基于实时数据的离散制造车间在制品流转状态研究

发布时间：2018-04-30 16:56

本文选题：频繁轨迹 + 预测制造　；参考：《南京航空航天大学》2017年硕士论文

【摘要】：制造物联技术的应用在制造环境中积累了大量宝贵的制造数据,为复杂制造系统的研究提供了坚实基础。作为制造系统的核心,生产任务往往围绕着在制品的加工过程进行分配,因此在制品制造数据为复杂制造系统的描述和解释提供了宝贵的逻辑知识,这些逻辑知识还能为生产状态的预测提供基础。本文针对离散制造车间中在制品的状态轨迹,提出了对在制品数据进行管理的数据模型,并在此基础上,提出了在制品状态轨迹模式的挖掘算法和在制品状态预测算法。本文的主要内容包括:(1)针对在制品数据中的知识,运用语义坐标和逻辑轨迹对在制品的状态轨迹链进行描述,对在制品的生产过程进行高层次抽象;(2)对在制品的状态轨迹链进行分析,在语义坐标的基础上,提出了在制品状态轨迹模式的挖掘算法;(3)在制品的状态轨迹模式为状态预测提供了逻辑知识。在轨迹模式的基础上,进一步提出了在制品状态的预测算法;(4)本文提出了物理空间和逻辑空间的融合框架,以及信息交换和推送机制。并且,本文对提出的方法进行了检验。
[Abstract]:The application of manufacturing technology accumulates a lot of valuable manufacturing data in manufacturing environment, which provides a solid foundation for the research of complex manufacturing system. As the core of manufacturing systems, production tasks are often distributed around the process of WIP, so WIP data provide valuable logical knowledge for the description and interpretation of complex manufacturing systems. These logic knowledge can also provide the basis for the prediction of production state. In this paper, a data model for managing WIP data in discrete manufacturing workshop is proposed. On the basis of this model, an algorithm for mining WIP state trajectory patterns and a WIP state prediction algorithm are proposed. The main contents of this paper include: (1) aiming at the knowledge in the WIP data, the state locus chain of the WIP is described by using the semantic coordinate and the logical locus. The state locus chain of WIP is analyzed on the basis of semantic coordinates. In this paper, an algorithm for mining state locus patterns of work-in-process is proposed. (3) the state locus pattern of work-in-process provides logical knowledge for state prediction. On the basis of trajectory pattern, a prediction algorithm of in-process state is proposed. (4) in this paper, the fusion framework of physical space and logical space, and the mechanism of information exchange and push are proposed. In addition, the proposed method is tested in this paper.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB49

【参考文献】