Laser Induced Luminescence Characterization of Mechanically Stressed PV Cells

Abstract

Electroluminescence (EL) and photoluminescence (PL) imaging are powerful solar cell and panel characterization techniques that are capable of detecting early stage solar cell failures and degradation with high sensitivity and level of detail required for proper preventive maintenance. The electrical contacts needed during measurement pose a challenge for many inspection scenarios, both for EL and PL. Light or Laser induced luminescence (LIL) is a contactless way to generate spatially resolved images of PV panels that highlight poorly connected cell areas, such as finger failures, areas with high series resistance or disconnected cell cracks. In this study, we analyze the characteristics of the LIL scan imaging method in detecting cracks in mono-Si cells, comparing single frames with reconstructed contactless EL images built with a full LIL scan image stack. A comparison between contactless EL generated by LIL and conventional contact EL is made, and finally an estimation of the required laser intensity for outdoor LIL is provided, in correlation to known contact EL performance during nighttime and daylight conditions.

A review of imaging methods for detection of photoluminescence in field-installed photovoltaic modules

“…This resulted in two images. An InGaAs camera was used with a 950 nm long-pass filter which lets through only the PL signal above 950 nm and blocks the reflected laser light below 950 nm [28]. The challenge with this imaging method is, as mentioned above in case of LED modules, that the intensity of the excitation light might be inhomogeneous [29].…”

“…The different approaches result in five types of acquired PL images. Imaging with laser excitation [28], LED modules excitation [16,17] as well as under sunlight excitation using ultranarrow band-pass filters [38] results in PL images consisting only of the PL signal. The stippled orange lines indicate that the approach with ultranarrow band-pass filters can be combined with instantaneous control of the OP in order to acquire images under different level of current extraction.…”

“…It was shown that the laser's scanning orientation, perpendicular or parallel to busbars, plays an important role as to how an isolated region appears on the PL image in question. Illuminating a line of a cell results in the possibility to detect finger interruptions which are difficult to detect if one illuminates the entire cell area during imaging [28,46]. Cracks and isolated regions (i.e.…”

Evaluation of PV Panel Power Loss Using Gabor Filter Bank

“…Akram et al performed automatic detection of PV module defects in EL images [2]. dos Reis Benatto et al, in their study, focused on obtaining EL images quickly in daylight and presented a drone-based system that could acquire EL images at 120 frames per second [11]. In their study, Dhimish and Holmes presented the development of a new technique used to increase the detection of microcracks in solar cells [12].…”

“…However, since thermal and RGB images can be taken from the system installed in the field, they can be processed faster. Taking into account the employed methods; [1,4,5,7,11,13,15,20,21] studies used image processing methods, while [18][19][20]22] studies tried to detect faulty panels via artificial intelligence algorithms. In image processing algorithms, the preprocessing is done by the programmer, while deep learning algorithms try to detect errors on the raw image.…”