Numerical Simulation And Optimization Of Hydroforming Of T-shaped Tube

The internal high-pressure forming(IHF) is an advanced technology of manufacturing light hollow structures, which is now widely applied in field of automobile and aerospace industry. The parts formed by IHF have many advantages, such as lightweight, high rigidity, less components, less subsequent machining time and shorter welding assembling cycle, etc. The mechanism of tube IHF is very complicated and the forming quality is related to many factors, among which, the loading paths of the process parameters and match of them are the biggest. Thus, how to design the loading paths and how to match them is a key issue in practical application of IHF.This dissertation, by means of finite element simulaton, compared thecontrol pattern in IHF of T-shaped tube based on the quality index, resultsshowed that force-control pattern could achieve a better quality, becausethe force-control pattern accords well with the rule of metal deformation.Further, the influence of loading paths of process parameters was studied,results of linear loading paths showed that enough axial feed is necessaryto obtain a certain amount of branch-tube, but too much axial feed willincrease the thickness severely; too large internal pressure or too smallback pressure will reduce the thickness severely or even cause burst; onthe contrary, too small internal pressure or too large back pressure willresult in a small amount of branch-tube, because the back pressure willplay a negative role on the growing of branch-tube; results of dual-linearloading paths showed that a proper dual-linear loading path could achievea much better quality. At last, the influence of friction coefficient andcorner radius were also studied, results showed that smaller frictioncoefficient or larger corner radius could produce a sounder T-shaped tube.As to the problem of process parameters match, by means of NeuralNetwork toolbox and Genetic Arithmetic toolbox in Matlab, the NeuralNetwork model of IHF of T-shaped tube was built, and the optimizationmatch of linear and dual-linear loading paths were also done. Resultsshowed that combined use of Neural Network and Genetic Arithmetic is a stable, feasible and intelligent optimization method, which can provide a new way to solve the process parameters optimization problem in IHF.