Optimizing Fixture Layout And Clamping Force Simultaneously Using Genetic Algorithm

This thesis keeps a close eye on the development of advanced manufacture technique and makes a study on fixture optimization. Based on finite element method, this thesis explores a feasible method that can optimize fixture layout and clamping forces simultaneously.Fixture is a tool that ensures part quality in machining and its design is an important step in the process of manufacture. The current design of fixture basically relies on the experience of the designer. Nowadays, the requirement for workpiece precision is higher and higher, consequently, the traditional design means can't adapt the demand of production. Fixture optimization is a new fixture design method that develops with the extensive application of optimization techniques. It includes fixture layout optimization and clamping force optimization. Most researchers study only one of both or study them separately. So, this thesis is focused on optimizing fixture layout and clamping forces simultaneously.This work aims at optimizing fixture design from the viewpoint of reducing workpiece deformation during machining. The deformations in consideration include contact deformation and elastic deformation. Contact can be usually found in workpiece fixture system and it has much impact on machining quality. Contact is a nonlinear problem that can only be solved iteratively; sometimes, the iteration process will not converge and affects the convergence of optimization. So, contact is often simplified in most cases. Here, contact is replaced by springs whose deformation is used to simulate contact deformation of workpiece and fixture element.The parameterization of physical problems is the base of optimization. The deformation of workpiece is very complicated, so this work chooses finite element method as an analyzing tool. Consequently, how to parameterize the finite element model of workpiece fixture system becomes a part of this work. Fixture optimization differs from structure optimization because its finite element model is difficult to parameterize. Here, a driver is developed to carry out parameterization.Fixture optimization is a complicated optimization problem that contains multi design variables, multi design field and so on. Usual optimization algorithm can't find the optimal solution, so, this work chooses genetic algorithm to solve the problem. Genetic algorithm is robust, self-adaptive, global optimal and so on. It is able to solve problems that contain multi variables, multi-objectives, multi-fields with poorconnectivity.The proposed fixture optimization method can optimize fixture layout and clamping forces simultaneously. Numerical results show that this method is an efficient fixture optimization method that can improve fixture quality greatly.