Failure prediction in power system components

电力系统组件的故障预测

• 批准号: 521914-2017
• 负责人: Hooshyar, Ali
• 金额: $ 1.82万
• 依托单位: York University
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=640044
• 关键词: Failure; prediction; power; system; components

Despite recent advances in power system protection technologies over the last decade, existing protectiverelays still operate only after a fault disrupts the healthy operation of a power system component. Therefore, theprotection technology has remained reactive in nature. This approach has a limit for reducing the outage timeand increasing the reliability of the grid. This project through a partnership between Professor Ali Hooshyarand partner organization, Hyperion Sensors, will attempt to fundamentally change this approach and developprotection technologies that predict a future failure in a power system component, e.g., an underground cable ora transformer winding. Therefore, the necessary maintenance to remove the root cause of that fault can becarried out before the end users experience power outage. The technique adopted by Hyperion Sensors toimplement this approach is based on distributed measurement of mechanical quantities (such as, temperatureand pressure) using a network of optical fibre. The optical fibre is distributed spatially alongside different partsof the power equipment, and operates both as the sensor and the communication medium to transfer themeasured quantities. The optical fibre can measure, e.g., temperature through a variety of principles, such asBrillouin principle.The challenges that are to be met for the realization of this technology are (i) identifying the appropriatemechanical quantity measured by the optical fibre, and (ii) finding the patterns in the measured quantities thatcan be used to predict a future failure. Hyperion Sensors is acquiring its first set of temperature measurementsfor an underground cable. The company will provide the data to the applicant to investigate differentwaveshape recognition and artificial intelligence techniques that can be used to address the above challenges.These techniques will be implemented in the protective device developed by Hyperion Sensors. Such aprotective device will reduce the outage time and increase the reliability of the power grid.

尽管在过去的十年中电力系统保护技术取得了进展，但现有的保护继电器仍然仅在故障破坏电力系统部件的健康运行后才运行。因此，保护技术在本质上仍然是反应性的。这种方法在减少停电时间和提高电网可靠性方面有一定的局限性。该项目通过Ali Hooshyar教授和合作伙伴组织CIMSensors之间的合作，将试图从根本上改变这种方法，并开发预测电力系统组件未来故障的保护技术，例如，地下电缆或Transformer绕组。因此，可以在最终用户经历停电之前进行必要的维护以消除故障的根本原因。采用的技术，由CIMSensors来实现这种方法是基于分布式测量的机械量（如，温度和压力），使用网络的光纤。光纤在空间上分布在电力设备的不同部分旁边，并且既作为传感器又作为通信介质来传输测量的量。光纤可以测量，例如，实现这种技术所要满足的挑战是（i）识别由光纤测量的适当的机械量，以及（ii）在测量的量中找到可以用于预测未来故障的模式。Sensors正在获取其第一套地下电缆的温度测量。该公司将向申请人提供数据，以研究可用于应对上述挑战的不同形状识别和人工智能技术。这些技术将在由WASTERN Sensors开发的保护装置中实现。这种保护装置将减少停电时间，提高电网的可靠性。