Study On The Key Technologies Of KBE Based Process Planning For Large Complicated Stampings

The large complicated stampings, such as automobile panels, have been widely applied to airplane, automobile and other vehicles, engineer machine, and metal structures. As the key point for the forming of large complicated stampings,process planning is the foundation for the die design and manufacturing, which directly affects product quality, cost, productivity and life of dies. Process planning for large complicated stampings is a complex decision-making and planning process, and is highly dependent on designer's experience so far, which leads to inefficiency and unreliability. It is necessary to adopt Knowledge-Based Engineering to realize automatized and intelligentized process planning for large complicated stampings to completely change the backward state. Supported by the Key Technologies R&D Programme in 10th national five-year plan and the national natural science foundation, the author has studied the key technologies in KBE based process planning for large complicated stampings, and developed a CAPP system, which is capable of carring out intelligent decision-making for process planning and 3D work-pieces design, and remarkably raising the automation and intelligentization level of the process planning. Process knowledge expression, process control, information association, design optimization and forming simulation are involved in the process planning for large complicated stampings. The author has proposed a master model for KBE based process planning. The master model is composed of knowledge base, process control structure, design evaluation, process information model, model link environment and simulation engine, which integrates the process design knowledge and methods, and provides an open, integrated design environment. Based on the product feature model, the author has set up a product/process knowledge model for large complicated stampings, which takes form feature, operation feature and process feature as the basic objects. The model can completely represent the product, and express the knowledge of process planning comprehensively. With object-oriented hybrid knowledge expression, product information, process knowledge and resource knowledge are expressed in the knowledge model, so that feature object and process knowledge are integrated to surport intelligent decision-making and knowledge expression and transfer. Operations sequencing is important for process planning. A hybrid reasoning model for process planning has been built to realize intelligent decision-making. Stamping formability analysis and operation combination and sequencing are realized by hybrid decision-making, which is based on Rule Based Reasoning, Model Based Reasoning, Case Based Reasoning and Genetic Algorithm. Considering the characters of process planning for large complicated stampings, a synthesis evaluation model has been proposed to evaluate process plans based on fuzzy theory. Drawing is the key operation in forming large complicated stampings, and product quality is directly affected by the design of drawn part. Studied on KBE based drawing workpiece design method, the author has proposed a control curves model for drawing workpiece design. It combines design guide rules, design experience, design constraints, and so on into the control curves. During the designing, the generated addendum section curves are constrainted by the control curves, and the addendum surfaces can be created based on section curves, hence knowledge-driven drawing workpiece design is realized, and design automation and intelligentization level is raized. Considering the vagueness for evaluation of drawing workpiece design, a two-level fuzzy evaluating model has been built to estimate drawing workpiece design, which provids an evaluating method to optimize drawing workpiece design, and improve its design quality. Numerical simulation has become a useful tool to find out forming defects. However, there is lack of explanation in the analysis results. How to improve design to remove the defects mainly depends on designer's experience and his judgement. The knowledge contained in the analysis results can not be effectively used. Therefore, an adaptive design model for drawing workpiece has been set up based on numerical simulation and knowledge discovery. With data clustering method and BP Neural Networks, design knowledge is dug from the data obtained by numerical simulation. The design parameters knowledge discoveried is used to drive the design model to realize drawing workpiece design optimized. The adaptive design method can support the procedure of "design—analysis and evaluation—optimizing design"for drawing workpieces. Based on the theoretical research, a computer aided process planning system for large complicated stampings —PDCAPP, has been developed on UG platform using UG/OPEN and Intent! of UG/KF. The main functions of the system include product/ process knowledge modeling, process planning, drawing workpiece design, design evaluation and modification. Knowledge based process planning for large complicated stampings is realized by the system. The design examples verify the correctness of the theoretical research results of KBE based process planning for large complicated stampings, and demonstrate the practicability of the system.