【作者】 李德刚；

【导师】 于德介；

【作者基本信息】 湖南大学 ， 机械工程， 2006， 博士

【摘要】 设备维护既是保证制造企业经营成功的关键因素，也对社会的可持续发展有重要意义；提升企业的设备维护水平不仅有助于企业实现其经营目标，也将推动生产、环境与社会的和谐发展。 鉴于设备维护的重要性，论文在湖南省科技攻关项目和中石化总公司科技攻关项目的资助下，针对先进制造环境和新维护价值观(强调安全、环保和人的价值)条件下设备维护在维护策略、维护管理机制、维护支持工具和维护资源配置等方面所面临的挑战，提出以预知维护的模式拓展作为应对挑战的解决方案，深入系统的研究了预知维护模式的体系理论、关键支撑技术与企业实施策略，取得了如下创新性研究成果： (1) 构建了预知维护模式的体系理论，借助科学技术的范式概念给出了设备预知维护模式的基本定义，归纳了预知维护模式的本质内涵，并在其基础上辨识了预知维护模式的构成要素，分析了构成要素的结构框架及相互关系，研究了预知维护模式的进化机制并探讨了预知维护模式的实施策略。体系理论的构建将为预知维护模式的拓展与实施提供理论支撑和方法论指导。 (2) 针对设备状态的“预知”问题研究了设备状态的预测技术，将更新迭代的思想引入灰色预测方法中，构建了基于在线监测的代谢递补灰色预测方法，并将该方法集成进大机组的群监测系统；针对一般支持向量机多步预测方法的不足，考虑在预测过程中不断利用预测结果修正支持向量机模型，构建了支持向量机迭代多步预测方法。上述预测方法的研究为设备状态的预知提供了技术支撑。 (3) 将仿真技术引入设备维护工作的优化中，构建了基于广义随机Petri网的设备维护流程仿真优化方法。该方法包括维护流程建模、流程模型的系统仿真平台(Artifex)映射和仿真运行三个步骤，能够量化的评估设备维护方案的可行性和优化度。设备维护流程仿真优化方法的研究为更好地发挥设备状态“预知”的效用提供了技术使能。 (4) 针对多机组低成本关键设备的在线监测问题研究了群监测技术，在背景企业选定的装置上设计开发了群监测系统，为预知维护的企业实施提供了必要的在线监测工具支持和关键设备(大机组)状态信息。 (5) 针对全面掌握设备状态的预知维护实施需求，设计了设备点检与在线监测相结合的设备状态监测完整解决方案，并将工作流技术引入设备点检的执行及管理中，研制开发了手持式点检仪，设计编制了设备点检工作流管理软件，实现了设备点检工作的数字化、规范化和绩效管理化，为预知维护的企业实施提供了必要的离线监测工具和成套装置状态信息。更多还原

【Abstract】 Plant maintenance is not only a critical factor in the success of manufacturing enterprises, but also of great significance in the sustainable development of the society. Upgrading the level of plant maintenance will not only help enterprises achieve their business objectives, but will promote the harmonious development of production, environment and the society.Because of the importance of plant maintenance, under the sponsor of Hunan provincial key scientific and technological project and the China Sinopec key scientific and technological project, and aiming at the challenges of maintenance strategies, maintenance management mechanisms, maintenance supporting tools and the allocation of maintenance resources under the condition of the advanced manufacturing environment and the new criterion of value judgment of plant maintenance (emphasizing safety, environment protection and human values), the present dissertation proposes an extended predictive maintenance pattern as a solution, and makes an in-depth investigation of the system theory , key supporting technology and business implementation strategy of predictive maintenance pattern. The main contributions of this dissertation are summarized as follows:(1) The system theory of predictive maintenance pattern has been established. The basic concept of predictive maintenance pattern based on the paradigm concept of science and technology is given and the essential connotation of predictive maintenance pattern is summarized. The compositional elements of predictive maintenance pattern have been identified. The structure and framework of the compositional elements of predictive maintenance pattern and their mutual relations are analyzed. The evolution mechanisms of predictive maintenance pattern are investigated and the implementation strategies of predictive maintenance pattern are explored. The establishment of system theory will provide a theoretical support and guidance to the extension and implementation of predictive maintenance pattern.(2)Aiming at the "predictive" issue of the maintenance condition, the author investigates the predictive technology of units condition, introduces the idea of updating into the grey predictive method, develops a metabolic filling grey predictive method based on the on-line monitoring of units , and integrates it to a group monitoring system of big units. Aiming at the deficiencies of general multi-step predictive method based on support vector machine, the author considers to constantly更多还原