云制造模式下建材装备企业制造任务执行关键技术研究

发布时间：2018-01-18 04:20

本文关键词：云制造模式下建材装备企业制造任务执行关键技术研究　出处：《武汉理工大学》2013年博士论文　论文类型：学位论文

【摘要】：随着云计算、物联网等先进技术的迅猛发展和制造业所面临的挑战,云制造模式和技术应运而生,其核心在于实现制造资源的协同和共享,为制造业的升级转型提供了新的推动力。然而,由于云制造模式的开放性及多用户等特征,导致多用户任务执行过程涉及到制造任务的异构性、任务资源服务的竞争性、状态监控和任务执行后的评价等问题。对此,本文从四个方面来研究云制造模式下制造任务的优化执行,并以建材装备企业为应用对象进行实践研究。主要研究工作可以概括为以下几个方面： (1)针对制造任务执行过程中的任务异构性问题,提出了云制造任务的信息模型和语义建模框架,首先应用本体建模技术建立了初始云制造任务本体(OCMT_Ontology),以建材装备企业制造任务文本库为基础结合文本处理技术,通过本体自学习模型建立并完善通用云制造任务本体(GCMT_Ontology);在此基础基础之上,通过匹配GCMT_Ontology,构建云制造任务子本体(CMTS_ontology)来实现云制造任务的语义描述。 (2)提出了云制造任务执行服务链构建框架,在分析单用户资源服务优选的基础上,重点研究多用户任务资源服务优选模型,针对多用户对资源服务的竞争,提出了应用演化博弈论的方法构建多用户资源服务的博弈模型,按照任务执行是否延期分四种类型求解博弈模型的演化稳定策略,并分析其动态复制方程的演化相图。最后,依据制造子任务的时序逻辑关系来构建可执行的协同制造资源服务链。 (3)提出云制造模式下任务执行过程状态监控框架,构建了基于制造任务子本体的多源数据一致映射模型,按照制造任务的结构和执行服务链实现任务执行进度的数据融合,根据制造资源实时监控的数据,应用隐性马尔科夫模型实现制造资源状态的实时监控。 (4)提出了云制造任务执行过程评估模型,并建立了不同执行阶段的评估指标及模糊化处理方法。提出直觉模糊OWA-TOPSIS的方法对任务执行过程进行评估,构架了六类不同的TOPSIS数据累积方法和最佳和最差理想点的识别方法,并通过案例研究验证了提出方法的实用性。 (5)从实际应用出发,针对建材装备制造企业产品的制造过程,开发了云制造模式下任务执行过程管理系统,并在协同制造的多个企业间实现任务执行计划管理、过程跟踪和评价。
[Abstract]:With the rapid development of cloud computing, Internet of things and other advanced technologies and the challenges facing the manufacturing industry, cloud manufacturing models and technologies emerge as the times require, the core of which is to realize the coordination and sharing of manufacturing resources. It provides a new impetus for the upgrading and transformation of manufacturing industry. However, due to the openness and multi-user characteristics of cloud manufacturing mode, the multi-user task execution process involves the heterogeneity of manufacturing tasks. The competitiveness of task resource service, state monitoring and evaluation after task execution are discussed. In this paper, the optimal execution of manufacturing tasks in cloud manufacturing mode is studied from four aspects. And take the building materials equipment enterprise as the application object to carry on the practice research. The main research work may be summarized as follows: 1) aiming at the problem of task heterogeneity in manufacturing task execution, the information model and semantic modeling framework of cloud manufacturing task are proposed. Firstly, the initial cloud manufacturing task ontology (OCMT Ontology) is established by using ontology modeling technology, which is based on the document library of manufacturing tasks of building materials and equipment enterprises combined with text processing technology. The general-purpose cloud manufacturing task ontology GCMT ontology is established and perfected by ontology self-learning model. On this basis, the semantic description of cloud manufacturing tasks is realized by matching GCMT Ontology and constructing the cloud manufacturing task sub-ontology (CMTS _ S _ T _ ontology). In this paper, a framework for constructing cloud manufacturing task execution service chain is proposed. Based on the analysis of single user resource service optimization, the multi-user task resource service optimization model is studied. Aiming at the competition between multi-user and resource service, a game model of multi-user resource service is proposed by applying evolutionary game theory. According to whether the task is delayed or not, the evolutionary stability strategy of the game model is solved in four types, and the evolutionary phase diagram of the dynamic replication equation is analyzed. According to the temporal logic relation of manufacturing sub-task, the executable collaborative manufacturing resource service chain is constructed. Thirdly, a framework for monitoring the state of task execution process in cloud manufacturing mode is proposed, and a consistent mapping model of multi-source data based on manufacturing task sub-ontology is constructed. According to the structure of manufacturing task and the chain of execution service, the data fusion of task execution progress is realized. According to the data of real-time monitoring of manufacturing resources, the hidden Markov model is applied to realize the real-time monitoring of manufacturing resources. Finally, the evaluation model of cloud manufacturing task execution process is proposed. The evaluation index and fuzzy processing method of different execution stages are established, and the intuitionistic fuzzy OWA-TOPSIS method is proposed to evaluate the task execution process. Six different kinds of TOPSIS data accumulation methods and identification methods of best and worst ideal points are constructed, and the practicability of the proposed method is verified by a case study. Based on the practical application, the task execution process management system in cloud manufacturing mode is developed for the manufacturing process of building materials and equipment manufacturing enterprises. Task execution planning management, process tracking and evaluation are implemented among multiple enterprises in collaborative manufacturing.
【学位授予单位】：武汉理工大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TU506

【参考文献】