Presentation Details
  1. Home
  2. Presentation List
  3. Presentation Details
Prediction Model of Rolling Mill Stiffness based on Particle Swarm Optimization-Back Propagation (PSO-BP) Neural Network
ID：65 View Protection：ATTENDEE Updated Time：2025-11-10 11:33:13 Hits：153 Oral Presentation

Start Time：2025-11-23 10:50（Asia/Shanghai）

Duration：20min

Session：S2 Parallel Session 2 » S2-2Parallel Session 2-23 AM

Presentation File

Tips: Only the registered participant can access the file. Please sign in first.

Abstract
The stiffness of the rolling mill roll system plays a crucial role in determining the shape of the hot rolled strip steel plate. It is directly influenced by the spatial position of the rolling mill roll system. This study establishes a numerical model of rolling mill stiffness through the spatial position of the roll system, and validates its accuracy using field measured data. Through the developed model, relevant data of the cross state and stiffness of the roll system are obtained. Focusing on the hot rolling mill, we propose a deep neural network (DNN) model called particle swarm optimization back propagation neural network (PSO-BP). The particle swarm optimization (PSO) algorithm is improved based on practical experience and simulation analysis: an adjustment factor α  is introduced to enforce physical constraints (influence weight of backup roll > work roll, drive side > operating side). Additionally, a nonlinear mapping mathematical model is established to quantify the relationship between the cross state of the roll system axis induced by the wear of the mill stand liner, and the mill stiffness. Our results demonstrate that the PSO-optimized BP model has higher prediction accuracy than conventional BP, genetic algorithm-BP (GA-BP), and PSO-BP without the adjustment factor.
Keywords
Mill stiffness, Prediction model, Support vector regression, Finite element analysis
Speaker
Qiu Bitao
‌Professor of engineer Wuhan University of Science and Technology

Submission Author
Qiu Bitao Wuhan University of Science and Technology
Dan Binbin Wuhan University of Science and Technology
Ruan Jinhua Wuhan University of Science and Technology
Guo Wanfu Wuhan University of Science and Technology
Submit Comment
Verify Code Change Another
All Comments
Important Date

  • Conference Date

    Nov 21

    2025

    to

    Nov 23

    2025

  • Oct 20 2025

    Draft paper submission deadline

  • Dec 08 2025

    Registration deadline

Sponsored By
IEEE Instrumentation and Measurement Society
South China University of Technology
Organized By
South China University of Technology
Contact Information
Login to view full contact details
Previous Conferences