基于经验的数控工艺知识挖掘算法研究

Research on knowledge mining algorithm of NC process based on experience

针对目前基于标准工艺模板的数控工艺复用方法在模板自动提取的研究中未考虑零件几何信息的问题,直接基于实例零件的复用方法存在检索效率低,提出将数据挖掘技术应用于数控工艺知识。利用三维极半径曲面余弦、正弦矩结合三维包络盒信息实现零件几何信息的数值化表达;选取关键加工参数并与几何信息进行关联,获得结构化的数控工艺数据。选用k-medoids聚类算法结合主成分分析和自编码器2种降维方法对历史工艺数据进行聚类分析以获得典型工艺模板,并以注射模电极零件为例,验证了方法的有效性和实用性。

In view of the problem that the current NC process reuse method based on standard process template without considering the geometric information of parts in the study of automatic template extraction,and the retrieval efficiency was low in the instance part reuse method,the data mining technology was proposed to apply to the knowledge of NC process.The three-dimensional polar radius surface cosine and sinusoidal moment combined with three-dimensional envelope information were used to numerically express the geometric information of parts.The key machining parameters were selected and associated with geometric information to obtain structured NC process data.The k-medoids clustering algorithm combined with principal component analysis and autoencoder was used to cluster and analyze historical process data to obtain typical process templates.The effectiveness and practicability of the method were verified by taking the electrode parts of injection mould as an example.