Robust Optimal Design Of Multi-objective Interaction

The research on robust optimization design is always the focus and difficulty in the field of optimization design,especially the robust optimization design considering the interaction of multiple response objectives.In the practical work of a flexible positioning platform with flexure hinge as its main functional component,there are many problems,such as unreasonable design of its main structural parameters,low positioning accuracy,displacement sensitivity not reaching the expected value,etc.In this thesis,a new method of robust optimal design of compliant mechanisms considering the interaction of multiple response objectives is proposed.The purpose of this method is to find the optimal combination of structural parameters considering the interaction of multiple objectives,so as to achieve the optimal design.Taking a flexure hinge micro positioning platform as the research object,the following works are carried out:Firstly,the dimension and order of magnitude of the original data are not fully considered in the multi-objective robustness optimization,and the matching degree between the data and the corresponding analysis method is not considered,which leads to the problem that the micro displacement of the designed positioning platform can not reach the expected value.In this paper,KMO test and Bartlett sphericity test are introduced to test whether the corresponding method can be applied to the data.The structural parameters of the most vulnerable component of the platform under the maximum stress: width,depth of incision,length and thickness are taken as the structural variables;at the same time,the four performance indexes of the component under the maximum stress: micro displacement,flexibility for X direction,flexibility for Y direction and flexibility for rotation center direction are used to establish the objective function through the proxy mathematical model,and the ANSYS finite element software is used In order to provide scientific and reliable data for the follow-up robust optimization design,Minitab is used to select the test scheme and standardize the data.Secondly,in the process of multi-objective robust optimization design of the platform,the interaction between response variables is difficult to determine due to the limitation of main structural parameters.Although the traditional response surface method can be used for robust optimization analysis of multi-objective problems,the essence of this method is to findthe optimal design of multiple structural parameters combination for a single response objective,ignoring the interaction between multiple response objectives,so that the results are not completely consistent with the actual situation.The factor analysis method introduced in this paper analyzes the main parameter variables and multiple response objectives in the multi-objective robust optimization design problem as a unified variable at the same time,calculates the total variance of the interpretation between multiple variables,and intuitively shows several common factors with large interpretation weight from the total variance.In order to further study the interaction of these common factors on the platform quality robustness interference Then,it forms a common factor matrix,and then rotates the matrix.The specific performance of the interaction can be obtained by factor scoring the rotated matrix into the factor comprehensive scoring formula,and then reasonable structural parameters can be selected to improve the positioning accuracy of the platform.Finally,in the current multi-objective response robustness optimization design,the influence of decision-maker's subjective will can not be completely avoided when setting the corresponding weight,which leads to the problem that the optimization results are not completely consistent with the actual situation.In this thesis,taking a micro positioning platform as an example,the weight of the main structural parameters of the micro positioning platform can be accurately obtained by calculating the factor load and factor comprehensive score through SPSS software.According to the requirements of its technical parameters,it provides a strong theoretical support for the setting of the weight of the main structural parameters of the platform under the consideration of multiple response interaction.The above shows that the research of this thesis will provide useful guidance for the theory and Application Research of robust optimization design of compliant positioning platform,which has better theoretical research significance and practical application value.