Modeling And Characteristic Of Deviation Field Propagation For The Assembly Of Large Thin-walled Structures

The large thin-walled structure is the skeleton of the rocket tank product,which is subjected to the huge static and dynamic loads during the launching procedure.The performance of the product is directly affected by the dimensional accuracy of the thinwalled structure.The geometric deviations of the large thin-walled structure are the results of the the spring-back deformation during manufacturing and the coordinate deformtion during the assembling procedure.The deviations of the part at different position caused by deformation are various due to the large size and the weakness stiffness of the thin-walled structure.As a result,the deviations of parts are distributed as the field form in the space.Meanwhile,the external clamping fixture errors and the welding deformation may cause the additional sources of deviations.The assembly deviation of the final product is difficult to predict accurately.Because of the coupling effect of various sources of deviations,the reason of the assembly deviation is difficult to be diagnosed when the assembly quality of large thin-walled structrures is poor.It leads to the difficult control of the deviation for the key components and assembly effectively.Therefore,it is of great significance to propose an accurate deviation prediction method for large-scale thin-walled structures during the assembling process,which may reveal the principle of deviation accumulation during the assembling process.And the deviation of the assembly and the components may be controlled finally.In terms of the deviation propagation and control in the assembly process of large thin-walled structure,the models for accurate description of structural deviation and propagation are proposed.Compared with the method of influence coefficient,the accuracy of the assembly deviation prediction with the proposed model is improved.First,the deviation of the large thin-walled structure is described in the field form based on the linear combination of basic deviation fields.A deviation field propagation model is established to quantitatively study the relationship of deviation between the components and the assembly.The accumulation mechanism of deviation in the assembling procedure may be revealed by the deviation field propagation model.Then a two-step optimization process is eatablished to build the mapping model of the basic deviation field and the key measurement points on the part.By the control of the deformation on multi-points,the effective control of the assembly deviation may be achieved by using the diagnosed process.Finally,the assembly deviation propagation of the rocked tank is studied.The deviation flows of different assembly schemes and assembly sequence may be obtained,which are used to optimize the assembly process and improve the geometric accuracy and assembly quality of the rocked tank.The main contents and conclusions are listed as follow.(1)An accurate description model of deviation for large thin-walled structure based on stiffness decompositionConsidering the non-uniform deviation distribution of large thin-walled structures on germatric surface,an eigen-decomposition method of the structure stiffness is proposed to calculate the basic deviation fields of large thin-walled structures,in which the boundary constraint of large thin-walled structure during assembly process are considered.Each basic deviation field corresponds to a generalized compliance matrix of part.The actual deviation field of the part may be described as a linear combination of the deformation with various generalized compliance.The linear certain deformation mode of part with the generalized compliance can be analyzed and the compliance coefficient decreases with the increasing order of generalized compliance.For the specific part,the deformation corresponding to high-order generalized compliance may be seen as the roughness behavior of part,which may be eliminated in the engineering application.The actual deviation field of part can be decomposed into the finite basic deviaion fields.The shortcomings of local feature point deviation in the existing deviation description methods are corrected.(2)The model of the deviation propagation and analysis for large thin-walled structures during the assembly procedureConsidering the N-2-1 fixture principle in the compliant assembly,the effect of the fixture errors is described with the basic deviation fields.The principal component analysis method is used to get the typical welding deformation patterns under various welding parameters.Moreover,by defining the deviation source feature points,the welding deformation is introduced into the deviation propagation model in a form of equivalent force.Considering the coupling effect of the initial manufacturing deviation,the fixture errors and the welding deformation,the deviation propagation model for large thin-walled structures is established based on the deformation coordination relationship during the assembling process.The basic deviation field is the basic unit in the model.The whole deformation coordination effect in the assembly process is effectively revealed in a form of field-to-filed deviation propagation.The contribution and sensitivity of the basic deviation field for the components and assembled structure is deduced and may be used in the accurate prediction of the deviation and diagnosis.The accuracy of the proposed model is verfied by the assembly experiment.The results show that the proposed method may improve the deviation prediction accuracy of thinwalled structure,when compared with that calculated by the influence coefficient method.(3)The control of assembly deviation based on multi-points coordinated deformationThe effective independence optimization method is adopted to gradually eliminate the measurement points with the least influence on the deviation coefficients.A set of optimization points is calculated to realize the approximate representation of deviation coefficients.Then the optimization model of the relationship between the optimization points and the basic deviation fields is built.The genetic optimization algorithm is used to calculate the key measurement points and its deviation adjustments to each basic deviation field.The specific basic deviation filed can be controlled by adjusting the deviation of the corresponding key measurement points.Finally,the optimal control of the assembly deviation can be achieved.Based on the testing results in Chapter 3,the optimal control effects of the proposed model are verified by the assembly simulation.(4)Study on the deviation field propagation for large thin-walled structures of fuel tank in rocket structureBased on the deviation propagation model,the software of deviation description and propagation for large thin-walled strcutrures is developed.The assembly deviation prediction is carried out for the bottom structure of fuel tank.The reasonability of the proposed method is verified compared with the testing data.Considering the effect of the assembly sequence for various components of fuel tank,the quantitative analysis of the deviation propagation is conducted.The reasonable assembly scheme is selected for the assembly of the whole fuel tank,which may provide the foundation for improving the manufacturing quality of large thin-walled structures in the fuel tank.A set of closed-loop methods for the deviation description,deviation propagation,deviation control based on stiffness decomposition are formed in this thesis,which provide basic theory and technical method for deviation prediction and the improvement of assembly quality.