Design investigation on 100 μm-thickness thin silicon PERC solar cells with assistance of machine learning

Zhu, Heng; Yan, Wensheng; Liu, Yiming; Hu, Die; Tu, Yiteng; Huang, Zhengtao; Tan, Xiaotian

doi:10.1016/j.mssp.2021.106198

Cited by 9 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the aforementioned reasons, to analyze the error performance of proposed studies, the calculation of MAE, MSE, RMSE, and R values are frequently used by many researchers. [15,20,23,27,29]…”

Section: Modeling Of I-v Characteristic Of Ni/n-gaas/in Sbdmentioning

confidence: 99%

See 1 more Smart Citation

Next‐Generation Application‐Based Artificial Intelligence in Modeling and Estimation for Ni/n‐GaAs/In Schottky Barrier Diode

Doğan

Koçkanat

2023

Physica Status Solidi (a)

View full text Add to dashboard Cite

Herein, for Ni/n-GaAs/In Schottky barrier diode, experimental measurement, modeling, data generation from the model, and parameter estimation processes are simultaneously carried out. In the experimental step, Ni/n-GaAs/In Schottky barrier diodes are fabricated and annealed from the temperature of 200 °C up to 600 °C with 100 °C steps. Current values are recorded by applying voltage to the diode contacts from À1 V up to 0.5 V. In the modeling step, 1503 experimental current-voltage data are used for 19 different regression models. For Adaptive Neuro Fuzzy System (ANFIS), when root mean square error, mean square error, mean absolute error, and coefficient of determination are calculated 6.0341e-07, 3.6410e-13, 2.3873e-07, and 0.9999 for training, they are obtained 5.8904e-07, 3.4697e-13, 2.3083e-07, and 0.9999 for testing. In the estimation step, the values of electrical parameters are estimated by using Mayfly algorithm. Estimations are performed for all annealing temperatures. In addition, current-voltage data for the annealing temperature of 350 °C are produced by the ANFIS model. Thus, a new-generation artificial intelligence application, that includes measurement, modeling, and estimation for the Ni/n-GaAs/In Schottky barrier diode with varying annealing temperatures, is realized and a new perspective is provided to researchers and practitioners.

show abstract

Section: Modeling Of I-v Characteristic Of Ni/n-gaas/in Sbdmentioning

confidence: 99%

“…Also, it is effectively used in the solution of many scientific problems. [11][12][13][14][15][16] Artificial intelligence is known the transfer of human abilities to computers with the aid of machine learning. At this point, machine learning plays an important role.…”

Section: Introductionmentioning

confidence: 99%

Next‐Generation Application‐Based Artificial Intelligence in Modeling and Estimation for Ni/n‐GaAs/In Schottky Barrier Diode

Doğan

Koçkanat

2023

Physica Status Solidi (a)

View full text Add to dashboard Cite

show abstract

“…The method was validated with experimental results, where the parameters obtained by the ML-based and DPSS approaches were shown to agree within an acceptable uncertainty range . ML-based regression methods for solar cell parameters to evaluate the impact of the thickness of different layers on the efficiency, , understand material properties, , and understand current–voltage curve analysis ,− were recently published. Regression tasks have also been performed on luminescence images. , Classification methods have mainly revolved around automated image analysis using deep learning algorithms such as convolutional neural networks (CNNs), where the objective is to classify defects − or identify their position in luminescence images. − …”

Section: Introductionmentioning

confidence: 99%

Section: ■ Introductionmentioning

confidence: 99%

Deep Learning Extraction of the Temperature-Dependent Parameters of Bulk Defects

Buratti¹,

Dick

Gia

et al. 2022

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Bulk defects in silicon solar cells are a key contributor to loss of efficiency. To detect and identify those defects, temperature-and injection-dependent lifetime spectroscopy is usually used, and the defect parameters are traditionally extracted using fitting methods to the Shockley−Read−Hall recombination statistics. In this study, we propose a deep learning-based extraction technique that is based on an alternative representation of the lifetime curves: lifetime mapping in the temperature and minority carrier concentration space. The deep learning approach successfully predicts all the defect parameters while addressing one of the main limitations of the traditional approach of locating the defect in the energy spectrum, which usually outputs two possible solutions. Furthermore, the approach is applied to temperature-dependent defect parameters where the traditional approach is not applicable, achieving satisfying levels of prediction of the defect parameters. Image representation and deep learning have the potential to bolster solar cell characterization techniques by extracting more insights from the characterization data.

show abstract

“…First generation monocrystalline and multicrystalline silicon technology still dominates the photovoltaic (PV) industry with a market share of over 90% due to high energy conversion efficiencies, constant price reduction and proven long-term reliability [Cui et al , 2022; National Renewable Energy Laboratory (NREL), 2020]. In PV manufacturing, 180 μm-thick crystalline silicon (cSi) wafers are used for fabrication of solar cells (Zhu et al , 2022). To date, the module price for monocrystalline (mono cSi) technology has reduced to less than US$0.25/Watt p [International Technology Roadmap for Photovoltaic (ITRPV), 2021].…”

Section: Introductionmentioning

confidence: 99%

Polyimide substrate textured by copper-seeding technique for enhanced light absorption in flexible black silicon

Omar

Abdulkadir

Hashim

et al. 2022

View full text Add to dashboard Cite

Purpose This paper aims to present investigation on textured polyimide (PI) substrate for enhanced light absorption in flexible black silicon (bSi). Design/methodology/approach Flexible bSi with thickness of 60 µm is used in this work. To texture the PI substrate, copper-seeding technique is used. A copper (Cu) layer with a thickness of 100 nm is deposited on PI substrate by sputtering. The substrate is then annealed at 400°C in air ambient for different durations of 60, 90 and 120 min. Findings With 90 min of annealing, root mean square roughness as large as 130 nm, peak angle of 24° and angle distribution of up to 87° are obtained. With this texturing condition, the flexible bSi exhibits maximum potential short-circuit current density (Jmax) of 40.33 mA/cm2, or 0.45 mA/cm2 higher compared to the flexible bSi on planar PI. The improvement is attributed to enhanced light scattering at the flexible bSi/textured PI interface. The findings from this work demonstrate that the optimization of the PI texturing via Cu-seeding process leads to an enhancement in the long wavelengths light absorption and potential Jmax in the flexible bSi absorber. Originality/value Demonstrated enhanced light absorption and potential Jmax in flexible bSi on textured PI substrate (compared to planar PI substrate) by Cu-seeding with different annealing durations.

show abstract

Design investigation on 100 μm-thickness thin silicon PERC solar cells with assistance of machine learning

Cited by 9 publications

References 20 publications

Next‐Generation Application‐Based Artificial Intelligence in Modeling and Estimation for Ni/n‐GaAs/In Schottky Barrier Diode

Next‐Generation Application‐Based Artificial Intelligence in Modeling and Estimation for Ni/n‐GaAs/In Schottky Barrier Diode

Deep Learning Extraction of the Temperature-Dependent Parameters of Bulk Defects

Polyimide substrate textured by copper-seeding technique for enhanced light absorption in flexible black silicon

Contact Info

Product

Resources

About