Gray-box modeling of 300 mm diameter Czochralski single-crystal Si production process

Kato, Shota; Kim, Sanghong; Kano, Manabu; Fujiwara, T.; Mizuta, Masahiko

doi:10.1016/j.jcrysgro.2020.125929

Cited by 9 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fig. 3 shows the structure of the GB model [12]. The GB model consists of two models: a statistical model and an energy transfer, hydrodynamic and geometrical (EHG) model.…”

Section: Gray-box Model Of Cz Processmentioning

confidence: 99%

“…The statistical model is then constructed by Gaussian Process Regression [15]. Details of the statistical model are described by Kato et al [12].…”

Section: Gray-box Model Of Cz Processmentioning

confidence: 99%

“…Although these initial values affect the prediction accuracy of the CVs, they are not measured in the industrial CZ process. The CVs can be predicted with high accuracy by calculating the initial values using the method proposed by Kato et al [12].…”

Section: Gray-box Model Of Cz Processmentioning

confidence: 99%

See 2 more Smart Citations

Gray-box model-based predictive control of Czochralski process

Kato

Kim

Mizuta

et al. 2021

Journal of Crystal Growth

Self Cite

View full text Add to dashboard Cite

“…Fig. 3 shows the structure of the GB model [12]. The GB model consists of two models: a statistical model and an energy transfer, hydrodynamic and geometrical (EHG) model.…”

Section: Gray-box Model Of Cz Processmentioning

confidence: 99%

“…The statistical model is then constructed by Gaussian Process Regression [15]. Details of the statistical model are described by Kato et al [12].…”

Section: Gray-box Model Of Cz Processmentioning

confidence: 99%

See 1 more Smart Citation

Gray-box model-based predictive control of Czochralski process

Kato

Kim

Mizuta

et al. 2021

Journal of Crystal Growth

Self Cite

View full text Add to dashboard Cite

“…Therefore, when operating conditions fluctuate, especially after long-term operation, the accuracy of the model may deviate [ 12 ]. In response to this issue, researchers have proposed a new adaptive nonlinear predictive control method for crystal diameter based on hybrid integration modeling [ 13 , 14 ]. This method combines engineering practice and decomposes the original data into several sub-sequences using the wavelet packet decomposition method.…”

Section: Introductionmentioning

confidence: 99%

Research on Classification Algorithm of Silicon Single-Crystal Growth Temperature Gradient Trend Based on Multi-Level Feature Fusion

Liu,

Mu,

Zhang

et al. 2024

Sensors

View full text Add to dashboard Cite

In the process of silicon single-crystal preparation, the timely identification and adjustment of abnormal conditions are crucial. Failure to promptly detect and resolve issues may result in a substandard silicon crystal product quality or even crystal pulling failure. Therefore, the early identification of abnormal furnace conditions is essential for ensuring the preparation of perfect silicon single crystals. Additionally, since the thermal field is the fundamental driving force for stable crystal growth and the primary assurance of crystal quality, this paper proposes a silicon single-crystal growth temperature gradient trend classification algorithm based on multi-level feature fusion. The aim is to accurately identify temperature gradient changes during silicon crystal growth, in order to promptly react to early growth failures and ensure the stable growth of high-quality silicon single crystals to meet industrial production requirements. The algorithm first divides the temperature gradient trend into reasonable categories based on expert knowledge and qualitative analysis methods. Then, it fuses the original features of actual production data, shallow features extracted based on statistical information, and deep features extracted through deep learning. During the fusion process, the algorithm considers the impact of different features on the target variable and calculates mutual information based on the difference between information entropy and conditional entropy, ultimately using mutual information for feature weighting. Subsequently, the fused multi-level feature vectors and their corresponding trend labels are input into a Deep Belief Network (DBN) model to capture process dynamics and classify trend changes. Finally, the experimental results demonstrate that the proposed algorithm can effectively predict the changing trend of thermal field temperature gradients. The introduction of this algorithm will help improve the accuracy of fault trend prediction in silicon single-crystal preparation, thereby minimizing product quality issues and production interruptions caused by abnormal conditions.

show abstract

“…One of the most important key raw materials in photovoltaic industry is single-crystal silicon (monocrystalline silicon). With the burgeoning photovoltaic industry, there is an increasing demand in high purity single-crystal silicon ingots, to satisfy the need of photovoltaic cell (Kato et al, 2021). Now, large diameter and high-quality single-crystal silicon ingots are mainly produced from the Czochralski (CZ) process (Qi et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Study on Waste Heat-Driven Refrigeration System for Energy Saving and Fast Cooling of Dust Collector in Monocrystalline Silicon Manufacture

Hou

2021

Front. Energy Res.

View full text Add to dashboard Cite

Single-crystal silicon is key raw material in photovoltaic industry. In its manufacture, silicon monoxide dust, a byproduct, is collected under vacuum environment. To clean the dust collector, air is recharged into the collector, reacting with the dust and causing very high temperature. Collector components may be damaged. It also takes several hours to cool down. In this paper, a cooling system based on ejection refrigeration cycle is proposed, which collects the reaction heat and simultaneously controls the collector temperature around 100°C. Then, it is driven by stored waste reaction heat and cools down the dust to a lower temperature. The designed cooling system, employing a 9.7972 m2 fin-tube heat exchanger, can simultaneously meet the cooling load of four dust collectors with 330L/S capacity. By a thermodynamic model established in this work, performance analysis is carried out. Generating temperature around 73°C and evaporating temperature around 6°C are recommended for system operation. Results also show the cooling system is able to provide 3270 kJ cooling energy that is needed by the collector, for fast cooling down the dust no longer than 620 s. It is about 92% shorter than the time of current collector, indicating the cooling system is effective and feasible.

show abstract

Gray-box modeling of 300 mm diameter Czochralski single-crystal Si production process

Cited by 9 publications

References 16 publications

Gray-box model-based predictive control of Czochralski process

Gray-box model-based predictive control of Czochralski process

Research on Classification Algorithm of Silicon Single-Crystal Growth Temperature Gradient Trend Based on Multi-Level Feature Fusion

Study on Waste Heat-Driven Refrigeration System for Energy Saving and Fast Cooling of Dust Collector in Monocrystalline Silicon Manufacture

Contact Info

Product

Resources

About