基于物理信息神经网络加工的无缝钢管
By using a temperature prediction model based on a physical information neural network, the temperature field of the entire cross-section of a seamless steel pipe can be predicted in real time and the power of the heating zone of the annealing furnace can be adjusted. This solves the problem of limited temperature information acquisition in the annealing process and achieves precise control and energy saving.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA COAL SCIENCE & TECHNOLOGY (TIANJIN) ROCK FORMATION INTELLIGENT CONTROL TECHNOLOGY CO LTD
- Filing Date
- 2026-04-17
- Publication Date
- 2026-07-17
AI Technical Summary
In the current annealing process of seamless steel pipes, the acquisition of temperature information is limited, which makes it impossible to obtain the temperature field distribution inside the steel pipe and the entire cross-section in real time and without damage. This leads to the reliance on experience in setting process parameters, resulting in energy waste and inconsistent microstructure and properties.
A temperature prediction model based on physical information neural networks is adopted. The model is trained by minimizing the total loss function, which includes data loss terms and physical constraint loss terms. Combined with the partial differential equation of heat conduction, boundary conditions and initial conditions, the temperature field of the entire cross section of the seamless steel pipe is predicted in real time, and the heating zone power of the annealing furnace is adjusted according to the predicted temperature field.
It achieves accurate prediction and adaptive control of the temperature field across the entire cross-section of seamless steel pipe, reduces temperature control lag, avoids local overheating or underheating, improves product microstructure consistency and quality reliability, and reduces energy waste.
Smart Images
Figure CN122044266B_ABST
Abstract
Citation Information
Patent Citations
KR20240167544A