Quality Control Of Steam Turbine Diaphragm Assembly And Construction Of Knowledge Graph

With the continuous development of manufacturing technology and production organization forms,automation technology,information technology and manufacturing technology are infiltrating each other,and the complementarity and integration of different fields are becoming more and more obvious.As a powerful driving force of economic development,the manufacturing industry is making great strides to combine the traditional manufacturing process with modern information technology to realize "manufacturing" + "intelligence".In this context,steam turbine,as one of the representatives of national defense equipment,its production mode is also changing towards informationization.As the main component of steam turbine,diaphragm is the main component of steam turbine,its assembly process is complicated,there are too many parts,and there are many factors that affect the final assembly quality.Variety.The traditional assembly quality control process has a low pass rate of one-time assembly,and requires multiple repairs to meet the qualification requirements.Therefore,it is necessary to improve the status quo from a data perspective,give full play to the value of the quality data of the partition assembly process,and find out the useful information hidden in the data for the assembly process through data mining and other means to realize a data-driven partition assembly.This paper takes the assembly process of a steam turbine diaphragm as the research object,and uses the relevant algorithms in data mining,and does not realize the data to guide the assembly process of the diaphragm as the following research.First,study and analyze the research status of quality control of the assembly process at home and abroad,analyze the relevant algorithms of data mining,and construct the overall framework of the quality control of the partition assembly process.Through comparative analysis,the feature selection algorithm combining filtering and wrapping is used as the data feature model construction algorithm;the association rule extraction algorithm is used as the key quality control point selection algorithm;and the neo4 j database is used as the database for the assembly process knowledge map construction.Secondly,because of the complexity of the partition assembly process,the quality characteristic data of the assembly process will be too high,which is not conducive to the data mining process,so the feature selection method is used to screen the partition data for key quality characteristics;through comparative analysis Choose to use a combination of filtering and wrapping two feature selection methods to carry out this process.Then,the Apriori association rule algorithm is used to extract key quality characteristics that are strongly related to the quality of the final product,and based on the neo4 j database,a knowledge map of the partition assembly process is constructed to guide the assembly process.Finally,based on the Django Web framework,Python development language,Pycharm and other development tools,the steam turbine diaphragm assembly quality control system was designed and developed,and relevant examples were verified.Through the above process,a quantitative basis is provided for the selection and control of the key quality characteristics of the partition assembly process,and it has certain guiding significance for the entire partition assembly process.It provides a corresponding theoretical basis for the digital transformation and intelligent manufacturing of partition assembly,and has certain application value for realizing the goal of "digital manufacturing".