Fault Diagnosis of Photovoltaic Panels Using Dynamic Current-Voltage Characteristics

• Published in: IEEE transactions on power electronics Vol. 31; no. 2; pp. 1588 - 1599
• Main Authors: Wenguan Wang, Liu, Alex Chun-For, Chung, Henry Shu-Hung, Lau, Ricky Wing-Hong, Jun Zhang, Lo, Alan Wai-Lun
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2016; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Capacitance; Computer architecture; Current-voltage characteristics; Devices; Dynamics; Electric currents; Failure; Fault diagnosis; Junctions; Panels; Parameter estimation; Photovoltaic cells; Photovoltaic systems; Prototypes; reliability; Solar cells; solar panels; Fault diagnosis; solar panels; photovoltaic (PV) systems; reliability
• ISSN: 0885-8993; 1941-0107
• DOI: 10.1109/TPEL.2015.2424079

A fault diagnosis technique for photovoltaic (PV) panels is presented. While a PV system is sampling the terminal voltage and current of its connected panel for tracking the maximum power point of the panels, the proposed technique utilizes the sampled data to estimate the intrinsic parameters of the panel simultaneously. Compared with the prior-art approach of using the static current-voltage characteristics, the proposed technique utilizes the dynamic current-voltage characteristics to determine the parameters. Apart from the fast parameter estimation, it also provides an in-depth understanding of the panel condition with the drift of the parameters. Several prototype devices with the proposed algorithm have been built. They are evaluated on a test bed with four 80-W panels, with two of them being healthy and the other two having different degrees of damage on the surfaces. Results reveal that the parameters of the cracked panels deviate significantly from their nominal values, giving a sign of panel failure. Furthermore, the device can communicate with and send the estimated parameters to the central control center over the panel cable via power line communication. The merits of this concept lie in its modularity, scalability, and remote fault diagnosis capability.