Online Health Monitoring of Silicon PV Panels by Converter-Based Impedance Spectroscopy: Panel-Level Equivalent Circuit Model and Health Feature Extraction

Converter-based impedance spectroscopy (IS) provides a promising tool to achieve online health monitoring of photovoltaic (PV) panels without additional equipment. However, to obtain a relatively complete IS spectrum of the PV panel, the frequency of the perturbation signal should reach at least 100 kHz, posing a great challenge for the current switching and sampling devices of power converters. Meanwhile, to correctly and clearly interpret the internal physical processes of the PV panel based on IS measurements, establishing an appropriate ac equivalent circuit model (ac-ECM) is essential. Nowadays, the existing ac-ECMs in PV domain are generally at the cell level, which exhibit inherent complexity for online implementation at the panel level. Moreover, the effect of bypass diodes is typically ignored, which affects the analysis under faulty conditions, such as partial shading. To address these issues, this work aims to propose a systematic health monitoring methodology for PV panels by converter-based IS. First, an appropriate panel-level ac-ECM is established by both simplifying the internal relaxation processes at the cell level and considering the existence of bypass diodes. Second, the proposed ac-ECM enables fitting the measured IS spectra over a lower frequency range of 1 Hz and 2 kHz, efficiently responding to variations in environmental conditions. Third, four health features are originally defined and extracted based on the proposed ac-ECM, including ln(R_p)/V_PV, ln(C_p)/V_PV to analyze the influence of temperature and output voltage, R_s + R_p to reflect the PV panel’s conductivity, and τ to describe the recombination process. Experimental verification has been performed under various operating conditions, including different temperatures, irradiance levels and partial shading percentages. The results indicate that the proposed methodology can efficiently characterize the different operating conditions of the PV panel.