Detection and prediction of photovoltaic panels malfunctions

Abstract

This article discusses issues of photovoltaic panels malfunctions and failures. Several computerized detection methods are applied on 300 real commercial crystalline Si photovoltaic installations to identify origin, quantity and time behaviour of most common malfunctions. This paper focuses only on operational failures while malfunctions derivable from fabrication process and transport are not in account. Failures influence on VA characteristic is deeply discussed. Possible usage of computerized real time VA characteristic measurements for backward failure detection and identification is evaluated. The main contribution of this research is the application and verification of this method on 8 channel pilot system.