A novel intelligent system for machinery condition monitoring and failure prognostics

一种用于机械状态监测和故障预测的新型智能系统

• 批准号: 402144-2011
• 负责人: Liu, Jie
• 金额: $ 1.68万
• 依托单位: Carleton University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=573174
• 关键词: novel; intelligent; system; machinery; condition

The development of reliable condition monitoring and failure prognostic systems has been the focus of various undertakings in a wide array of industries to prevent machine/system performance degradation, malfunction, or even catastrophic failure. Although a variety of methods have been proposed in the literature, reliable assessment of machine current and future health conditions still remains a challenging task in both research and real-world industrial applications. This is because most machines are complex in structure and are operating in a noisy and time-varying environment, wherein the traditional signal processing techniques, pattern classification schemes, and failure prognostic approaches cannot be effectively implemented. Correspondingly, the objective of my proposed research program is to develop a novel intelligent system to address these challenges. The ultimate goal is to develop more reliable techniques for industrial monitoring and prognostic applications so as to improve product quality and to reduce the manufacturing cost in Canadian companies. The proposed intelligent system mainly consists of three innovative functional modules: feature extraction using a new wavelet spectrum analysis technique, fault diagnosis using a structure-evolving smart diagnostic scheme, and failure prognostics using a knowledge-model-fusion framework. The applied strategy is to develop advanced techniques at each processing module to improve the diagnostic and prognostic reliability for engineering systems under time-varying operational conditions. Particular focus in the proposed research program will be placed on the fault diagnosis and failure prognosis of machine drivetrains, helicopter engines, as well as the end-of-discharge and end-of-life prediction of lithium-ion batteries. The proposed intelligent system will become very useful for safety-critical applications (e.g., aircrafts, power plants or chemical plants), for cost-critical applications (e.g., spacecrafts, offshore wind turbines), and for volume-critical applications (e.g., automotive assembly lines, large manufacturing facilities).

可靠的状态监测和故障预测系统的开发一直是各种行业中各种事业的焦点，以防止机器/系统性能下降，故障，甚至灾难性故障。虽然在文献中已经提出了各种方法，机器当前和未来的健康状况的可靠评估仍然是一个具有挑战性的任务，在研究和现实世界的工业应用。这是因为大多数机器结构复杂，并且在噪声和时变环境中操作，其中传统的信号处理技术、模式分类方案和故障预测方法不能有效地实施。相应地，我提出的研究计划的目标是开发一种新的智能系统来应对这些挑战。最终目标是为工业监测和预测应用开发更可靠的技术，以提高产品质量，降低加拿大公司的制造成本。 提出的智能系统主要由三个创新的功能模块：特征提取使用一种新的小波谱分析技术，故障诊断使用的结构进化的智能诊断方案，和故障诊断使用的知识模型融合框架。应用的策略是在每个处理模块开发先进的技术，以提高工程系统在时变操作条件下的诊断和预测的可靠性。拟议研究计划的重点将放在机器传动系统，直升机发动机的故障诊断和故障预测以及锂离子电池的放电和寿命终止预测上。所提出的智能系统对于安全关键应用（例如，飞机、发电厂或化工厂），用于成本关键的应用（例如，航天器，海上风力涡轮机），以及对于体积关键应用（例如，汽车装配线、大型制造设施）。