Peramalan Keluaran Photovoltaic Kapasitas 120WP Menggunakan Metode Neural Network
DOI:
https://doi.org/10.36277/jteuniba.v9i2.1265Keywords:
PLTS, PV, Neural Network, Peramalan, MAPE.Abstract
Photovoltaic (PV) adalah komponen utama dalam Pembangkit Listrik Tenaga Surya (PLTS) yang mengubah energi matahari menjadi listrik. Namun, daya PV bersifat fluktuatif, berbeda dengan pembangkit konvensional yang memiliki daya konstan. Untuk menyeimbangkan kebutuhan daya dan produksi, diperlukan pembangkit tambahan yang disebut load tracking power plant. Peramalan keluaran daya PV menjadi penting untuk memastikan load tracking power plant dapat merespons secara cepat dan akurat. Penelitian ini memfokuskan pada peramalan daya PV menggunakan algoritma Neural Network (NN), yang terbukti memiliki akurasi tinggi. PV yang diuji memiliki efisiensi maksimum 13,16% pada intensitas cahaya 82.574,66 Lux dengan daya 60,67 W. Dalam intensitas cahaya 15.432,43 Lux, efisiensinya mencapai 15,48% dengan daya 13,34 W. Percobaan dilakukan dengan 6 skenario, menghasilkan nilai Mean Absolute Percentage Error (MAPE) terkecil 1,22% saat 80% data digunakan untuk training dengan dua hidden layer. Hasil peramalan untuk tiga hari pengujian menunjukkan nilai MAPE sebesar 9,43% (hari pertama), 6,11% (hari kedua), dan 21,33% (hari ketiga). Nilai MAPE rata-rata adalah 12,29%. Input data berupa radiasi matahari, suhu, dan kelembapan dari empat minggu sebelumnya digunakan untuk melatih model NN.
Downloads
References
L. Hu et al., “Ultra-Short-Term Solar PV Power Forecasting Method Based on Frequency-Domain Decomposition and Deep Learning,” 2020 IEEE Ind. Appl. Soc. Annu. Meet. IAS 2020, 2020, doi: 10.1109/IAS44978.2020.9334889.
M. M. Hasan Shawon, S. Akter, M. K. Islam, S. Ahmed, and M. M. Rahman, “Forecasting PV panel output using prophet time series machine learning model,” IEEE Reg. 10 Annu. Int. Conf. Proceedings/TENCON, vol. 2020-Novem, pp. 1141–1144, 2020, doi: 10.1109/TENCON50793.2020.9293751.
H. Chai et al., “Convolutional Auto-encoder Based Sky Image Prediction Model for Minutely Solar PV Power Forecasting,” 2020 IEEE Ind. Appl. Soc. Annu. Meet. IAS 2020, pp. 0–6, 2020, doi: 10.1109/IAS44978.2020.9334923.
S. Theocharides, V. Venizelou, G. Makrides, and G. E. Georghiou, “Day-ahead Forecasting of Solar Power Output from Photovoltaic Systems Utilising Gradient Boosting Machines,” 2018 IEEE 7th World Conf. Photovolt. Energy Conversion, WCPEC 2018 - A Jt. Conf. 45th IEEE PVSC, 28th PVSEC 34th EU PVSEC, pp. 2371–2375, 2018, doi: 10.1109/PVSC.2018.8547375.
J. Shi et al., "Bayesian Optimization - LSTM Modeling and Time Frequency Correlation Mapping Based Probabilistic Forecasting of Ultra-short-term Photovoltaic Power Outputs," in IEEE Transactions on Industry Applications, vol. 60, no. 2, pp. 2422-2430, March-April 2024, doi: 10.1109/TIA.2023.3334700.
Y. Wang, J. Shi, Y. Ma, S. Wang, Z. Fu and J. Gao, "Ultra-short-term interval prediction model for photovoltaic power based on bayesian optimization", Proc. IEEE/IAS Ind. Commercial Power System Asia, pp. 1138-1144, 2022.
J. Shi, Y. Chen, X. Cheng, M. Yang and M. Wang, "Four-Stage Space-Time Hybrid Model For Distributed Photovoltaic Power Forecasting," in IEEE Transactions on Industry Applications, vol. 59, no. 1, pp. 1129-1138, Jan.-Feb. 2023, doi: 10.1109/TIA.2022.3205570.
B. Liu, J. Nowotarski, T. Hong and R. Weron, "Probabilistic load forecasting via quantile regression averaging on sister forecasts", IEEE Trans. Smart Grid, vol. 8, no. 2, pp. 730-737, Mar. 2017.
M. S. Hossain and H. Mahmood, "Short-term photovoltaic power forecasting using an LSTM neural network and synthetic weather forecast", IEEE Access, vol. 8, pp. 172524-172533, Sep. 2020.
A. Borré et al., "Machine fault detection using a hybrid CNN-LSTM attention-based model", Sensors, vol. 9, no. 23, Apr. 2023.
A. Mellit, A. M. Pavan and V. Lughi, "Deep learning neural networks for short-term photovoltaic power forecasting", Renewable Energy, vol. 172, pp. 276-288, Jul. 2021.
Khairunnisa, Prediksi Daya Pembangkit Listrik Pv Satu Hari Ke Depan Untuk Memudahkan Manajemen Energi Pada Sistem Menggunakan Neural Network. 2020.
K. Nisa’, L. Mahendra, H. Eko, and I. W. Farid, “PV Power Forecast Based on Artificial Neural Network at Indonesia Shopping Mall PV Rooftop,” 2021 Int. Conf. Adv. Mechatronics, Intell. Manuf. Ind. Autom. ICAMIMIA 2021 - Proceeding, pp. 335–338, 2021, doi: 10.1109/ICAMIMIA54022.2021.9807694.
H. T. Yang, C. M. Huang, Y. C. Huang, and Y. S. Pai, “A weather-based hybrid method for 1-day ahead hourly forecasting of PV power output,” IEEE Trans. Sustain. Energy, vol. 5, no. 3, pp. 917–926, 2014, doi: 10.1109/TSTE.2014.2313600.
N. Jittratorn, G. W. Chang, and G. Y. Li, “A Hybrid Method for Hour-ahead PV Output Forecast with Historical Data Clustering,” Proc. - 2022 IET Int. Conf. Eng. Technol. Appl. IET-ICETA 2022, pp. 1–2, 2022, doi: 10.1109/IET-ICETA56553.2022.9971576.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Imam Wahyudi Farid, Joko Priambodo, Irfan Hamzah Fansuri, Lucky Putri Rahayu, Faiza Alif Fakhrina
This work is licensed under a Creative Commons Attribution 4.0 International License.
