Modeling and application of Czochralski silicon single crystal growth process using hybrid model of data-driven and mechanism-based methodologies

Ren, Jun-Chao; Liu, Ding; Wan, Yin

doi:10.1016/j.jprocont.2021.06.002

Cited by 30 publications

(10 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fourier heat conduction equation (heat equation) is an important PDE, which describes how the temperature in a region changes with time. For example, in the process of growing monocrystalline silicon by the Czochralski method, a stable and suitable thermal field is the key to growing a perfect monocrystalline silicon, so accurate simulation of the thermal field is an important research [37,46]. This paper considers the heat conduction of monocrystalline silicon part.…”

Section: Heat Conduction Equationmentioning

confidence: 99%

An Adaptive Physics-Informed Neural Network with Two-Stage Learning Strategy to Solve Partial Differential Equations

Shi¹,

Liu²,

null³

et al. 2023

NMTMA

View full text Add to dashboard Cite

Physics-Informed Neural Network (PINN) represents a new approach to solve Partial Differential Equations (PDEs). PINNs aim to solve PDEs by integrating governing equations and the initial/boundary conditions (I/BCs) into a loss function. However, the imbalance of the loss function caused by parameter settings usually makes it difficult for PINNs to converge, e.g. because they fall into local optima. In other words, the presence of balanced PDE loss, initial loss and boundary loss may be critical for the convergence. In addition, existing PINNs are not able to reveal the hidden errors caused by non-convergent boundaries and conduction errors caused by the PDE near the boundaries. Overall, these problems have made PINN-based methods of limited use on practical situations. In this paper, we propose a novel physics-informed neural network, i.e. an adaptive physics-informed neural network with a two-stage training process. Our algorithm adds spatio-temporal coefficient and PDE balance parameter to the loss function, and solve PDEs using a two-stage training process: pre-training and formal training. The pre-training step ensures the convergence of boundary loss, whereas the formal training process completes the solution of PDE by balancing various loss functions. In order to verify the performance of our method, we consider the imbalanced heat conduction and Helmholtz equations often appearing in practical situations. The Klein-Gordon equation, which is widely used to compare performance, reveals that our method is able to reduce the hidden errors. Experimental results confirm that our algorithm can effectively and accurately solve models with unbalanced loss function, hidden errors and conduction errors.

show abstract

Section: Heat Conduction Equationmentioning

confidence: 99%

An Adaptive Physics-Informed Neural Network with Two-Stage Learning Strategy to Solve Partial Differential Equations

Shi¹,

Liu²,

null³

et al. 2023

NMTMA

View full text Add to dashboard Cite

show abstract

“…Currently, monocrystalline silicon is manly produced by the Czochralski method (80-90%) [36,37], in which purified polycrystalline silicon is melted in a crucible from highly pure quartz in a high-vacuum furnace. This method is used to grow the largest single crystals.…”

Section: Mono-and Polysilicon Ingots Productionmentioning

confidence: 99%

Integration of Kazakhstan Technologies for Silicon and Monosilane Production with the Suitable World Practices for the Production of Solar Cells and Panels

et al. 2022

View full text Add to dashboard Cite

In this review article, the state of the art of the complete processing chain in the production of solar photo-electric modules from raw materials (quartzites, quartz sand) is detailed. In particular, the silicon and silane production technologies of the Institute of Physics and Technology of Almaty, Kazakhstan, can become part of an expansive technologies chain. Such integration could present a number of benefits in comparison with the analogs, including less environmental pressure and increased safety. The combination of innovative production technologies of highly effective solar cells and modules with competitive production technologies of solar-grade silicon and silane constitutes a basis for the creation of an industrial cluster in the field of silicon solar photo energy with a complete vertically integrated production cycle.

show abstract

“…The greater the sintering temperature it is possible the faster the crystal formation process [25]. The size of the temperature also affects the shape and size of the gap and also affects the structure of crystal growth [26]. Sintering temperatures can be determined from thermal experiments such as DTA, DTG, and DSC.…”

Section: Sintering Processmentioning

confidence: 99%

Effect of Dredging Rate, Sintering Time, Cooling Media on Corrosion Results of Carbon Steel Dredging

Irawan¹

2021

ijseit

View full text Add to dashboard Cite

Metal reinforcement that has an impact on improving mechanical properties can occur in various ways, including by strain hardening mechanisms, solid solutions, second phase, prespitations, disperse, grain refinement and texture. In recent years, another method has been developed to obtain good mechanical properties without adding alloy elements, namely by the method of comprehensive plastic deformation (Severe Plastic Deformation). The purpose of this research is to conduct a study of rolling using medium carbon steel with a treatment temperature above recrystallization, rolling rate, sintering time, cooling media as an effort to develop and design new innovations about steel structures being rolled. The study of developing rolling techniques that can be applied in industry and can be applied as learning material / media in higher education requires structured research stages. The best impact value from this temper process is a temperature of 550 ° C for 1 hour where the impact value is 1.58 J / mm² with air conditioning media. In the rolling process above the recrystallization temperature 800 ° C deformation 20% the impact value is 1.7 J / mm² where the toughness indicates a ductile fracture while the deformation is 10% the impact value

show abstract

Modeling and application of Czochralski silicon single crystal growth process using hybrid model of data-driven and mechanism-based methodologies

Cited by 30 publications

References 29 publications

An Adaptive Physics-Informed Neural Network with Two-Stage Learning Strategy to Solve Partial Differential Equations

An Adaptive Physics-Informed Neural Network with Two-Stage Learning Strategy to Solve Partial Differential Equations

Integration of Kazakhstan Technologies for Silicon and Monosilane Production with the Suitable World Practices for the Production of Solar Cells and Panels

Effect of Dredging Rate, Sintering Time, Cooling Media on Corrosion Results of Carbon Steel Dredging

Contact Info

Product

Resources

About