Analysis And Application Of Data Mining Based On PDM System

With the rapid development of information technology,PDM(Product Data Management)system has been widely used in enterprises and has accumulated a large amount of data.The hidden knowledge in these huge data has not been fully utilized,that is,the status of "rich data and poor knowledge" presented by most enterprises.Data mining technology can discover hidden valuable information or knowledge from a large amount of data accumulated by enterprises.Therefore,it is of great significance to apply data mining technology to PDM systems.This paper is mainly based on data mining methods to research data in PDM system.The main contents include the following aspects:First,the data of the PDM system is analyzed from the perspective of data sources and data types.Analyze the data from the perspective of data sources to prepare for the purpose of clear data mining;analyze the data from the data type perspective to prepare for the data mining preprocessing and method selection.Secondly,based on the five major data types in the PDM system,a PDM-oriented data mining system model is constructed,including data layer,method layer,knowledge layer,and application layer.The data layer mainly discusses preprocessing methods for structured data,semi-structured data,and unstructured data to prepare for data mining.The method layer first proposes a data mining strategy based on the life cycle,and then analyzes and selects the correlation analysis and aggregation.Four kinds of data mining methods,such as class analysis,sequence pattern analysis,and predictive analysis,knowledge layer analysis of knowledge expression and knowledge classification management,and application layer analysis of possible application scenarios of various types of knowledge.Finally,based on the data mining system model,data analysis and data mining of product design data and product processing data are carried out.For the product design data,taking the motor as an example,Apriori algorithm was used to analyze the relationship between the design parameters in the design knowledge base and the motor model.For product processing data,the Weka(Waikato Environment for Knowledge Analysis)data mining platform was used to analyze the relationship between the three cutting elements,vibration and roughness,and several different algorithms were performed.Comparative analysis.