基于改进Apriori算法的船用柴油机装配工艺文件关联规则挖掘研究

柴油机作为船舶关键部件之一,其装配工艺设计的合理与否与船舶的整体性能直接相关。针对企业历史装配工艺文件不能充分为新机型装配任务发布时装配工艺设计提供支撑的问题,提出一个基于关联规则挖掘的柴油机装配工艺规则挖掘系统,该系统充分利用已有装配工艺文件,结合数据挖掘技术,对文件进行挖掘分析,建立装配规则库,为后续装配工艺智能设计提供数据支持。本文提出一种改进的Apriori算法,建立了关联规则挖掘的数学模型和基于自定义词典装配工艺文件分词算法,将分词处理过后的文件可以作为数据挖掘算法的输入,并在传统Apriori算法的基础上提出一种改进的Apriori算法,该算法核心思想是在对候选频繁事务集剪枝时,降低算法的时间复杂度,从而提高挖掘效率直到获得柴油机装配工艺文件中的关联规则。最后以某船厂的历史机型柴油机气缸盖装配工艺文件为例,验证了改进Apriori算法在挖掘柴油装配工艺文件关联规则时的可行性和有效性。

Research on Association Rules Mining of Marine Diesel Engine Assembly Process Documents Based on Improved Apriori Algorithm

As one of the key components of the ship, the diesel engine is reasonable in its assembly process design and directly affects the overall performance of the ship. Aiming at the problem that the enterprise history assembly process document can not fully support the assembly process design when the new model assembly task is released, a diesel engine assembly process rule mining system based on association rule mining is proposed.. The system makes full use of the existing assembly process documents and combines the data. Mining technology, build a database of assembly rules, and provide support for subsequent assembly process intelligent design. In this paper, an improved Apriori algorithm is proposed, and the mathematical model of association rule mining and the word segmentation algorithm based on custom dictionary assembly process are established. The file processed by word segmentation can be used as input of data mining algorithm and based on traditional Apriori algorithm. An improved Apriori algorithm is proposed, which can reduce the time complexity of the algorithm and improve the operating efficiency until the candidate rules of the diesel engine assembly process file are obtained. Finally, the feasibility and effectiveness of the improved Apriori algorithm in mining the association rules of diesel assembly process documents are verified by taking the assembly paper of all models of diesel engine cylinder heads in a shipyard as an example.