Study On Self-compensating Control Algorithm For Springback Of Tube Bending Machine

This paper is aimed at the S-YXGWK-200hydraulic bending machine equipmentdeveloped by the new platform. Original software platform to operating personnel accordingto the experience, relation curve between the tube shape and hydraulic cylinder feed input, butthis mode of operation is not intuitive and each batch of pipe to try bending, tube shape changealso need to bend test, according to the I in the factory’s actual knowledge, only for skilledolder workers each trial bending to waste5tube. Therefore, inefficient waste materials.To overcome this problem to realize bending process automation, high efficiency, developa set of reliable control system, to good engineering drawing "translated" into thecorresponding hydraulic cylinder feed curve; and due to the pipe of a rebound in the presence,we also choose a practical and effective springback solutions.To achieve this purpose, we in the software of AutoCAD for secondary development, theuseful information of the interception of AutoCAD file export DXF file and the tube shape ofengineering drawing information description. And because in actual use in the plant werecompared using NURBS curve to describe the tubular, therefore we also in-depth study splinecurve mathematical principles. In the completion of the tube shaped curve after a read, needaccording to the actual status of elbow, the curves of a series of changes and to get thecorresponding to the position of hydraulic cylinder feed, and form the feed rate curve is wewant to get the final curve.On the other hand, overcome the pipe of springback. We read and study, discussion of theresearch results of domestic and foreign scholars in this discovery:The pipe will rebound in bending deformation occurs, this is the main factor affecting thequality of the elbow. And the influence factors of springback are numerous, such as pipematerial, shape, size, wall thickness deformation. NC tube bending confidential development,inevitable to face the problem of springback compensation, in accordance with the tube type characteristics of effective springback compensation in order to make a fully automated CNCbending machine. But on the tube springback problem is a hard problem of scholars at homeand abroad. Currently a lot of research and experiments on springback of the pipe: from elasticplastic mechanical properties of elasto-plastic mechanics of six basic assumption (assumptionof continuity, uniform assumption, the assumption of isotropy, perfectly elastic, smalldeformation and without initial stress) solution of the rebound and the tube shape, materials,size and deformation, pipe wall thickness between function [1]; from Perspective of finiteelement analysis and the use of finite element analysis software, a lot of [2] of the magnitudeof springback prediction; using BP neural network learning sample and to predict the actualpipe [3] the relationship between type and rebound; application principle of minimumpotential energy analysis and springback prediction [4]. However, the basic assumption ofthese models require too much, applicable scope is very small: only limit a tube or smallradius bending or thin plate bending. So far, there is no a is widely recognized and accepted bythe theoretical model.Function relationship between in view of this of bold innovation, avoid tube type, wallthickness, pipe etc. many factors and rebound, the method of black box design, secondarymeasurement of rebound after the tube type, acquisition is transmitted to the computer, in abend in the compensation, the successive approximation, eventually generated to meet therequirements of a tube type.For data acquisition software for the hardware circuit, we in the lab built the actual circuitis verified; in the development of software for simulation trial bending. The experimentalresults show that the correctness of the hardware circuit and the software algorithm. And in thecar factory to install the software to the IPC test curve, obtained satisfactory results.