Multiple-dimensional Feature-based Integration Of Design, Machining And Inspection For Complex Structural Parts

Complex structural parts such as aerospace structural parts have complex structures with great accuracy, which require sophisticated machining processes. With the increase of large span structures, complex surfaces, thin-walled structures and deep pockets, the NC machining of complex structural parts are facing the new challenge. The dissertation focuses on multiple dimensional feature-based integration of design, machining and inspection for complex structural parts. The research achievements include:(1) The concept of multiple dimensional feature is proposed, and the multiple dimensional feature model is established. As the process dimension, the time dimension is added to the traditional three-dimension model to establish the multiple dimensional feature model. The multiple feature model not only represents the final status of the part, but also represents the complex intermediate status of the part. The constraint network is used to represent the machining process knowledge such as typical machining process which facilitates the creation of drive geometries and the removal volumes. The drive geometries and the removal volumes are used to generate the boundary represent model(B-Rep) of each status of the part, which supports the generation of inspection points. Benefitting from this method, the geometries of the intermediate status of the part are well associated with the machining process.(2) A process capability indexes calculation method is proposed for multiple dimensional feature. The sample size in multi-variety and small batch manufacturing is too small to calculate the process capability indexes accurately. The relationships between the process, dimension and tolerance are researched, and the ratios of the standard deviation to the tolerance of the machining features with the same machining process and tolerance but distinguished dimensions are constant. Then, these machining features are grouped so that the sample size is increased. The individuals of the sample are unified so that they have the same distribution, then process capability indexes can be calculated by point estimation method instead of interval estimation method.(3) A creation method of drive surfaces for multiple dimensional feature is proposed. For flank milling surface, the machining allowance is hard to obtain due to the deformation of the part. The mapping relationships between the machining allowance and the thickness tolerance of the intermediate status of the part and the thickness tolerance of the final status of the part are researched and the calculation methods of the machining allowance and the thickness tolerance of the intermediate status of the features are given. The machining allowance of the intermediate status of the features is used to optimize the grazing curves between the cutter and the surface which are employed to design the drive geometry. The drive geometry is used to generate tool path. The grazing curves are equally precision discretized to inspection points which facilitate the processing of measure data and reconstruction of the actual drive geometry. The thickness tolerances of the intermediate status of the parts are used to decide the deformation of the parts. Then, the drive geometries for the next machining operation are created for adaptive tool path generation. These researches will provide the theoretical and technological reference for the paradigm shift from the trade-off solution based on personal experiences to the optimized solution based on the inspection information in NC machining.(4) A multiple dimensional feature-based integration system of design, machining and inspection for complex structural parts is developed. And the system is verified in the manufacturing of several complex structural parts from some aerospace manufacturing companies in China.