A survey of fault detection and identification methods for Photovoltaic systems based on I-V curves

Photovoltaic systems (PVS), like all energy production systems, must be monitored in order to be able to detect failures in near real-time so that they will maintain their performance to an optimum level, thus achieving the greatest possible reliability. There are several algorithms for identifying faults during the operation of a PVS based on I-V curves. These algorithms can be applied to PVS telemetry data either locally or remotely. Implementations can be simple like the recording and comparing measurements, but they can also be more advanced like the use of neural networks to detect PV faults. In the present paper, fault detection and identification methods based on I-V curves are presented. The discussed methods are selected according to the recent literature, are presented and analyzed so that their differences, advantages, and limits are pointed out.