Integrasi Deteksi, Segmentasi, dan Klasifikasi Material Atap Berbasis YOLOv8s-seg untuk Estimasi Potensi PLTS Atap di Desa Adat Penglipuran, Bali

Amanda Austin Herlambang; I Nyoman Satya Kumara; I Wayan Gede Ariastina; Putu Arya Mertasana; I Wayan Sukerayasa

doi:10.51454/decode.v6i3.1834

Authors

Amanda Austin Herlambang Magister Teknik Elektro Universitas Udayana
I Nyoman Satya Kumara Magister Teknik Elektro Universitas Udayana
I Wayan Gede Ariastina Magister Teknik Elektro Universitas Udayana
Putu Arya Mertasana Teknik Elektro Universitas Udayana
I Wayan Sukerayasa Teknik Elektro Universitas Udayana

DOI:

https://doi.org/10.51454/decode.v6i3.1834

Keywords:

Desa Penglipuran, Deteksi Objek, PLTS, YOLOv8

Abstract

Peningkatan pemanfaatan energi terbarukan mendorong pengembangan Pembangkit Listrik Tenaga Surya (PLTS) atap untuk mendukung target Net Zero Emission (NZE) Indonesia 2060 dan Bali 2045. Penelitian ini mengimplementasikan deep learning YOLOv8 untuk mendeteksi, melakukan segmentasi, mengklasifikasikan material, dan mengestimasi luas atap sebagai dasar penentuan potensi PLTS di Desa Adat Penglipuran, Bali. Kontribusi utama penelitian ini adalah mengintegrasikan deteksi, segmentasi, klasifikasi material atap, dan estimasi luas dalam satu pendekatan berbasis citra udara untuk menghasilkan penilaian potensi teknis PLTS atap pada skala wilayah, yang tidak hanya berfokus pada deteksi objek atap. Menggunakan citra udara beresolusi 5 cm/piksel dan 26 citra pelatihan, model menghasilkan mAP@50 sebesar 0,665 untuk deteksi dan 0,658 untuk segmentasi. Atap beton dan metal ditetapkan layak, sedangkan atap asbes, sirap bambu, dan ilalang dikecualikan. Luas atap layak mencapai 25.190,46 m² dari 76 pekarangan, dengan potensi kapasitas PLTS sebesar 4,54–5,00 MWp dan produksi energi 6,20–6,82 GWh/tahun. Potensi tersebut setara 62,8–69,0 kali kebutuhan listrik tahunan 76 rumah tangga sebesar 98,8 MWh/tahun. Hasil menunjukkan YOLOv8s-seg efektif untuk rapid asesment potensi teknis PLTS atap pada skala wilayah.

References

Ahmed, M. S., Arman, M. S., Tasnim, N., Imran, M. H., Sammak, M. H., & Bhuiyan, T. (2025). Rooftops Detection With Yolov8 From Aerial Imagery And A Brief Review On Rooftop Photovoltaic Potential Assessment. IAES International Journal of Artificial Intelligence, 14(3), 2282–2290. https://doi.org/10.11591/ijai.v14.i3.pp2282-2290

Ahmed, S., Ali, A., & D’Angola, A. (2024). A Review of Renewable Energy Communities: Concepts, Scope, Progress, Challenges, and Recommendations. Sustainability 2024, 16(5), 1749; https://doi.org/10.3390/su16051749

Amir, A. A., Sukman, Lihara, A. D., & Nabil, M. (2026). Sistem Analisis Kinerja Lalu Lintas Berbasis Deep Learning Dengan Arsitektur You Only Look Once (YOLO). Decode: Jurnal Pendidikan Teknologi Informasi, 6(1), 179–190. https://doi.org/10.51454/decode.v6i1.1503

Biswas, S., Acharjee, S., Ali, A., & Chaudhari, S. S. (2023). YOLOv8 based Traffic Signal Detection in Indian Road. 2023 7th International Conference on Electronics, Materials Engineering and Nano-Technology, IEMENTech 2023, 1-6. https://doi.org/10.1109/IEMENTech60402.2023.10423520

Castellanos, S., Sunter, D. A., & Kammen, D. M. (2017). Rooftop Solar Photovoltaic Potential In Cities: How Scalable Are Assessment Approaches?. Environmental Research Letters, 12(12), 125005. https://doi.org/10.1088/1748-9326/aa7857

Direktorat Jenderal Survei Dan Pemetaan Pertahanan Dan Ruang. (2026). Model Dasar Pertahanan Retrieved January 3, 2026. https://petadasar.atrbpn.go.id/

Fakhraian, E., Alier, M., Dalmau, F. V., Nameni, A., & Guerrero, J. C. (2021). The urban rooftop photovoltaic potential determination. Sustainability (Switzerland), 13(13). https://doi.org/10.3390/su13137447

Global Solar Atlas. (2026). Global Solar Atlas Retrieved June 25, 2026. https://globalsolaratlas.info/map?s=-8.422177,115.358908&m=site&c=-8.422177, 115.358908, 11

