Analyses of Efficiency Degradation and Electroluminescence Imaging in Crystalline-Si and CdTe Thin-Film PV Modules via Accelerated Lifecycle Testing

Abstract

The efficiency degradation of photovoltaic (PV) modules is a critical issue that affects the performance and economic viability of solar energy systems. In general, there are two primary methods for observing PV aging behavior: in-situ test yard studies and laboratory chamber-based simulations. Field studies offer the advantage of monitoring the performance of a centralized, outdoor installation of PV modules under actual environmental exposure over real time, but the interpretation of power efficiency data is heavily dependent on the availability and quality of test-condition data. In contrast, laboratory studies simulate important environmental stresses to potentially mimic the aging process, and by adopting accelerated testing conditions, it can increase the degradation rate under any preset weather profile. Although IEC 61215 and IEC 61646 serve as the present design qualifications and type approval for crystalline-Si and thin-film modules, they do not fully account for degradation patterns that may occur in specific local climates and system setups, and these standards do not necessarily distinguish between products that may have varying lifetimes.This study aimed to combine the realistic property of test yard studies and the accelerating manner of chamber-based experiments, in order to illustrate the usage of a full-scale, industrial-standard environmental chamber to assess in-situ performance in an Accelerated Lifecycle (ALC) testing fashion under realistic weather profiles specific to Southern Arizona climates. A direct comparison of power conversion performance between fielded modules and a chambered panel of similar PV technology was first performed to validate the usage of the environmental degradation chamber. Upon several test yard upgrades - including important temperature sensors, pyranometers, data loggers etc. - initial results showed that the generated output power and raw conversion efficiency of the fielded module and the selected panel of similar technology inside the chamber responded accordingly to temperature variation and showed a close resemblance of temperature response between them. Their power performance over a period of 12 months also revealed a comparable reduction in relative efficiency (2.5% and 2.4%, respectively). This not only highlighted the importance of accurate and reliable data collection on climate and module characteristics at the test yard, but further demonstrated the feasibility of migrating an outdoor scenario into a finely-controlled indoor experimental research. By closely matching the local climate data collected at the Tucson Electric Power (TEP) test yard, the application of precise realistic weekly weather conditions in the test chamber provided an opportunity for a direct comparison of in-situ PV accelerated degradation studies. In this paper, five (labeled as A, B, C, D, E) commercially available PV modules of three different technologies (one monocrystalline-Si, three polycrystalline-Si, and one thin-film PV) were subjected to a 12-month ALC testing inside the Envirotronics chamber, and their temperature- and time-dependent contributions to efficiency degradation were quantitatively investigated. The temperature-dependent de-rating effects were individually characterized by performing a temperature sweep after a solar exposure of 500 hours, and the experimentally measured temperature de-rating coefficients was proven to be in close agreement with the manufacture’s rating. Removal of such effect further enabled the computation of time-dependent efficiency degradation, highlighting significant differences in their efficiency aging behaviors and hence their expected service lifetime. Visual inspection and EL imaging were performed to examine potential catastrophic defects in the packaging materials and solar cells. It is interesting to note that visually only the studied monocrystalline-Si panel (Module A) displayed a visible (but slight) hazing on the module surface in isolated regions, but no apparent scattering that might impact the irradiance level reaching the cell junction was confirmed; the polycrystalline-Si and thin-film PVs, on the other hand, remained seemingly new and did not exhibit any visual defects. For Modules C to E, raw electroluminescence images were taken in the middle and at the end of each simulated month using an Andor iKon-M Infra-red camera, and all of their images appeared to be darker at the end of the one-year equivalent ALC cycles, indicating a reduction in the electrical conductivity of the solar cells. For one of the tested polycrystalline-Si panels (Module D, which was a half-cut panel), alternating aging behavior was observed between the two half planes, suggesting a possible mismatch of loads or/and cell qualities. In MATLAB, image processing was performed and pixel intensities were collected. Two analyzing techniques were attempted on the active cell areas to quantify variation seen in the EL: horizontal line scan and cell intensity histograms. While the line scan method required an additional installment of a reference light source to normalize month-to-month intensity variation, the cell histogram approach minimized the error without a calibration source as it relied on area integration of many extracted cells. The study also investigated the relationship between efficiency degradation and electroluminescence (EL) imaging. It was evidential to see that the EL intensity reduction of two (of the three) imaged panels and their efficiency degradation were in good agreement: for Module C, the normalized intensity reduced by 2.5%, while the relative efficiency degraded by 2.4%; for Module D, both intensity and efficiency showed a 5.6% decrease. The results of this study provide significant contributions to our understanding of the factors that contribute to degradation of PV modules, emphasizing the importance of information accuracy and data completeness at the test yards, and validating the feasibility of adopting an integrated, full-scale environmental chamber as an effective alternative to real-time field observations. Using this approach, researchers can control the environmental conditions to a high degree to replicate local weather profiles while facilitating PV module aging tests at a rapid pace and subsequently leading to more effective mitigation strategies for a more reliable, resilient and sustainable PV system. Overall, these insights can have significant technical and economic implications for advancing the state-of-the-art in the design and deployment of solar energy systems.