Research On CAD/CAM/CAPP Integrated System For Reciprocated Traveling WEDM

Wire electrical discharge machining(WEDM)is an important branch of electrical discharge machining(EDM)technology.It is a type of non-traditional machining based on the eroding effect of spark discharge.Compared with traditional cutting,WEDM process has the characteristics of no cutting force and high processing efficiency.It has been widely applied in mould manufacturing,automobile and military industry.In the process of WEDM,precise planning of machining trajectory and reasonable process design are quite important.In order to solve the trajectory planning and process optimization problems in reciprocated traveling WEDM.This thesis carries out a series of research works on intelligent CAD/CAM/CAPP integrated system.A networked CAD/CAM system has been developed based on cross-platform software development technology.The function modules were developed by Qt C++ and organized by event-driven model.To achieve the compatibility of various programming and control modes,Socket was used to establish the C/S communication model between CAM and machine controller,and the network transmission of tasks can be carried out via TCP/IP protocol.The CAD/CAM includes the basic drawing and trajectory planning functions for reciprocated traveling WEDM,and supports advanced programming functions,such as Multiple-cutting,ruled surface and taper trajectory planning.The system can run on the Windows and the Linux operating systemTo solve the entity sorting problem of graphics which contains huge amount of geometric entities,a new algorithm was proposed in this thesis,which was called memory search algorithm.Compared with the traditional algorithm,memory search algorithm achieves the local optimal search and realizes the dimensionality reduction of algorithm complexity.It provides support to the efficient and accurate trajectory planning of precision and complex geometries.The process of reciprocated traveling WEDM has the characteristics of complexity and diversity.Due to its excellent performance in nonlinear regression analysis,support vector machine regression algorithm(SVR)were used to solve the problems of multiple-cutting processing prediction and process optimization.The verification showed that compared with the traditional regression model and the RBF neural network model,the SVR model has better prediction accuracy and generalization ability,and it can be used for reliable prediction of process indexes.On this basis,reciprocated traveling WEDM CAPP system was constructed based on grid search method.Process experiments were conducted using the recommended parameters of the CAPP system.The results showed that the process indexes were optimized in a certain degree while satisfying the acceptable conditions.