Photovoltaic fault detection algorithm based on theoretical curves modelling and fuzzy classification system

TLDR: A fault detection algorithm based on the analysis of the theoretical curves which describe the behavior of an existing PV system can accurately detect different faults occurring in the PV system, where the maximum detection accuracy of before considering the fuzzy logic system is equal to 95.27%.

About: This article is published in Energy.The article was published on 2017-12-01 and is currently open access. It has received 101 citations till now. The article focuses on the topics: Fault detection and isolation & Fuzzy classification.

Figures

TABLE 3 Comparative Results between the Proposed Algorithm and the One Presented in Ref. [4], Ref. [7] and Ref. [8]

Fig. 11. MPPT unit output power (A) Examined partial shading condition, (B) P-V curve including the output LMPP and GMPP obtained by the MPPT unit

Fig. 12. Theoretical detection limits vs. real-time long-term data measurements for one faulty, two faulty, three faulty and four faulty photovoltaic modules

Fig. 13. Out of region samples processed by the fuzzy logic classification system

Fig. 14. Theoretical curves vs. real time long term measured data. (A) Hot spot images taken from two different PV modules using FLIR thermal imaging camera, (B) Theoretical fault detection curves vs. measured data obtained from a PV module without hot spots, PV module contains only one hot spot and PV module contains two hot spots

Fig. 8. Theoretical vs. Measured output power during 5 different days

Frequently asked questions

Q1. What are the contributions in this paper?

9 This work proposes a fault detection algorithm based on the analysis of the theoretical curves which 10 describe the behaviour of an existing grid-connected photovoltaic ( GCPV ) plant. 13 Furthermore, a third order polynomial function is used to generate two detection limits ( high and low 14 limit ) for the VR and PR ratios obtained using LabVIEW simulation tool. Furthermore, 17 samples that lies out of the detection limits are processed by a fuzzy logic classification system which 18 consists of two inputs ( VR and PR ) and one output membership function.