International collaboration framework for the calculation of performance loss rates: Data quality, benchmarks, and trends (towards a uniform methodology)

Self Cite

This article presents an initial performance analysis of a database of photovoltaic (PV) system performance time series collected within the European funded COST Action PEARL PV. The database contains monitoring data of over 8400 PV systems with accompanying metadata. The PV plants are small residential systems, primarily installed in Europe, with a high density in Belgium. In this initial study, the annual average performance ratio, the annual energy yield, and the performance loss rate of the systems are determined and evaluated. The systems have an average lifetime of 30.5 months. The annual mean performance ratio across all systems is 76.7% and the average yield is 954.9 kWh/kWp per year. The performance loss rate is calculated using three different statistical approaches and one irradiance data source. Average performance losses between −0.74%/year and −0.86%/year are calculated depending on the used approach. Furthermore, certain weather-dependent correlations are detected, such as decreasing performance ratio and increasing yield values with increasing irradiation. This study is a stepping-stone for further populating the present database, lessons learnt for handling large amounts of PV performance data, and carrying out performance studies of PV system fleets installed across Europe.

Section: A Methodologymentioning

confidence: 99%

Section: Experimental Setup-data Descriptionmentioning

confidence: 99%

Performance Analysis and Degradation of a Large Fleet of PV Systems

Lindig

Ascencio‐Vásquez

Leloux

et al. 2021

Self Cite

“…Besides knowing the "real" Rd values and change-point positions, synthetic data also have the advantage of being independent of sensor drift, temperature uncertainty, soiling, maintenance issues etc., that may affect the accuracy and/or uncertainty of the calculations. Typically, when Rd or PLR methodologies are compared using field data, the "true" value is considered to be the mean of all estimates; assuming that this is the closest to the "real" value [8]. However, caution is recommended in the presence of outliers, which can dominate the mean value; in this case, the median should be used.…”

Section: A Generation Of Synthetic Datasetsmentioning

confidence: 99%

“…However, it is known that PV performance fluctuates due to a number of seasonally related factors such as temperature [3], spectrum [4], soiling [5], or even abrupt changes caused by failures [6], etc. Therefore, although the statistical tools are available and relatively easy to use, it is inherently challenging to extract reproducible and accurate PV Rds [7], [8].…”

Section: Introductionmentioning

confidence: 99%

Comparative Analysis of Change-Point Techniques for Nonlinear Photovoltaic Performance Degradation Rate Estimations

Theristis

Livera

Micheli

et al. 2021

Self Cite

A linear performance drop is generally assumed during the photovoltaic (PV) lifetime. However, operational data demonstrate that the PV module degradation rate (Rd) is often nonlinear, which, if neglected, may increase the financial uncertainty. Although nonlinear behavior has been the subject of numerous publications, it was only recently that statistical models able to detect change-points and extract multiple Rd values from PV performance time-series were introduced. A comparative analysis of six open-source libraries, which can detect change-points and calculate nonlinear Rd, is presented in this article. Since the real Rd and change-point locations are unknown in field data, 960 synthetic datasets from six locations and two PV module technologies have been generated using different aggregation and normalization decisions and nonlinear degradation rate patterns. The results demonstrated that coarser temporal aggregation (i.e., monthly vs. weekly), temperature correction, and both PV module technologies and climates with lower seasonality can benefit the change-point detection and Rd extraction. This also raises a concern that statistical models typically deployed for Rd analysis may be highly climaticand technology-dependent. The comparative analysis of the six approaches demonstrated median mean absolute errors (MAE) ranging from 0.06 to 0.26%/year, given a maximum absolute Rd of 2.9%/year. The median MAE in change-point position detection varied from 3.5 months to 6 years.

“…Different algorithms and decomposition models have been tested and compared in the literature [9], [13], [14]. The results of an international benchmarking exercise of linear performance loss rate calculation methodologies suggest that averaging across multiple calculation methods provides the most consistent results across different PV systems [15]. Based on these and similar activities, it is visible that the current state of the art focuses on a linear PLR representation.…”

Section: Introductionmentioning

confidence: 99%

New PV Performance Loss Methodology Applying a Self-Regulated Multistep Algorithm

Lindig

Louwen

Moser

et al. 2021

Self Cite

Analyses of performance loss rates in photovoltaic (PV) systems are not yet standardized, and are typically carried out by a linear regression of the evolution of a certain performance metric (e.g., performance ratio, yield, etc.) over time. In this article, we propose a novel methodology of advanced PV system performance loss rate modeling applying a self-regulated multistep algorithm. The developed algorithm automatically detects the number and positions of breakpoints in nonlinear performance time series and divides the performance trend into an adequate number of linear segments. Instead of calculating one linear performance loss rate, given in percentage per year, as is common practice, multiple linear performance loss values are determined, depending on the trend of the time series and subsequently the number of breakpoints. The algorithm is fully automated. We have applied our methodology on data of an experimental PV installation in Bolzano/Italy, which consists of 26 different PV systems. The overall linear performance loss rate of the facility's crystalline silicon systems is between −0.5 and −1.3%/year, whereas the thin-film PV systems experience values between −0.6 and −2.4%/year. Based on our results, the algorithm appears to be stable and accurate. The methodology is to be used as a fast and automated check of PV systems in operation to detect anomalies affecting performance (in an early stage). By building up a large database of detected issues in the field this algorithm will enable us to better understand the performance evolution of different PV system types in varying climates.