Advanced Prognostics and Maintenance Optimization Methodologies for System Health Management

对于油气系统、管道系统、可再生能源系统等工程系统而言，确保其可靠性、降低整体成本，以提升系统竞争力、确保安全、增加利润、保护环境至关重要。预测和健康管理（PHM）旨在根据系统健康状况提高系统可靠性并降低操作和维护成本。在PHM框架下，利用故障学方法预测设备未来的健康状况和故障时间。可以根据预测的故障时间优化维护活动，以提高可靠性并最大限度地降低生命周期成本。PHM具有关注特定于单元的健康状况信息的能力，具有显著提高可靠性和降低总体成本的巨大潜力。为了实现有效的PHM计划，需要在动力学和维护优化方面取得重大进展。 拟议研究计划的目标是开发准确，高效的先进性能和维护优化方法和工具，以支持有效的PHM计划，并将其应用于工程系统，特别是石油和天然气，可再生能源和运输系统。在拟议的计划中，我们计划通过利用物理模型和各种状态监测数据开发准确的集成故障诊断方法，基于深度学习工具的故障诊断方法，以及复杂系统的高效准确的基于状态的维护（CBM）优化方法。他们的应用旋转设备，特别是齿轮箱，和管道将进行调查。 预计拟议的研究计划将导致有效的方法和工具，准确的设备健康状况和剩余使用寿命预测，有效的维护优化，以及处理复杂设备和系统的能力。这项研究的结果将有助于实现高度可靠，安全和具有成本效益的工程系统，并将为加拿大石油和天然气，可再生能源，运输，制造业和航空航天工业带来重大利益。将培训高素质的人员。

For engineering systems such as oil and gas systems, pipeline systems and renewable energy systems, it is critical to ensure their reliability and reduce the overall cost so as to enhance system competitiveness, ensure safety, increase profits and protect the environment. Prognostics and health management (PHM) aims to improve system reliability and reduce operating and maintenance costs based on system health conditions. Under the PHM framework, prognostics approaches are utilized to predict equipment future health conditions and failure times. Maintenance activities can be optimized based on the predicted failure time to achieve improved reliability and minimized life-cycle cost. With the capability to focus on unit-specific health condition information, PHM presents great potential to significantly improve the reliability and reduce the overall cost. Significant advances are needed in prognostics and maintenance optimization for achieving effective PHM programs. The objective of the proposed research program is to develop accurate and efficient advanced prognostics and maintenance optimization methods and tools to support effective PHM programs, and apply them to engineering systems particularly oil and gas, renewable energy, and transportation systems. In the proposed program, we plan to develop accurate integrated prognostics methods by leveraging physical models and various condition monitoring data, prognostics methods based on deep learning tools, and efficient and accurate condition-based maintenance (CBM) optimization methods for complex systems. Their applications to rotating equipment, particularly gearboxes, and pipelines will be investigated. The proposed research program is expected to result in effective methods and tools for accurate equipment health condition and remaining useful life prediction, efficient maintenance optimization, and the capability to deal with complex equipment and systems. The results from this research will contribute to achieving highly reliable, safe and cost-effective engineering systems, and will generate major benefits for Canadian oil and gas, renewable energy, transportation, manufacturing and aerospace industries. Highly qualified personnel will be trained.