Pipe bending methods and pipe bending systems

By generating mapping models and using machine learning methods, the problem of multiple trial adjustments in the tube bending process in existing technologies has been solved, achieving an efficient and precise tube bending process that can adapt to changes in various materials and states.

CN117415200BActive Publication Date: 2026-07-17HEXAGON INNOVATION CENTER LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HEXAGON INNOVATION CENTER LTD
Filing Date
2023-07-14
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies require multiple trials and adjustments during pipe bending to achieve the desired geometry, resulting in wasted time and resources, and making it difficult to provide accurate bending results across different pipe materials and properties.

Method used

By generating a mapping model, machine learning methods are used to determine the mapping from bending parameters to input parameters from 3D measurement data. The model is then trained using machine learning processes to predict processing input parameters, taking into account material, geometry, and machine condition, to achieve real-time correction.

Benefits of technology

It reduces the number of tests, improves the accuracy and efficiency of the pipe bending process, reduces the number of discarded parts, and adapts to changes in different pipe materials and target geometries.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117415200B_ABST
    Figure CN117415200B_ABST
Patent Text Reader

Abstract

This invention relates to a tube bending method and a tube bending system. A method and system for bending tubes using a tube bending machine are provided, wherein the values ​​of input parameters defining the processing steps of the tube bending machine are determined based on a mapping from bending parameters defining a target tube bending geometry to input parameters. The mapping is determined by a data-driven method, wherein a machine learning-based mapping model is fitted to tube bending machine processing data from an ongoing or previous bending process, thereby providing a machine learning correlation between the input parameters and the target bending parameters. For training the mapping model, the values ​​of the bending parameters and corresponding values ​​of the input parameters, as well as comparison information between the bending parameter values ​​and measured actual values ​​of the bending parameters produced by the tube bending process, are used to achieve the bending parameter values ​​within defined target tolerances.
Need to check novelty before this filing date? Find Prior Art