Overview of non-destructive PV imaging technology
Ultraviolet fluorescence (UV-FL)
Best for: Cell crack detection; Limitation: Depends on aging time
Best for: Fast inspection; Limitation: Limited to visual defects
Best for: Various defects; Limitations: Experience needed for interpretation
Best for: Hot spots; Limitation: Only detects heat signatures
- EL is the most powerful, information-rich and mature tool for detection of PV module defects.
- Using EL for onsite inspection, we can accurately pinpoint and detect defective PV modules.
Quantitative Electroluminescence Analysis (QELA)
- EL is not widely used for field application due to some challenges
- EL images are dependent on the measurement systems.
- Require dark environment for image capturing.
- Analysis are subjective and qualitative.
- Quantitative ElectroLuminescence Analysis (QELA) enables the use of EL for outdoor application
- Calibrate EL image for various measurement system and environment. First EL round robin involving 17 labs around the world showed reduction in uncertainty of EL images captured between different labs from 15% to less than 3% using QELA technology.
- 100% comply with IEC TS 60904‑
- Quantitative analysis of EL images.
- Read more:
On top of QELA algorithms, we develop machine learning and artificial intelligence models to detect and analyse every module in PV plants and identify potential defects that might reduce the performance of the asset.
Drone coupled EL imaging
Our semi-autonomous drone inspection patent-pending technology performs inspection of large-scale solar assets in an easy, fast and user-friendly manner. The use of drone opens the possibility of site inspection for hard-to-reach places such as floating PV plants. Built-in auto search functions and constant communication with the ground power station enable the drone to fly to the required region automatically for measurements. An embedded processor adjusts camera parameters and lens focus to enable optimum EL imaging. With advanced telemetry, the module location can be accurately determined without the need of human intervention during post-processing.