An integrative approach to spatial mapping of pressure distribution in microrolling

Abstract

Microrolling is an advanced manufacturing process for the mass production of miniaturized components. At the micrometer scale, small deviation in material thickness or roll misalignment can lead to significant variation of the pressure distribution across the product, resulting in dimensional inaccuracy of the imprinted texture. To address this problem, an embedded sensing system has been developed to monitor in situ the spatial pressure distribution across the tool–workpiece interface. A key element of the sensing system is a pressure reconstruction algorithm, which determines the deformation across the roll structure from a limited number of sensing points, and converts the data subsequently into pressure distribution across the workpiece. This paper investigates how backprojection and interpolation can be integrated synergistically to improve the accuracy and robustness of pressure reconstruction. Specifically, interpolation is performed as a pre-possessing step to increase the resolution in capacitance variation estimation and improve the robustness of back-projection in solving the inverse problem of capacitance–pressure conversion. Experimental study confirmed the performance of the developed method.