Research On Locating Layout Planning For Aeronautical Sheet Metal Part Based On Prior Constraint

Sheet metal part is widely used in various fields of aerospace,automobile,shipping,and so on.Among them,aeronautical sheet metal part?ASMP?especially always tends to deform at machining,assembly,and measuring stages during the whole manufacturing process because of its properties of thin wall,large size,and low rigidity.As a result,it is difficult to locate ASMP accurately and guarantee its aerodynamic outline veracity.Generally,the tooling or fixture is used to constrain the excessive deformation of ASMP to ensure its locating accuracy and outline veracity during the manufacturing process.However,the current design of fixture basically relies on the experience of the designer,lacking of scientific theoretical guidance.Reasonable and efficient fixture locating layout planning is an important way to control the deformation of ASMP and to improve its manufacturing precision and efficiency.For ASMP fixture design,the core task is how to determine the optimal fixture locating layout so that the ASMP deformation in the manufacturing process is minimized,while meeting the requirements of the short computing time and high design quality.Aiming at the problems above,this dissertation accomplishes the ASMP fixture locating layout planning by taking the design features of ASMP as prior constraints,while with the goal of reducing the deformation of ASMP.To this end,a design method for ASMP fixture locating layout planning based on prior constraint is presented.Meanwhile,a prototype system for ASMP fixture layout planning is developed and an experimental platform for multi-point flexible locating is established to verify the effectiveness and correctness of the proposed method.Therefore,the ASMP fixture locating layout planning technique combining surrogate model and evolutionary algorithm under prior constraint of key locating features is researched based on“?N₁+N₂?-2-1”locating criterion.The method can reduce the cost of finite element calculation while optimizing the fixture locating layout.Moreover,it meets the requirement of controlling deformation during the ASMP manufacturing process,and can improve the fixture design efficiency at the same time.The main research contents and achievements are as follows:?1?For the optimization design of ASMP fixture locating layout considering the design features prior,an overall planning method for ASMP fixture locating layout given the prior constraint is proposed and a new“?N₁+N₂?-2-1”locating criterion is constructed.Based on the“N-2-1”locating principle,the N locating points on the primary datum of ASMP are divided into two parts for partition optimization:the key feature constraint locating point N₁ and the deformation control locating point N₂.The locating point N₁ is determined using a features recognition and prior selection method.Then under prior constraint of N₁,the locating layout of N₂ is optimized.Thus,An ASMP locating layout overall planning method combining the recognition and prior selection of locating features,and surrogate model and evolutionary algorithm is constructed based on the“?N₁+N₂?-2-1”criterion.?2?By analyzing and modeling the design features of ASMP,a locating features recognition and prior selection?classification and ranking?method based on attribute adjacency graph,self-defined rules and analytic hierarchy process?AHP?is presented.Through analyzing the attributes of the ASMP locating features,a 9-tuple locating feature model is constructed.Then the topological relations in the MBD model are transformed into the edge and face attribute adjacency graph,and the locating features in the attribute adjacency graph are extracted and identified.According to the geometric and physical information of the locating features,the self-defined feature classification rules are built and the locating features are classified.Based on AHP,the degree of membership of each candidate locating feature to the target key locating features is calculated.Then the final ranking of the locating features is achieved by analyzing the combined weight values of the various candidate schemes and the key locating features are determined.?3?Taken the key locating features as prior constraint,a support vector regression?SVR?parameters optimization approach based on flower pollination algorithm?FPA?,termed as FPA-SVR,is presented to enhance the prediction ability of SVR model and the FPA-SVR based ASMP fixture locating layout optimization surrogate models are constructed.In this dissertation,FPA is used to find the best hyper-parameters of SVR,and the proposed FPA-SVR is a relatively effective model in terms of accuracy and efficiency in comparison with GA-SVR and PSO-SVR models through the testing analysis on the public standard datasets.Then the parametric finite element model for ASMP fixture locating layout is established,and a limited number of training and testing samples are generated by combining Latin hypercube design?LHD?with FEA.Learning from the limited training and testing samples,the FPA-SVR surrogate models can be founded for ASMP fixture locating layout optimization.?4?A new multi-objective optimization method is proposed in this dissertation combining the FPA-SVR surrogate models and the elitist nondominated sorting genetic algorithm?NSGA-II?to realize the multi-objective optimization of ASMP fixture locating layout and to reduce the time of FEA calculation during the iterative optimization.The fixture locating layout taken as design variables and the overall deformation and maximum deformation of ASMP under external forces taken as the multiple objective functions,the multi-objective optimization mathematical model for fixture locating layout under the prior constraint is established.Combined with the FPA-SVR surrogate models,NSGA-II is employed to search for the Pareto front and Pareto optimal solutions for the ASMP fixture locating layout,which provides theoretical guidance for the ASMP fixture design.At last,an experimental validation scheme for the planning of ASMP fixture locating layout is designed.Here,a prototype system for ASMP fixture locating layout planning considering prior constraint is developed,and a multi-point flexible locating experimental platform for curved surface sheet metal part is built.Then,an aircraft fuselage skin scale part is chosen as an example.Through the experimental data analysis,the results of the parametric finite element model,the FPA-SVR surrogate models,and the multi-objective optimization are verified.