DronEL - Fast and accurate inspection of large photovoltaic plants using aerial drone imaging

  • Séra, Dezso (PI (principal investigator))
  • Spataru, Sergiu Viorel (Projektdeltager)
  • Parikh, Harsh Rajesh (Projektdeltager)

Projektdetaljer

Beskrivelse

Fault detection and regular maintenance of both small and large PV installations, is important to secure the expected ROI, however the frequency and detail-level are limited by cost of manpower. Fast PV plant inspection, based on drone-mounted infrared (IR) cameras, reduces the inspection time and cost significantly, and is an emerging alternative to traditional methods. However, the detection accuracy of IR is limited by weather conditions, and to faults causing a sufficient increase in temperature.
This project – DronEL, will develop a fast and accurate automatic drone-based inspection system for PV plants that combines IR, luminescence (EL or PL) imaging, and visual images (VI). The system will be able to detect a wider range of PV panel failures: visual defects, hot-spots, solar cell cracks, potential-induced degradation, and more.
DronEL is a significant leap forward in PV plant inspection technology, combining the speed of drone-based IR inspection with the in-depth analysis of EL/PL. The project will carry out R&D activities in three main areas: Image acquisition and processing, Image interpretation – correlating images with known PV fault types, and drone control system and deployment. Accordingly, the project consists of the following main tasks: (i) R&D of suitable PL/EL imaging techniques. (ii) Integration and optimization of the imaging system on a drone (iii) Development of IR, PL/EL, and VI analysis for automatic fault detection and identification (iv) Integration and test of the drone system with the image analysis and automatic fault detection.
Kort titelDronEL
StatusAfsluttet
Effektiv start/slut dato01/01/201731/12/2019

Samarbejdspartnere

  • Technical University of Denmark
  • Sky-Watch A/S
  • Kenergy
  • Skive Kommune
  • SiCon • Silicon and PV consulting

Finansiering

  • Innovationsfonden: 19.993.065,00 kr.

Fingerprint

Udforsk forskningsemnerne, som dette projekt berører. Disse etiketter er oprettet på grundlag af de underliggende bevillinger/legater. Sammen danner de et unikt fingerprint.
  • Solar Cell Cracks and Finger Failure Detection Using Statistical Parameters of Electroluminescence Images and Machine Learning

    Parikh, H. R., Buratti, Y., Spataru, S. V., Villebro, F., A. dos Reis Benatto, G., Poulsen, P., Wendlandt, S., Kerekes, T., Séra, D. & Hameiri, Z., dec. 2020, I: Applied Sciences. 10, 24, s. 1-15 15 s.

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

    Åben adgang
    Fil
    24 Citationer (Scopus)
    31 Downloads (Pure)
  • A Photovoltaic Module Diagnostic Setup for Lock-in Electroluminescence Imaging

    Parikh, H. R., Spataru, S. V., Séra, D., A. dos Reis Benatto, G., Poulsen, P., Mantel, C., Forchhammer, S., Larsen, M., Frederiksen, K. H. B. & Vedde, J., jun. 2019, Proceedings of 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC). IEEE Press, s. 0538-0543 6 s. 0160-8371. (I E E E Photovoltaic Specialists Conference. Conference Record).

    Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

    Åben adgang
    Fil
    4 Citationer (Scopus)
    80 Downloads (Pure)
  • Machine learning prediction of defect types for electroluminescence images of photovoltaic panels

    Mantel, C., Villebro, F., Alves dos Reis Benatto, G., Parikh, H. R., Wendlandt, S., Hossain, K., Poulsen, P., Spataru, S. V., Séra, D. & Forchhammer, S., sep. 2019, Proceedings of SPIE Optical Engineering + Applications: Applications of Machine Learning. Zelinski, M. E., Taha, T. M., Howe, J., Awwal, A. A. S. & Iftekharuddin, K. M. (red.). SPIE - International Society for Optical Engineering, 9 s. 1113904. (Proceedings of SPIE, the International Society for Optical Engineering, Bind 11139).

    Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

    Åben adgang
    Fil
    22 Citationer (Scopus)
    193 Downloads (Pure)