Title:	Utilizing System Efficiency Evaluations to Determine DC Output of PV Systems
Author(s):	B. Jaeckel, M. Pander, D. Daßler, S. Malik, J. Froebel
Keywords:	Degradation, Energy Yield, Prediction
Topic:	PV Systems Engineering, Integrated/Applied PV
Subtopic:	Operation, Performance and Maintenance of PV Systems
Event:	8th World Conference on Photovoltaic Energy Conversion
Session:	4BV.5.37
Pages:	1432 - 1435
ISBN:	3-936338-86-8
Paper DOI:	10.4229/WCPEC-82022-4BV.5.37
Price:	0,00 EUR
Document(s):	paper

Abstract/Summary:

---

A novel approach to investigate the state of health (SoH) of PV systems is presented. It utilizes simple monitoring data from the past minimum three years of the PV system. The data must include energy yield of the PV system, irradiance data and for good filtering logfiles to understand missing or flowed data sets. The available data should be at least on daily basis for trend charting of several years. The PV system itself consist of PV modules, cabling, and other balance of system components (BOS) including inverters. All those impact the herein used system efficiency SE evaluation. The level of detail obviously also depends on the number of individual measurement points (e.g. string, combiner box or inverter level) and can range from a few kWp to MWp. By applying a linear regression fit into monthly or annual data a robust system efficiency SE can be determined even if some data outliers exist. Depending on scope of work even different day profiles can be separated e.g. for clear and sunny days vs days low light levels. However, the main goal here is to determine current and past DC output power of the PV modules from monitoring data while applying some assumption such as the system consist of always the same BOS components, same orientation, and homogeneous degradation of modules. Those, at least partly, can be easily validated by flash test or electroluminescence (e.g. for testing for PID). The determination of DC output power requires reference points. Those can lay in the past for instance from system commissioning or / and can be from today (fielded modules are remeasured in laboratory). From that a reference data set for a particular system can be created with known DC power. Against this system a sister system nearby e.g. from a different technology or otherwise poorly performing system can be evaluated to detected possible system efficiency losses. This, for instance, can be applied to a c-Si system with a thin film sister system nearby (same irradiance profile) with very similar BOS (same non-PV module related losses), same orientation and known DC rating. Results of four different PV technologies will be exemplarily presented and evaluated by the introduced methodology within the paper. The applied methodology predicts current DC ratings of the PV modules precisely and can determine degradation rates.