A Method and System for Predicting the Minimum Coil Length in Hot-Rolled Finishing Based on Regression Algorithm
By constructing a minimum coil length prediction model based on a multiple linear regression algorithm, the problem of lack of quantitative standards in the hot rolling finishing process was solved, which improved the hot rolling yield and reduced production costs, while avoiding the accident of small strip coils turning over.
CN122133867APending Publication Date: 2026-06-02BAOSTEEL ZHANJIANG IRON & STEEL CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BAOSTEEL ZHANJIANG IRON & STEEL CO LTD
- Filing Date
- 2026-02-27
- Publication Date
- 2026-06-02
Smart Images
Figure CN122133867A_ABST
Abstract
This invention relates to the technical field of hot-rolled finishing production lines, and in particular discloses a method for predicting the minimum coil length in hot-rolled finishing based on a regression algorithm. The method includes the following steps: collecting state data of multiple strip coils during extreme condition transportation; analyzing the factors affecting the minimum coil length based on the state data to obtain a first variable; analyzing the interaction relationship between the first variables to obtain a second variable; constructing an initial minimum coil length prediction model, and training and optimizing the initial minimum coil length prediction model based on the second variable to obtain an optimized minimum coil length prediction model. This invention features clear cutting standards, improved hot-rolled yield, and reduced production costs.
Need to check novelty before this filing date? Find Prior Art