Abstract

Rotary draw bending (RDB) is a forming process that is commonly used to bend tubes with small wall thicknesses and small bending radii. One of the limitations of this process is the formation of wrinkles caused by compressive stress on the inner bend. In order to design the bending process without wrinkles and to determine the necessary process parameters, adjustment tests are required. Within this work, a fuzzy controller is to be developed which automatically prevents the formation of wrinkles and thus eliminates the need for time-consuming set-up tests to determine the necessary process parameters. The fuzzy controller is based on fuzzy set theory and fuzzy logic. In connection with a rule base it is possible to simulate the human decision process. A fuzzy controller is programmed based on a max-min controller, with the required rules resulting from previous bending tests. After the fuzzy controller has been implemented, it must be connected to the bending machine by suitable interfaces. The input values, which indicate wrinkles, are measured by sensors during the bending process and provide the controller with data. The fuzzy controller then uses the control base to specify the required control variables. After programming has been completed, practical validation tests were carried out. In the validation tests using different tube wall thicknesses and materials, a significant reduction of wrinkles is achieved. Bending of completely wrinkle-free tubes is also possible due to the automated finding of optimal tool settings. Using the fuzzy controller eliminates the need for costly adjustment bends, resulting in significant time and cost savings.

Highlights

  • Wrinkle formation is one of the process limits of rotary draw bending

  • The fuzzy controller is based on fuzzy set theory and fuzzy logic

  • The fuzzy controller is based on fuzzy set theory and fuzzy logic, which, coupled with a set of adjustment rules, can simulate human decision making [3]

Read more

Summary

Introduction

Wrinkle formation is one of the process limits of rotary draw bending. Predicting the formation of wrinkles on the inner bend of a tube is only possible approximately, since there are many influencing parameters to consider [1]. For this reason, bending processes in rotary draw bending are often based on the experience of skilled employees [2]. By means of test bends the process parameters are adjusted until the required forming is achieved. In order to avoid this adjustment work, the aim of this project was to develop a fuzzy controller that recognizes wrinkling during the bending process and counteracts it by readjusting control variables.

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call