Product Assembly Error Analysis And Selective Assembly Technology Considering Part Deformation

In order to ensure the quality of assembly,the influence of the deformation of the parts on the assembly accuracy must be considered in the process of assembling high-precision complex products.Part deformation is used as the source of deviation to participate in the analysis and selection of assembly errors.How to deal with part deformation and geometric errors has become a key issue.In this paper,an assembly error analysis method that considers part deformation is implemented,including a superposition analysis method of part deformation and part processing error,an equivalent and transfer method of deformation amount to geometric error.Focusing on the parts selection problem,the data modeling and selective assembly strategies are improved to ensure assembly accuracy and increase the success rate of part assembly.The first chapter discusses the research background and objectives of the subject,and explains the problems caused by the accuracy analysis of assembly deformation.This article reviews the current research status at home and abroad,including point cloud segmentation and point cloud registration technology,assembly error analysis methods,and multi-target selective assembly technology.Finally,the research content and organizational framework of this paper are introduced.The second chapter proposes to use the registration of point cloud model and mesh model to realize the superposition analysis of part deformation and machining error.The point cloud model is used to express the error information of the actual parts,the finite element is used to calculate the deformation of the ideal model,and the pre-registration and accurate registration of the point cloud model and the ideal mesh model are achieved based on the principal component analysis and the ICP algorithm respectively.According to the deformation offset of each node of the grid model,the deformation offset of each discrete point position of the point cloud is calculated by trilinear interpolation and superimposed with the coordinates containing the error information to achieve the coupling of the two.The third chapter builds a new point cloud model based on the superposition of part deformation and processing errors,segments part point cloud models to obtain feature point clouds.For planar features and freeform surface features,according to the distance from each feature point cloud to the ideal grid,Assess the dimensional and shape errors of part features.The rotation matrix and translation matrix obtained from the registration of the point cloud and the grid are used to represent the characteristic error with a small displacement spin.The Jacobian matrix is modified according to the new dimensional error,so that the deformation of the part equivalent to geometric error can be transmitted in the assembly chain.The calculation method of the influence of the deformation amount on the assembly accuracy is given.In the fourth chapter,for the selection of parts,parameters such as the amount of deformation are added to the part parameter database,and a unified selection information model is established according to the assembly requirements and assembly relationships.Due to the low yield of some parts,on the premise of meeting the accuracy requirements,try to increase the assembly success rate of parts as much as possible and reduce the number of remaining parts.Based on the genetic algorithm,a comprehensive selection of strong and weak Pareto relationships and population density is designed.For the problem of genetic exchange of the same type parts in different positions,the directional exchange strategy was added to selective assembly strategy,the mutual exchange channels of this type of parts were opened to increase the number of part combinations,and the assembly success rate of parts was further improved.The fifth chapter applies the methods proposed in this paper to solve the problems caused by the deformation of the thermal protection parts and the low yield of the parts,verify the feasibility of the method.The sixth chapter summarizes the research results of this article and looks forward to the future research work.