Topology Optimization of Geometrically Nonlinear Compliant Constant-Force Mechanisms Considering Actual Output Displacement with a Meta-graph Strategy

Abstract

This study presents an optimal design procedure to design compliant constant-force mechanisms based on topology optimization A constant-force mechanism can generate a nearly constant output force over a range of input displacements without the need of additional sensors for output force control This study uses the density filter scheme to overcome the mesh dependence and checkerboard problems and makes use of parameterized projection function to reduce the existence of the gray elements Because of the nonlinear behavior of the compliant constant-force mechanisms geometrically nonlinear analysis is considered in finite element analysis The method of moving asymptotes is used to update design variables To enhance the stability of the optimal design procedure and the connectivity of topological results a composite objective function and a meta-graph strategy suitable for the SIMP method are proposed The numerical optimization problem of the composite objective function considering the actual output displacement is to minimize the sum of squares of errors Through the proposed optimal design procedure two types of compliant constant-force mechanisms and one type of the compliant constant-force gripper have been presented The effect of low physical density elements additional hyperelastic elements and the smoothing of topological edges are discussed in this study The prototype of the compliant constant-force gripper is manufactured by 3D printing using thermoplastic elastomer material The prototype has a constant output force in the input displacement region from 13 to 33 mm the value of the constant output force is 41 15 N and the average absolute error is 1 88% A three-finger constant-force gripper module is developed and installed on a four-axis robotic arm for grasping application Test results show the presented design is with a stable gripping ability

Date of Award	2020
Original language	English
Supervisor	Chih-Hsing Liu (Supervisor)