Modeling And Diagnosis Of Dimensional Variation In Rigid Multistation Assembly Of Sheet Metal

Sheet metal is widely used in automobile and aircraft industry, the dimensional deviation of sheet metal is one of the important indicators for the quality of products. Especially in the automotive industry, the vehicle’s sealing, appearance and parts assembly quality is closely linked with the assembly dimension deviation of sheet metal. The traditional way of controling deviation is measured to realize the analysis of the dimension deviation and deviation source diagnosis by using multivariate statistical, only relying on experience, needing to spend a lot of time and for the affairs of behavior. So we need to develop a model that can be able to accurately predict dimensional deviation in multistation assembly of sheet metal, and can rapidly diagnose deviation source by the model.This paper takes sheet metal as the research object, on the basis of analyzing the main assembling deviation source of sheet metal, and establishes the state space model of variations stream in multistation assembly of sheet metal, and studies the deviation source diagnosis by using modern control theory. The method of deviation source diagnosis for multistation assembly of sheet metal is proposed based on surface scanning. Based on the predecessors’ work, the main research work is divided into th e following:1) Consider deviation source: fixture deviation and parts deviationParts positioning hole and fixture positioning pin, positioning groove, etc have different size tolerance, so fit clearance exists between them. Fixture pin and part hole together decide the assembling deviation of sheet metal, therefore it is necessary to consider fit clearance of the pins and locating holes, locating slot for the model.2)Deviation modeling for multistation assembly of sheet metal based on state space methodOn the basis of considering deviation source for multistation assembly, we analyse the fixture deviation, parts positioning hole deviation and reorientation deviation, establishing the deviation model for multistation assembly of sheet metal by state space method. We simulate the assembly process deviation by MATLAB to get the propagation law of assembling deviation. The calculated theoretical data and the scan data are compared to prove the validity of the model by mutual authentication.3) The method of assembling deviation source diagnosis based on surface scanningWe study the deviation source diagnosis model of multistation assembly used observability of control theory, and anslyse the defect of above model for deviation source diagnosis, properly modificating the expression of assembling deviation, thus getting complete diagnosable conditions of deviation source. Fixture deviation can be rapaidly diagnosis by using surface scanning.4) The specific assembly verifies deviation model and diagnosis modelOn the basis of deviation model and the diagnosis model, we analyse and calculate the floor assembly samples, and the assembling deviation is measured by surface scanning. The fixture deviation is obtained by diagnosis model compared with three coordinate measuring valu es, verifying the correctness and validity of the model.