Authors:
Masaya Iwata
;
Yuji Kasai
;
Eiichi Takahashi
and
Masahiro Murakawa
Affiliation:
National Institute of Advanced Industrial Science and Technology (AIST), Japan
Keyword(s):
Photovoltaic Modules, Monitoring, Malfunction Detection, Maintenance, Power Line Communications.
Related
Ontology
Subjects/Areas/Topics:
Embedded Sensor Networks
;
Energy and Economy
;
Energy Monitoring
;
Energy-Aware Systems and Technologies
;
Sustainable Computing and Communications
Abstract:
Although photovoltaic (PV) modules occasionally fail, it is difficult to identify which module is malfunctioning. In order to detect malfunctioning PV modules, we have developed a malfunction detection method for individual PV modules by continuously monitoring their data. This method can automatically identify a malfunctioning module where output power declines at an early stage. Thus, the method provides faster and more accurate detection of malfunctions. Moreover, the method considerably reduces workloads for maintenance personnel because it eliminates the need for conventional inspection procedures to identify a malfunctioning module. A feature of the method is the utilization of two kinds of information among the PV modules, namely, spatial and temporal correlations, to distinguish between generation declines due to some malfunction and those due to climate conditions. To confirm the effectiveness of the method, we conducted a malfunction-detection experiment with actual data fr
om our PV module monitoring system which we have already implemented. The experiment used 24 PV modules installed within the monitoring system, and simulated a malfunction by covering 10% of a module. The system was able to detect the period of the simulated malfunction, which confirms the effectiveness of the method.
(More)