New method for fault detection of PV panels in domestic applications

Abstract

This research is presented a new algorithm and method for fault detection of photovoltaic panels in Residential Photovoltaic System (RPS). Our proposed PV performance diagnosis system is consist of two parts, passive and active part. In the passive diagnosis part, residue value is generated with Model Base Fault Diagnosis (MBFD) method to observe clear alarm signal from arbitrary data. As well, statistical method such as Wald test technique is used to detect primary alarm in passive part. After observing the clear alarm signal, flash test is applied for investigating closer scrutiny of I-V characteristic to determine different type of faults (active part). This proposed Simple Flash Test Driver (FTD) device could be integrated in the modern inverters in order to increase the efficiency and reliability of the RPS without increasing any additional cost of materials and device in commercial DC/DC inverter.