier, J. Rodriguez-resendiz, and J. M. Álvarez-alvarado, “Optimized Random Forest for Solar Radiation Prediction Using Sunshine Hours,” 2022.
M. Sari, “G-Tech : Jurnal Teknologi Terapan Indonesia Using the Random Forest Method,” vol. 9, no. 4, pp. 1771–1781, 2025.
E. A. Handoyo et al., “Perancangan dan Pengujian Lensa Fresnel pada Kolektor Surya Plat Datar,” vol. 17, no. 2, 2020, doi: 10.9744/jtm.17.2.48.
A. I. Press, A. Wagyana, I. Artikel, J. T. Elektro, and P. N. Jakarta, “Jurnal Ilmiah Setrum,” vol. 8, no. 1, pp. 238–247, 2019.
W. Satria, W. Indrasari, and R. Fahdiran, “KARAKTERISASI SENSOR ARUS DAN TEGANGAN UNTUK APLIKASI MAXIMUM POWER POINT TRACKER PADA SISTEM PENYIMPANAN ENERGI,” vol. IX, pp. 77–82, 2020.
D. Aribowo, G. Priyogi, S. Islam, P. T. Elektro, U. Sultan, and A. Tirtayasa, “APLIKASI SENSOR LDR ( LIGHT DEPENDENT RESISTOR ) UNTUK EFISIENSI ENERGI PADA LAMPU PENERANGAN,” vol. 9, no. 1, 2022.
BMKG, “DataOnlineBMKG.” [Online]. Available: https://dataonline.bmkg.go.id/dataonline-home
M. Ichsan, “MACHINE LEARNING DETEKSI PENYAKIT PADA KUCING MENGGUNAKAN,” vol. 13, no. 3, 2000.
L. E. O. Breiman, “Random Forests,” pp. 5–32, 2001.
F. L. Rashid, M. A. Al-obaidi, and A. J. Mahdi, “Advancements in Fresnel Lens Technology across Diverse Solar Energy Applications : A Comprehensive Review,” 2024.
G. A. Sehangunaung et al., “KEPUASAN KONSUMEN PENGGUNA APLIKASI LAZADA DI KOTA MANADO THE INFLUENCE OF SERVICE QUALITY ON CUSTOMER SATISFACTION USING DELIVERY Jurnal EMBA Vol . 11 , No . 3 . Juli 2023 , Hal . 1-11,” vol. 11, no. 3, pp. 1–11, 2023.
J. Adi and S. No, “PENGUKURAN PERFORMA SUPPORT VECTOR MACHINE DAN NEURAL NETWOK DALAM Diyah Ruswanti,” vol. 13, no. mm, pp. 66–75, 2020.
H. H. Nuha, U. Telkom, and J. Barat, “Mean Absolute Percentage Error ( MAPE ) dan Penggunaannya,” vol. 3, no. December 2024, pp. 3–5, 2025.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
