A robust optimization design method for sheet metal roll forming and its application in roll forming circular cross-section pipe

Abstract

Sheet metal roll forming (SMRF) is affected by many factors, including forming speed, forming passes, roller diameter, and material properties. It is easy to produce quality defects, such as longitudinal bending, edge wave, and springback, which eventually lead to poor quality stability of SMRF. At present, SMRF design relies mainly on the experience of engineers and its technological parameters must be adjusted repeatedly through experiments, resulting in a long product design cycle and difficult quality stability control. This paper discusses a mathematical model for a robust optimization design for SMRF. Support vector machine (SVM) classification and support vector regression (SVR) are adopted to construct a model of the response surface and to separate feasible variables and infeasible variables. High-dimensional robust optimization problems in SMRF are solved. Using roll forming of a circular cross-section pipe as an example, the result shows that the roundness error standard deviation of the robust optimization design method is 62.2% less than the roundness error standard deviation of the deterministic optimization design method, along with significant improvement in reliability probability. The method is validated as effective in reducing quality defects, including longitudinal bending, edge wave, and springback. This robust optimization design method can improve the quality stability of SMRF.