Paper
2 November 2023 Research on diameter control of Czochralski Silicon Single Crystal based on PSO parameter tuning for RNN-PFDL-MFAC
Shengzhe Xu, Dedong Gao, Shan Wang, Haohao Wu, Haixin Lin, Wenyong Zhang
Author Affiliations +
Proceedings Volume 12919, International Conference on Electronic Materials and Information Engineering (EMIE 2023); 1291912 (2023) https://doi.org/10.1117/12.3010748
Event: 3rd International Conference on Electronic Materials and Information Engineering (EMIE 2023), 2023, Guangzhou,, China
Abstract
This paper addresses the issues of large diameter fluctuations and wire breakage in the Cz-si monocrystal. A real-time diameter control scheme for Cz-si monocrystal pulling, found on particle swarm optimization (PSO) parameter tuning for RNN-PFDL-MFAC, is proposed to achieve accurate diameter control. By analyzing the control parameters that affect the diameter of the Cz-si monocrystal, a three-input single-output RNN diameter prediction model is established for the isothermal stage of crystal growth. A real-time diameter control scheme for Cz-si monocrystal pulling is determined based on the prediction model, the PFDL data model, and the MFAC algorithm. The optimal values of hyperparameters in the control scheme are determined using the particle swarm optimization algorithm. Finally, the validity and stability of the control project are demonstrated by control simulation.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Shengzhe Xu, Dedong Gao, Shan Wang, Haohao Wu, Haixin Lin, and Wenyong Zhang "Research on diameter control of Czochralski Silicon Single Crystal based on PSO parameter tuning for RNN-PFDL-MFAC", Proc. SPIE 12919, International Conference on Electronic Materials and Information Engineering (EMIE 2023), 1291912 (2 November 2023); https://doi.org/10.1117/12.3010748
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Crystals

Silicon

Particle swarm optimization

Data modeling

Correlation coefficients

Control systems

Neural networks

Back to Top