Creation of LIVE Digital Twin for Prognostics and Predictive Maintenance of Rotating Machineries (LIVE for 2PM)

创建用于旋转机械的预测和预测性维护的实时数字孪生（LIVE for 2PM）

• 批准号: 570439-2021
• 负责人: Barari, Ahmad
• 金额: $ 12.39万
• 依托单位: University of Ontario Institute of Technology
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=723390
• 关键词: Creation; LIVE; Digital; Twin; Prognostics

Due to the pandemic, the regular in-person inspection of the critical Rotating Machineries (RM) in hydropower, wind energy, and transportation were interrupted. Unfortunately, RM typically deteriorate their balance condition dynamically, due to wear, thermal bending, dirt collection, insulation failure, partial discharge, fault currents, thermal stress due to overloading, etc., and gradually develop concerning vibratory unbalance responses affecting RM performance and lifespan. Failure of RM in applications such as turbomachines, electromotors, generators, pumps, jet fans, etc. creates enormous unscheduled repair costs and more importantly may put human life in danger. This project aims to develop methodologies required to employ the concept of LIVE Digital Twin for Prognostics and Predictive Maintenance (LIVE for 2PM) of RM highly critical assets. The term "LIVE" refers to the first letters of the four phases in the methodology, i.e. Learn, Identify, Verify, and Extend. The key feature in LIVE digital Twins is multi-physics simulation solutions developed in alternative levels and modes, which are supported by artificial neural network and deep learning tools. LIVE Digital Twin allows Learning the behavioral patterns of individual RM considering their specifications, environment and customized factors. It Identifies the locations and types of sensors and the sensing procedures to Verify the possibilities for progressively increasing defects. It Extends the life of RM by advising predictive maintenance avoiding its failure. Implementation of LIVE Digital Twin not only reduces the needs for in-person inspection during the pandemic, but also can result in significant improvement in asset management of the critical Canadian infrastructures for energy and transportation sectors with predictive maintenance, reduced cost of repair, reduced frequency in replacing parts, shorter down-time of the corresponding highly critical infrastructures, reduced waste, and increased life-span of critical machines. All of these result in significant economic growth for a chain of industries and also a noticeable welfare of the society.

受疫情影响，水电、风能及交通运输等关键旋转机械的定期现场检查中断。不幸的是，由于磨损、热弯曲、灰尘收集、绝缘故障、局部放电、故障电流、由于过载引起的热应力等，RM通常动态地恶化其平衡条件，并逐渐发展关于振动不平衡响应影响RM的性能和寿命。RM在诸如涡轮机、电动机、发电机、泵、射流风扇等应用中的故障产生巨大的计划外维修成本，更重要的是可能使人的生命处于危险之中。该项目旨在开发所需的方法，采用实时数字孪生的概念，用于RM高度关键资产的预测和预测性维护（2PM实时）。术语“LIVE”是指方法中四个阶段的首字母，即学习、识别、验证和扩展。LIVE Digital Twins的主要特点是以替代级别和模式开发的多物理场仿真解决方案，这些解决方案由人工神经网络和深度学习工具支持。LIVE Digital Twin允许学习单个RM的行为模式，考虑他们的规格，环境和定制因素。它描述了传感器的位置和类型以及传感程序，以验证逐渐增加的缺陷的可能性。它通过建议预防性维护来延长RM的使用寿命，避免其故障。LIVE Digital Twin的实施不仅减少了疫情期间亲自检查的需求，而且还可以通过预测性维护、降低维修成本、减少更换零件的频率、缩短相应高度关键基础设施的停机时间、减少浪费，并延长关键机器的使用寿命。所有这些都为一系列产业带来了显着的经济增长，也为社会带来了显着的福利。