Hamzaoglu, A., Erduman, A., & Kırçay, A. (2025). Deep Learning-Based Rooftop PV Detection and Techno Economic Feasibility for Sustainable Urban Energy Planning. Sustainability (Switzerland), 17(15), 6853. https://doi.org/10.3390/su17156853

Hu, M., Liu, Z., Huang, Y., Wei, M., & Yuan, B. (2023). Estimation of Rooftop Solar Photovoltaic Potential Based on High-Resolution Images and Digital Surface Models. Buildings, 13(11), 2686. https://doi.org/10.3390/buildings13112686

IESR. (2025). Bali’s Roadmap to Net Zero Emissions by 2045 in the Electricity Sector. https://iesr.or.id/en/pustaka/balis-roadmap-to-net-zero-emissions-by-2045-in-the-electricity-sector/

Jinko Solar. (2024). 66HL4M-(V) 610-635 Watt Anti-PID guarantee SMBB Technology Mechanical Load Enhanced Durability Against Extreme Environment N-type Technology N-type modules with Tunnel Oxide Passivating. www.jinkosolar.com

Kim, J., Bae, H., Kang, H., & Lee, S. G. (2021). Cnn Algorithm For Roof Detection And Material Classification In Satellite Images. Electronics (Switzerland), 10(13), 1592. https://doi.org/10.3390/electronics10131592

Kusuma, N. T. R., Suarjaya, I. M. A. D., & Vihikan, W. O. (2024). Implementasi Metode Yolov8 Pada Mobile Apps Untuk Klasifikasi Kain Endek Bali. Decode: Jurnal Pendidikan Teknologi Informasi, 4(3), 787–797. https://doi.org/10.51454/decode.v4i3.610

Li, Q., Krapf, S., Mou, L., Shi, Y., & Zhu, X. X. (2024). Deep Learning-Based Framework For City-Scale Rooftop Solar Potential Estimation By Considering Roof Superstructures. Applied Energy, 374, 123839. https://doi.org/10.1016/j.apenergy.2024.123839

Longi Solar. (2024). Longi Datasheet Hi-Mo X10 LR7-72HVH 655 - 670M.

Marcelleno, D. J., & Putra, M. P. K. (2025). Performance Evaluation Of Yolov8 In Real-Time Vehicle Detection In Various Environmental Conditions. Jurnal Teknik Informatika (Jutif), 6(1), 269–279. https://doi.org/10.52436/1.jutif.2025.6.1.3916

Qi, Q., Zhao, J., Tan, Z., Tao, K., Zhang, X., & Tian, Y. (2024). Development Assessment Of Regional Rooftop Photovoltaics Based On Remote Sensing And Deep Learning. Applied Energy, 375, 124172. https://doi.org/10.1016/j.apenergy.2024.124172

Silva, A. S., de Azevedo, A. R., Neto, F. H. de A. M., & Ferreira da Silva, P. H. (2025). YOLOv8-based Model For Automatic Detection Of Residential Roof Damage. Revista Alconpat, 15(1), 50–63. https://doi.org/10.21041/ra.v15i1.783

Solar Co Ltd. (2025). CanadianSolar Datasheet 435 W ~ 465 W TOPHiKu6. www.csisolar.com/na

Trina Solar. (2024). Vertex S Dual Glass N type i-TOPCon Module Outstanding Visual Appearance Dual-glass Design, High Reliability. www.trinasolar.com

Ultralytics. (2026). COCO Dataset Retrieved July 9, 2026. https://docs.ultralytics.com/datasets/detect/coco

Ultralytics. (2023). Explore Ultralytics YOLOv8 . https://docs.ultralytics.com/models/yolov8#overview

Xu, J., Guo, Z., Yu, Q., Dong, K., Tan, H., Zhang, H., & Yan, J. (2025). Spatiotemporal Feature Encoded Deep Learning Method For Rooftop PV Potential Assessment. Applied Energy, 394, 126171. https://doi.org/10.1016/j.apenergy.2025.126171

Zeng, J., Yang, M., Tang, X., Guan, X., & Ma, T. (2025). Intelligent Segmentation Of Urban Building Roofs And Solar Energy Potential Estimation For Photovoltaic Applications. Journal of Imaging, 11(10), 334. https://doi.org/10.3390/jimaging11100334

Zhang, Y., He, W., Hu, J., Zhou, C., Ren, B., Luo, H., Tian, Z., & Liu, W. (2025). Assessment of Urban Rooftop Photovoltaic Potential Based on Deep Learning: A Case Study of the Central Urban Area of Wuhan. Buildings, 15(15), 2607. https://doi.org/10.3390/buildings15152607

Zhou, Z., Li, Y., Guo, Z., Yan, Z., & Chow, M.-Y. (2024). A White-Box Deep-Learning Method for Electrical Energy System Modeling Based on Kolmogorov-Arnold Network. IEEE Transactions on Industrial Informatics, 22(1) 107-118. https://doi.org/https://doi.org/10.48550/arXiv.2409.08044