Online pre-perception of forming state based on real-time measurement in spinning of thin-walled shell component

Abstract

Die-less spinning is an advance incremental forming process widely used for the manufacturing of thin-walled shell components. During the spinning, workpiece shape and forming state both change continuously, and the influence of process parameters on the forming results is also time-varying, which make it difficult to control the forming quality. To this end, this work develops a real-time measurement system and an online pre-perception method to provide technical support for dynamic control of the spinning process. Specifically, the real-time measurement system is constructed firstly by placing two laser profilers bilaterally at two sides of the workpiece. Based on the measuring data of two laser profilers, the profile data of workpiece cross-section is obtained by coordinate transformation, data denoising and correcting the error caused by inclined workpiece surface. Then, an algorithm is proposed to identify the critical geometric parameters (flange width, roller action radius, wall thickness and flange fluctuation degree) of workpiece shape from the profile data. In contrast to the measurement results by a three-coordinate measuring machine, the developed system presents a real-time measurement error less than 4 %. Moreover, an online pre-perception method of wall thickness and wrinkling defect is developed based on the real-time measured workpiece shape parameters. The online pre-perception of wall thickness and wrinkling defect presents high accuracy with the relative error less than 2.1 %. The above results indicate both the workpiece dimensions and spinning state can be well real-time measured and online pre-perceived, which can provide important foundation for the study of time-varying influence and dynamic control of the spinning process.