A Scalable Sensor Network for Pipeline Vibration Analysis

用于管道振动分析的可扩展传感器网络

• 批准号: 571908-2022
• 负责人: Nikolaidis, IoanisI
• 金额: $ 2.91万
• 依托单位: University of Alberta
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=750463
• 关键词: Scalable; Sensor; Network; Pipeline; Vibration

The pipelines used in the mining, oil and gas industry deteriorate over time. Pipeline maintenance is expensive and often based on best guesses of the condition of individual pipeline segments. We propose to use a network of wireless sensors that monitor and report the vibrations sensed at each pipeline segment. The relative amplitudes of the vibrations at various frequencies, and across the length of a pipeline, form a vibration profile for the pipeline. We will develop the sensor platform and protocols for the acquisition, communication, and study of the vibration profiles and their changes over time as they reflect the changing condition of the pipeline. Using the platform, pipeline operators will be able to determine trouble spots as early as possible. Machine learning techniques will be used to classify the profiles as normal or problematic, to infer the most likely location of recent change that could mean need for maintenance, and to determine the points in time where the profiles change to a degree that warrants inspection and maintenance. We will use advanced data compression and aggregation techniques adapted to the limited computation and communication resources of the wireless sensors nodes. We will integrate a widely available Internet-of-Things (IoT) protocol, LoRaWAN, with 5G IoT protocols to aggregate and relay the sensed data in compressed form to monitoring stations. With the help of the partner organizations, we will test the platform and its networking aspects to ensure that it meets the performance envelope of the heterogeneous, LoRaWAN and 5G, communication environment. The overall objective is to automate the process of identifying, early and effectively, pipeline segments in need of maintenance, thus reducing the corresponding operating costs.

采矿、石油和天然气行业使用的管道会随着时间的推移而老化。管道维护是昂贵的，并且通常基于对各个管道段的状况的最佳猜测。我们建议使用一个无线传感器网络，监测和报告在每个管道段感测到的振动。在各种频率下以及在管道的长度上的振动的相对振幅形成管道的振动轮廓。我们将开发传感器平台和协议，用于采集、通信和研究振动曲线及其随时间的变化，因为它们反映了管道的变化状况。使用该平台，管道运营商将能够尽早确定故障点。机器学习技术将用于将配置文件分类为正常或有问题，以推断可能需要维护的最近变化的最可能位置，并确定配置文件变化到需要检查和维护的程度的时间点。我们将使用先进的数据压缩和聚合技术，以适应有限的计算和通信资源的无线传感器节点。我们将广泛使用的物联网（IoT）协议LoRaWAN与5G IoT协议集成，以压缩形式将传感数据聚合并中继到监测站。在合作伙伴组织的帮助下，我们将测试该平台及其网络方面，以确保其满足异构LoRaWAN和5G通信环境的性能要求。总体目标是使尽早和有效地确定需要维修的管道段的过程自动化，从而降低相应的运营成本。