Dynamics of backsheet-driven insulation issues

doi:10.1016/j.solmat.2023.112398

Dynamics of backsheet-driven insulation issues

2023-08-01

3

PRE

AI

C

Claudia Buerhop‐Lutz *

DOI：10.1016/j.solmat.2023.112398

原文链接

原文求助

摘要

En 中文

To reap in the full benefit of high-efficiency solar cells and modules, the quality of the polymer encapsulant and backsheet (BS) materials is essential. Recent field studies show that early degrading backsheets cause safety issues and inverter shutdowns, resulting in yield losses. The dynamic development of degradation is important but is rarely studied because of the lack of proper datasets with meaningful and sufficient data. We studied inverter data of solar panels with a combined capacity of about 1 MWp for more than nine years including ground impedance (GI), and we labelled BS-types using our in-house identification method. Every inverter was connected to modules with exclusively a single BS-type, namely three-multilayer BSs denoted by their outer air layer: PA (polyamide), FC (fluorinated coating), and PVDF (polyvinylidene fluoride). We present, for the first time, an analysis of the degradation dynamics of GI, using a trained Gaussian Process Regression model. Using this model, we derived key degradation parameters, namely GI loss rates and time points for the onset of degradation. We found that PVDF BSs are associated with very low GI loss rates, while loss rates for PA are threefold higher with an onset at 6.6 year of operation for PA. FC BS-related loss rates were humidity dependent and two times higher those of PA; the onset was determined at 4.9 year of operation.

Keyword:

Backsheet

Degradation rate

Insulation resistance

Machine learning

Gaussian Process Regression

Dynamics of backsheet-driven insulation issues

摘要

期刊

机构

引用论文