Contactless InlineIVMeasurement of Solar Cells Using an Empirical Model

Abstract

The current–voltage measurement is the most important measurement in solar cell quality control. As the contacting process of cells results in mechanical stress and consumes a significant amount of measurement time, this work presents an IV characterization based on contactless measurements only. An empirical model is introduced that can derive the full IV curve and IV parameters as the open‐circuit voltage, short‐circuit current density, fill factor, and efficiency. As a basis, a series of photoluminescence and contactless electroluminescence images and spectral reflectance measurements are used. An advantage of the model's convolutional neural network design lies in the semantic compression of local image structures across the input data. Within an ablation study, it is shown that the empirical model is well suited to combine these data sources, which is the optimal input configuration for contactless IV derivation. The accuracy, e.g., with an error in efficiency of 0.035 % abs and correlation of over 99%, is similar to comparing two contacting IV measurement devices. The contactless IV curves also have a close fit to their contacted counterparts. Within simulations on module level, it is demonstrated that contactless binning performs as well as contacting binning and does not result in any additional mismatch loss.

Shingle Cell IV Characterization Based on Spatially Resolved Host Cell Measurements

“…The problem could be reduced by using more data in which defects are partially occluded or by selectively covering known defects while training enforcing the model to stitch together partly occluded defects. Furthermore, it would be thinkable to use only PL images, possibly in combination with a partial shading [1], in order to measure series resistance properties as well. In addition, a GridTouch contact unit could be used covering less cell area.…”

“…The fixed voltage values were not set to be uniformly distributed, but were rather positioned to more densely map key regions around the short-circuit current (J sc ), maximum power point (MPP), and open-circuit voltage (V oc ). Also, as suggested in another work [1], the loss function was extended to the first and second derivatives as described in Equation (1).…”

“…In other works, it has been shown that CNNs are well suited to incorporate spatially resolved inline measurements into cell characterization. There exist various approaches to derive quality parameters of cells from EL or PL images [1][2][3][4][5][6][7][8][9][10][11]. Those, however, are still focusing on full cells only.…”

Predicting Loss Analysis from Luminescence Images in Si Solar Cells with Convolutional Neural Networks

“…[ 17 ] CNNs can also be used to derive I–V parameters from the photoluminescence (PL) image of the wafer [ 18 ] or a series of PL and contactless EL images as well as spectral reflectance measurements. [ 19 ] Semi‐supervised machine‐learning (ML) techniques are useful to derive spatially resolved quality parameters from the network activations trained from measurements. However, a study using ANN to infer power loss analysis in solar cells could be an important addition to available software tool, especially those that take spatially nonuniform device parameters into consideration.…”