智能化潜在失效模式与影响分析研究

发布时间：2018-05-03 08:54

本文选题：FMEA + 本体　；参考：《山东大学》2017年硕士论文

【摘要】：现代质量管理的一项研究表明,高质量的产品都是通过事前采取措施预防质量问题的发生,而不是通过事后检测来改进质量。因此,在产品设计之初就考虑产品的质量,对企业来讲至关重要。潜在失效模式与影响分析(PotentialFailure Mode and Effect Analysis,FMEA)就是一种典型的事前预防的可靠性分析方法,它从衡量产品或系统潜在故障的角度评估其可靠性程度,并将这些故障信息作为设计规划人员以及工艺人员预防失效与改进方案的依据。因此,准确、高效的FMEA分析对提高产品的质量具有重要的意义。本文智能化FMEA主要从FMEA知识建模、知识检索以及风险评估等几个方面来展开研究。首先,深入分析了 FMEA知识的特点,提出了一种基于本体的FMEA知识表示框架。将FMEA本体分为FMEA对象、功能、失效分析、FMEA团队四类核心概念,详细分析了这四类核心概念中主要的概念以及属性关系,建立了FMEA本体顶层概念的本体主模型,为后续FMEA知识的检索和重用奠定了基础。其次,针对目前常用的检索技术存在的缺陷,对基于本体的FMEA知识检索方法进行了研究,并提出了一种综合考虑语义距离、语义属性、语义重合度、语义深度、语义密度等因素的改进的语义相似度计算方法,最后针对内燃机活塞本体知识模型对该计算方法的有效性进行了验证。再次,通过分析目前FMEA风险评估方法存在的问题,提出基于模糊共因失效分析的FMEA风险评估方法的模型,该模型综合采用三角模糊数、灰色关联分析和决策试验与评价实验室法对失效模式的风险进行排序,既考虑了风险评估因子的模糊性,又考虑了失效模式和失效原因间的影响关系,使得排序结果更加符合工程实际。最后,基于上述研究,以中国航天山东山大华天软件有限公司的产品全生命周期管理系统——InforCenterPLM为基础,设计实现了系统中有关智能化FMEA的相关功能,并在某机械制造企业中的FMEA分析中得到成功的应用验证。
[Abstract]:A study of modern quality management shows that high quality products are improved by taking prior measures to prevent the occurrence of quality problems, rather than by testing afterwards. Therefore, it is very important for enterprises to consider the quality of products at the beginning of product design. Potential failure mode and impact analysis potential failure Mode and Effect Analysis (FMEA) is a typical pre-prevention reliability analysis method, which evaluates the degree of reliability of a product or system from the point of view of measuring potential failures. The fault information is used as the basis for the design planners and process personnel to prevent and improve the failure. Therefore, accurate and efficient FMEA analysis is of great significance to improve the quality of products. In this paper, intelligent FMEA is studied from several aspects, such as FMEA knowledge modeling, knowledge retrieval and risk assessment. Firstly, the characteristics of FMEA knowledge are analyzed in depth, and an ontology-based FMEA knowledge representation framework is proposed. The FMEA ontology is divided into four core concepts: FMEA object, function and failure analysis. The main concepts and attributes of the four core concepts are analyzed in detail, and the main ontology model of the top-level concept of FMEA ontology is established. It lays a foundation for the retrieval and reuse of FMEA knowledge. Secondly, aiming at the defects of the commonly used retrieval technology, this paper studies the Ontology-based FMEA knowledge retrieval method, and proposes a semantic knowledge retrieval method which considers the semantic distance, semantic attributes, semantic coincidence degree and semantic depth. An improved semantic similarity calculation method based on semantic density and other factors is presented. Finally, the validity of this method is verified by the piston ontology knowledge model of internal combustion engine. Thirdly, by analyzing the problems existing in the current FMEA risk assessment method, a model of FMEA risk assessment method based on fuzzy common cause failure analysis is proposed. The grey relational analysis and the laboratory method of decision testing and evaluation are used to rank the risk of failure mode, which not only considers the fuzziness of risk assessment factor, but also considers the influence relationship between failure mode and failure reason. The sequencing results are more in line with the engineering practice. Finally, based on the above research, based on the product life cycle management system of China Aerospace Shandong Shanda Huatian Software Co., Ltd., this paper designs and implements the related functions of intelligent FMEA in the system. It has been successfully applied in the FMEA analysis of a mechanical manufacturing enterprise.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH186

【相似文献】