Research On Key Technologies Of Intelligent CNC System Based On STEP-NC

With the development of NC(numerical control,NC)technology,the open,intelligent and networked numerical control system has become main development trend.To adapt to this trend,an open,intelligent and networked machining data model is needed.However,most CNC systems has been adopting the ISO6983 standard,namely the G/M code.This standard has lots of shortcomings,such as incomplete information,one-way data flow,incompatibility and so on,which seriously hinders the development of open,intelligent and networked CNC systems,and has become the bottleneck of the development of modern manufacturing.The STEP-NC(STEP-compliant Numerical Control,STEP-NC)standard describes the object in a manufacturing feature-based manner,covers all the machining information,supports the bidirectional data flow and networked manufacturing,which provides a good condition for the intelligent CNC system.After the comprehensive analysis of the research about STEP-NC-based CNC system and its key technologies at home and abroad,focusing on the STEP-NC standard,this paper takes the intelligent CNC milling system STEP-NC-based as the research object,and studies its three key technologies.The main research contents are described as follows:(1)It introduces the background of the research,analyzes the research status of NC STEP-NC-based CNC system and its key technologies at home and abroad,analyzes the existing problems.Based on it,the purpose and contents of the research are presented.(2)Facing the issues of current feature recognition based on STEP-NC,an intelligent feature recognition method based on STEP and improved neural network is proposed.In the method,the geometric and topological information firstly is extracted from the STEP AP 203 file of a part.Followed by it,the minimum subgraph of the part is built by the judgment of the convexity of edges.Then,an improved BP neural network is proposed by combining artificial bee colony algorithm and BP neural network algorithm.Finally,by inputting the information data from the minimum subgraph of the part to the improved BP neural network,the efficient and accurate feature recognition for STEP-NC manufacturing features in the part is achieved.(3)According to the issues of process planning based on STEP-NC,an intelligent nonlinear process planning method based on STEP-NC is studied.The STEP-NC-based process planning is divided into macro process planning,detailed process planning and micro process planning.In the stage of macro-process planning,the improved BP neural network is used to generate nonlinear machining schemes of features,and the machining schemes for STEP-NC-based manufacturing features are intelligently obtained.In the stage of detailed process planning,according to the constraint principle of machining steps,bidirectional linked list algorithm is adopted to realize the efficient generation of reasonable machining steps sequence.In the stage of micro-process planning,the rules-based algorithms are proposed to realize the intelligent selection of machining resources(machine tools and tools),and then the artificial bee colony algorithm is used to realize the intelligent generation of the optimal machining route with the goal of the shortest machining time.Finally,with the goal of the shortest machining time and the lowest machining cost,an optimization model of multi-objective milling process parameters is established,and the intelligent optimization of process parameters is realized by using artificial bee colony algorithm..(4)Aiming at the existing problems of tool path planning based on STEP-NC,a tool path planning algorithm based on network optimization and Dijkstra algorithm for STEP-NC-based complex pocket is proposed.In the method,STEP-NC-based complex pocket enclosed isometric ring is generated by certain spacing.Then the network optimization model is established by using the offset curve of linear and circular arc properties(vertex,direction and length).Finally the shortest network circuit of the model is obtained based on Dijkstra algorithm,the tool path for STEP-NC-based complex pocket without interference is generated.(5)Choosing Visual C++6.0 as the development environment,and using the MFC application development framework,the STEP-NC-based intelligent CNC prototype system is developed under the Win7 system.