Research On Key Technology Of Intelligent CAPP System For Key Parts Of Marine Diesel Engine

Under the general trend of developing intelligent manufacturing,the existing computer-aided process planning system of a large marine diesel engine manufacturing company has some problems about low intelligence,low historical process data reuse rate,low process design efficiency and inability to meet the needs of intelligent manufacturing,etc.Combined with the actual situation of the enterprise,the business requirements were analyzed,the overall plan of the intelligent process design system was designed,and the key technology of the intelligent CAPP system was studied.Construct a structured historical process database,and on this base,make full use of historical process data to study the core technologies such as processing feature recognition technology,feature processing plan decision technology,process sorting rule mining technology,and process route intelligent generation technology.On the base of the original system,the above-mentioned technical function modules are added,which improves the intelligence of the system and improves the efficiency of process design.Taking the connecting rod of the key diesel engine as the verification object,the validity and reliability of the system are verified.The specific research content is as follows:(1)Build an intelligent CAPP system architecture.Combined with the actual situation of the enterprise,a demand analysis was carried out,and the overall system framework and various functional modules were designed.(2)Research on process feature recognition technology based on convolutional neural network.By creating a custom processing feature data set,importing convolutional neural network training modeling to obtain the optimal neural network model;based on MBD technology,quickly and accurately obtaining the processing features to be recognized of the part;converting the processing features into triangular mesh data,and gives an algorithm of triangular mesh data processing to realize the transformation of processing features into the data required by the neural network model to complete the feature recognition.(3)The intelligent generation technology of process route is studied.Constructed a structured process information database;obtained a large number of feature-processing plan data from the structured process information database,and trained the BP neural network to decide the feature processing plan;through the process routes of different types of parts in the structured process database The data is sorted based on the Prefix Span algorithm to obtain the process sorting rules of this type of part,which is convenient for the subsequent process sorting;the decision-making feature processing plan is sorted according to the process sorting rules,and related non-machining processes are automatically added to generate a complete Routing.(4)Developed and verified the intelligent CAPP prototype system.Combining the above key technologies,the processing feature recognition module and the process route automatic generation module are developed,and the new module is organically integrated with the original system module to form an intelligent system.Taking the key component connecting rod as the verification object,the three-dimensional process design was carried out to verify the effectiveness and reliability of the system.