Design And Robust Optimization Of A Multistage Amplified Flexible Microgripper

Piezoelectric ceramic actuators have the advantages of high resolution,high output stiffness,and short response time,and are widely used in precision positioning and operation.However,their output displacement is about 0.1% of their own length,and they need to be used in conjunction with displacement amplification mechanisms.Therefore,this paper designs a multi-level amplification flexible micro gripper,which amplifies the output displacement of piezoelectric ceramics,and performs mechanical analysis on it.Based on the mechanical analysis,a robust design is carried out to enable it to output a larger magnification while ensuring its compact structure.On the one hand,it solves the shortcomings of traditional micro gripper design and expands its scope of use.On the other hand,it combines parameter and tolerance parallel design with process capability index,providing a new idea for robust design.The specific research work of this article includes:(1)Structure design of micro gripper.In view of the control difficulties and limitations brought by the traditional "single input-double output","double input-double output" is selected for synchronous control of clamping force and displacement.In the selection of flexure hinges,choose the blade flexure hinges with better deformation ability.In the displacement amplification part,the bridge amplification mechanism and the lever amplification mechanism are selected to form a two-stage amplification.The two bridge mechanisms are respectively driven by two piezoelectric ceramics to form a "double input".The lever mechanism is designed as two connecting rods in parallel to form a parallelogram mechanism,which can enlarge the displacement and realize parallel guidance.(2)Mechanical analysis and optimization of micro gripper.Firstly,the Euler-Bernoulli beam theory is used to carry out the static analysis of the micro gripper,calculate the magnification,and do not ignore the deformation of the connecting beam.The results are compared with the pseudo-rigid body method and the calculation method of ignoring the deformation of the connecting beam.The simulation results show that the Euler-Bernoulli beam theory,which does not ignore the deformation of the connecting beam,is more accurate in predicting the magnification.In addition,the Lagrange equation is used to establish the dynamic model of the micro gripper and analyze its dynamic characteristics.Secondly,the parameters of the microgripper are preliminarily optimized.The RBF neural network is used to establish the optimization model with the amplification factor and natural frequency as the response and the structural size as the parameters.The multi-island genetic algorithm is used to optimize the theoretical model of the micro-gripper.The optimized amplification factor and natural frequency are increased by 7.85% and 5.65% respectively.(3)Parallel design of parameters and tolerance considering process capability index.In view of the influence of tolerance on the optimization of parameters,a parallel design method of parameters and tolerance considering process capability index is proposed.Firstly,the key structural parameters are screened out through Spearman rank correlation coefficient to reduce the calculation amount of response surface modeling and reduce the test cost.The screened structural parameters are used for central composite test design and the mean and variance models of each quality characteristic are obtained from the data;Then the asymmetric mass loss function is constructed,and the fluctuation of the mean and variance of the mass characteristics is transformed into mass loss by Taylor expansion;Then fit the tolerance cost model according to the tolerance cost data of design variables;Finally,the asymmetric quality loss function and tolerance cost model are substituted into the process capability index.The optimal combination of design variables and their tolerances are obtained through the second generation genetic algorithm.Compared with other literatures,the optimization results show that the quality loss and manufacturing cost are reduced by 10%～20%,which not only saves costs,but also improves product robustness.