Research On Methodologies Of Part Structure Design And Geometric Analysis Of Its Machining Process With Transformation Of Feature Information

Along with increasing of market competition, industries generally adopt the strategy of product family based part design and variant CAPP in the product developing process. Now, there still exist several severe problems in the implementation of this strategy. For example, it is difficult to reuse the old design and determine the machining ability of old machining resources because the current design tools are not intelligent enough and transformation of design information from design phase to the downstairs is not smooth enough, too. Based on analyses of the pervious research work regarding these issues, in this dissertation, an idea for part structure design and geometric analysis of machining ability based on transformation of feature information is put forward. First, a novel method is presented for automatically generating engineering-richened part structure models according to function requirements and rules of sub-function combinations. In this method, an atomic function dependency graph (AFDG) is generated first according to the design requirements and the complete neighbor definitions of atomic functions in a pre-defined design rule library. Then, a feature dependency graph (FDG) of a part is synthesized from the AFDG using pre-defined mappings between atomic functions and local feature structures. Finally, the FDG is transformed to a CSG-type feature structure model, which can be input to currently available commercial CAD systems to generate final CAD part models. Second, an approach is given for checking equivalency between two different geometric models based on solving CSG equations. In this approach, a general form for parametric CSG model was proposed based on the concept of parametric half spaces. The inequalities constraining the model parameters are deduced from the relationships among the parametric half spaces of different CSG models, which are required to check their equivalency. The inequalities defining the equivalent parameter domain are solved and an explicit representation of their solutions is obtained by utilizing the Γ-algorithm from the linear algebra field. Third, an approach is developed for optimally selecting locating positions of workpieces and identifying feasible clamping regions that meet requirements of the form-closure principle for fixture layout that can be used to fix multi-workpieces. In this approach, firstly finds an optimal configuration of locating points on a set of faces of the workpiece, which minimize the error transfer from locating points to machining features. In formulation of the optimization, the error transfer is modeled using an error transfer matrix. Eigenvalues of the matrix are used as objective functions of the optimization. Add coefficient of accuracy to the max eigenvalue of every workpiece for condition of one fixture fix multi workpiece. Secondly, this approach computes the clamping wrench cone and its member wrench can form a form-closure fixture layout together with the formerly selected locating points. Finally, it generates the feasible clamping regions by projecting the clamping wrench cone onto faces of clamping feature faces. In addition, the clamping wrench cone projection method for generation of feasible clamping regions is more suitable to handling the cases with concave clamping faces than traditional convex-combination based methods. Finally, a feature-based geometric analysing method of manufacturability for workpiece is presented. In this method, the machining ability of manufacturing process to the given workpiece can be determined by comparing the precedence tree of machining features for a workpiece with a sequence of operations in a process. If all the operations in the process can machine all the machining features of the workpiece, we can get the conclusion that the process can be adopted to machine the workpiece. The capability checking mainly includes the steps of machining feature matching, fixture constraint satisfaction checking, and form-closure condition checking. The general architecture of a prototype system for part structure design and geometric analysis of machining ability based on transformation of feature information has been given to validate the ideas and methods proposed in this dissertation. Using Visual C++6.0,Matlab 6.5, and UG II NX2.0, four sub-modules, including Part structure design, Feature conversion, Fixture layout design of multi-workpieces, and geometric analysis of machining ability, are developed and implemented. Some practical examples are used to validate the correctness and effectiveness of the research results. In the end of the dissertation, some conclusions have been drawn and the future directions in the field are given.