Simulation Of Multiple-step Incremental Roll-bending Forming Of Large-scale Sheet Metal With U Shape

Roll-bending is a common process method of sheet metal forming. The roll-bending with Single-step can only form the workpieces of simple profile, for the workpiece that having the semi-elliptical cross section, single-step roll-bending process is impossible to form, so we must choose the multiple-step incremental roll-bending processing method. First discrete the semi-elliptical cross section of the workpiece, transform the geometry information of the workpiece into the roll-bending parameters which consist of single-step, then control the sheet feed and the downward amount of upper roller in each single-step, and get high precision workpiece step by step. This process is suitable for forming a crane boom, boom of bridge, pressure piping, telescopic arm of pump truck, and other large size, small quantities products. It is able to significantly reduce product development time and production costs.During the roll-bending forming process of the sheet, plastic deformation occurs and the springback phenomenon is inevitable. The generation of springback reduces the geometric accuracy of the workpiece, affects future assembly work, and increases the correction step and new product development time. All of this restricts further promotion and application of incremental roll-bending process, especially for high-strength sheet machining large scale parts. The springback phenomenon during the incremental roll-bending process is very complex issue, many factors affect the springback and coupled with each other, showing a severe nonlinear. This situation makes it is difficult to model and forecast the springback of the roll-bending forming process. The analysis reports about incremental roll-bending forming with large scale sheet metal part involving springback are rare. So explore the incremental roll-bending process, research effective method of modeling, predicting and compensating springback, reduce the error of the parts got by roll-bending forming to obtain high-precision parts, are problems needed to solve urgent in the incremental roll-bending forming field.In this paper, based on the classic theory of the sheet metal forming, fully integrate the actual situation of roll-bending process, derive the formula of stress and strain, position and movement of the neutral layer, changing of the length during roll-bending forming. Determine the influence factors of the springback radius of sheet metal based on the orthogonal test and range analysis. Correct the geometric radius formula of roll-bending forming and establish a mathematical model of springback radius. Using BP neural network methods, obtain a high precision forecast model of the springback radius. Using software ABAQUS as a platform, In view of the semi-ellipse shape of workpiece, a geometric plan using fives arcs with different radius to approach the original shape in piecewise is proposed. On the basis of the programming, reasonable arrange the rolling steps and simplify their boundary conditions, using optimized process parameters to form a workpiece with semi-elliptical cross section, create a finite element analysis model of incremental roll-bending forming, and then solve the model.The results show that:using the mathematical model in this paper to predict springback radius of roll-bending, the error is relatively large, but the prediction model based on BP neural network has high accuracy and can be used to predict springback radius of roll-bending forming. It plays a very good guide to the actual roll-bending process. For the large scale workpiece with semi-elliptical shape, it can't obtain the high precision workpiece only by geometric programming. It need to optimize the roll-bending process according to the specific circumstances of each arc segment, adjust the ABAQUS parameter, reduce the effect of offset and the springback, eliminate the dislocation. After the optimization of process parameters, the maximum error of curvature radius of each segment is less than 6%, and meets the dimensional accuracy requirements. It provides an effective reference data and theoretical basis for further application of ABAQUS in the incremental roll-bending forming.