Article
Automation & Control Systems
Zhuwen Yan, Henan Bu, Changzhou Hu, Bo Pang, Hongyu Lyu
Summary: In this paper, a rolling force prediction method based on IQGA-WNN ensemble learning is proposed to further improve the calculation accuracy of rolling force in the FGC process of tandem cold rolling. The traditional QGA is improved by quantum variation and is used to optimize the initial parameters of the network to enhance the prediction ability of WNN. The improved WNNs are used as base learners for ensemble learning and effectively integrated through the bagging algorithm, resulting in improved accuracy and stability of the rolling process.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Sang-Kon Lee, Kyung-Hun Lee
Summary: This study focused on designing die groove profile for cold pilger rolling process to predict rolling force in manufacturing seamless pipes. The theoretical calculations and experimental data verification indicated the successful application of the design method in controlling tube shape and tolerance accurately.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Manufacturing
Yu Wang, Changsheng Li, Lianggui Peng, Ruida An, Xin Jin
Summary: This paper develops a prediction model based on CNN to accurately predict strip flatness under different conditions. Data preprocessing methods like the isolated forest algorithm and data folding technique were utilized. The model achieved high accuracy by modifying the loss function and using an Inception module as the basic network structure.
JOURNAL OF MANUFACTURING PROCESSES
(2021)
Article
Automation & Control Systems
Yu Zhang, Wei Yu, Zhicheng Cheng, Qingwu Cai
Summary: The temperature gradient rolling (GTR) method effectively addresses the issue of uneven deformation of surface and core in the rolling of extra-thick plate. However, there has been limited research on rolling force prediction methods specifically for GTR.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Automation & Control Systems
Wenying Zeng, Jinkuan Wang, Yan Zhang, Yinghua Han, Qiang Zhao
Summary: This paper presents a model-free controller based on Deep Deterministic Policy Gradient (DDPG) for thickness and tension control in the unsteady rolling process of cold rolling. By utilizing strategies such as dividing state space variables with the mechanism model, defining reward function and state normalization, the DDPG controller copes with random disturbance and complex uncertainties, achieving better accuracy, stability, and rapidity than the proportional integral (PI) controller.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Metallurgy & Metallurgical Engineering
Pengfei Wang, Shuren Jin, Xu Li, Huagui Huang, Haifeng Wang, Dianhua Zhang, Wentian Li, Yulin Yao
Summary: The study proposes an accurate acquisition approach of flatness actuator efficiency (FAE) based on the combination of simulation modeling and data-driven modeling for flatness control of cold-rolled strip. The method involves constructing a 3D rolling feature space model, establishing a 3D finite element simulation model, developing an online optimization model, and proposing a linear output prediction model to achieve precise acquisition of FAE under any rolling conditions. The data analysis of the production process shows that the model is adaptive and accurate, capable of realizing high precision flatness control.
STEEL RESEARCH INTERNATIONAL
(2022)
Article
Automation & Control Systems
Da-Wei Zhang, Chao Zhang, Chong Tian, Sheng-Dun Zhao
Summary: This paper studies the forming characteristics of the thread cold rolling process with round dies. The results show that metal flow and plastic deformation mainly occur in the superficial area near the thread profile. The radial load is the main load of rolling dies and the characteristics of metal flow, stress distribution, and strain distribution in the thread profile are significantly different.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Automation & Control Systems
Dong Chen, Rui Zhang, Zhenlei Li, Yunjie Li, Guo Yuan
Summary: The paper proposes a temperature distribution prediction method based on recurrent neural network, fully considering the dynamic characteristics of variable-velocity rolling. By evaluating the temperature distribution prediction performance of the model with different recurrent cells and time steps, the results show that the proposed model can achieve temperature distribution prediction. The model based on bi-LSTM and 48 timesteps has the highest determination coefficient value of 0.976, the lowest root mean square error of 8.03, and a mean absolute error of 5.7.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Automation & Control Systems
Lixin Wei, Bohao Zhai, Hao Sun, Ziyu Hu, Zhiwei Zhao
Summary: In the cold tandem rolling process, the accuracy of rolling force prediction directly affects product quality and yield. This study proposes an ensemble just-in-time-learning modeling method based on multi-weighted similarity measures to address the limitations of fixed prediction models and single similarity measures.
Article
Metallurgy & Metallurgical Engineering
Cui Xiying, Wei Baomin, Wang Jianhui, Bai Xiaoshuai, Bai Zhenhua
Summary: In this study, an optimization model for deformation resistance during the cold rolling process in the hot rolling process was established using analytical methods and big data regression. The accuracy of the model was validated through actual data and field tests, and the application of the research results greatly improved the quality and shape of the cold rolled strip.
IRONMAKING & STEELMAKING
(2023)
Article
Automation & Control Systems
Yue Huang, Xiaomin Zhou, Zhiying Gao
Summary: This paper proposes a radial basis function neural network based on VBGM-RBF to predict the thickness of cold-rolled thin strip, which shows high accuracy and generalization performance in predicting the thickness of cold-rolled strips, and can be well-applied to the production of thin strip steel.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Computer Science, Theory & Methods
Shuqing Gong, Zhenyuan Guo, Shiping Wen
Summary: This paper focuses on the synchronization problem of T-S fuzzy memristive neural networks with time delay. A delay-independent nonlinear fuzzy control is designed, and two kinds of finite-time synchronization criteria are obtained under this control. The settling time is also estimated. A numerical simulation example is provided to demonstrate the effectiveness and feasibility of the theoretical results, and an application in the pseudorandom number generator is presented.
FUZZY SETS AND SYSTEMS
(2023)
Article
Multidisciplinary Sciences
John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Zidek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, Jonas Adler, Trevor Back, Stig Petersen, David Reiman, Ellen Clancy, Michal Zielinski, Martin Steinegger, Michalina Pacholska, Tamas Berghammer, Sebastian Bodenstein, David Silver, Oriol Vinyals, Andrew W. Senior, Koray Kavukcuoglu, Pushmeet Kohli, Demis Hassabis
Summary: Proteins are essential for life, and accurate prediction of their structures is a crucial research problem. Current experimental methods are time-consuming, highlighting the need for accurate computational approaches to address the gap in structural coverage. Despite recent progress, existing methods fall short of atomic accuracy in protein structure prediction.
Article
Metallurgy & Metallurgical Engineering
Ze-dong Wu, Xiao-chen Wang, Quan Yang, Dong Xu, Jian-wei Zhao, Jing-dong Li, Shu-zong Yan
Summary: The traditional rolling force model in tandem cold rolling mills does not consider the actual size and mechanical properties of the incoming material, leading to a mismatch between the deformation resistance setting and the actual state of the material, affecting the accuracy of the rolling force. This study developed an inverse calculation method to obtain the actual deformation resistance and established a support vector regression (SVR) model based on the cross-process dataset. The GWO algorithm was used to optimize the SVR model, resulting in improved accuracy and quality of the rolling process.
JOURNAL OF IRON AND STEEL RESEARCH INTERNATIONAL
(2023)
Article
Computer Science, Artificial Intelligence
Ning Wang, Witold Pedrycz, Wen Yao, Xiaoqian Chen, Yong Zhao
Summary: This article proposes a new network approach, called disjunctive fuzzy neural networks (DJFNNs), for implementing Takagi-Sugeno fuzzy models. The proposed DJFNN involves a novel network architecture and greedy learning algorithm, which overcomes the curse of dimensionality and forms more interpretable models. Experimental results confirm the effectiveness of the DJFNN in terms of accuracy, interpretability, and computational cost.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)