Deep‐learning‐based pipeline for module power prediction from electroluminescense measurements

Abstract

Automated inspection plays an important role in monitoring large-scale photovoltaic power plants. Commonly, electroluminescense measurements are used to identify various types of defects on solar modules, but have not been used to determine the power of a module. However, knowledge of the power at maximum power point is important as well, since drops in the power of a single module can affect the performance of an entire string. By now, this is commonly determined by measurements that require to discontact or even dismount the module, rendering a regular inspection of individual modules infeasible. In this work, we bridge the gap between electroluminescense measurements and the power determination of a module. We compile a large dataset of 719 electroluminescense measurements of modules at various stages of degradation, especially cell cracks and fractures, and the corresponding power at maximum power point. Here, we focus on inactive regions and cracks as the predominant type of defect. We set up a baseline regression model to predict the power from electroluminescense measurements with a mean absolute error (MAE) of 9.0 ± 8.4W P (4.0 ± 3.7%). Then, we show that deep learning can be used to train a model that performs significantly better (7.3 ± 6.5W P or 3.2 ± 2.7%) and propose a variant of class activation maps to obtain the per cell power loss, as predicted by the model. With this work, we aim to open a new research topic. Therefore, we publicly release the dataset, the code, and trained models to empower other researchers to compare against our results. Finally, we present a thorough evaluation of certain boundary conditions like the dataset size and an automated preprocessing pipeline for on-site measurements showing multiple modules at once.

Characterization of Electrical Parameters of Cracked Crystalline Silicon Solar Cells in Photovoltaic Modules

“…Furthermore, cracks and power loss in modules has been evaluated using light and dark I-V measurements over several stages of degradation [19]. Characterization of electrical parameters and statistical analysis of cracked PERC and Al-BSF cells in mini-modules was analyzed by Whitaker [20] and very recently, deep learning has been used to make power predictions from EL images of modules with solar cell cracks [21].…”

Identification of the Backsheet Type of Silicon Photovoltaic Modules from Encapsulant Fluorescence Images

“…Recent trends in global climate policy stimulated a dramatic growth of photovoltaic (PV) deployment with an expected level of tens of terawatts worldwide to be reached in one to two decades. , This enormous growth serves as a locomotive for the development of associate technologies in the production, monitoring, and recycling of PV modules. In particular, the expected TW scale of PV installations requires efficient high-throughput procedures for automated characterization of the status of deployed PV modules, their integrity, performance, and state of degradation. − The most promising technologies for the high-throughput characterization include different imaging techniques, such as thermography, − electroluminescence (EL), , and photoluminescence (PL) imaging, , which can be realized on automated moving/flying platforms and provide essential information on the module performance and type/population of defects.…”

“…In particular, the expected TW scale of PV installations requires efficient high-throughput procedures for automated characterization of the status of deployed PV modules, their integrity, performance, and state of degradation. 3−6 The most promising technologies for the high-throughput characterization include different imaging techniques, such as thermography, 7−10 electroluminescence (EL), 11,12 and photoluminescence (PL) imaging, 13,14 which can be realized on automated moving/ flying platforms and provide essential information on the module performance and type/population of defects.…”

Physics-Based Method for Generating Fully Synthetic IV Curve Training Datasets for Machine Learning Classification of PV Failures

“…Such machine learning techniques have shown great potential as diagnostic tools for PV systems, boasting high detection and diagnostic accuracy for a wide range of failure modes [8]. Typical input features for the machine learning models include 1-dimensional (1D) data such as environmental time-series data (irradiance, ambient temperature, module temperature, wind speed), electrical parameter data (voltage, current, power, energy) [3,15,16], and IV characteristics [17][18][19][20] or 2-dimensional (2D) data such as visual (Vis) images [21], electroluminescence (EL) images [22][23][24][25][26], infrared (IR) images [3,27,28], and full IV curves [29,30].…”