Data Analysis for Condition Monitoring and Fault Detection in Induction Motors

三相感应电机是艾伯塔省许多工业的重要组成部分，因为它们是许多工业过程中具有成本效益的工具，特别是在艾伯塔省，因为石油和天然气，林业和石化工业。一般来说，工业部门60-80%的电力是由感应电动机消耗的。因此，从经济和环境的角度来看，感应电机的能源效率和可靠运行是一个主要的兴趣。电机故障减少了生产时间，造成了巨大的成本。因此，在艾伯塔省的工业中，大规模、连续监测设备完整性的潜在好处是显著的，包括：提高操作过程的效率；减少计划外停机时间；增加设备使用寿命；增强对失败原因的理解。该项目是先前Engage赠款在同一主题上的延续，新的重点是适合自定义低成本数据采集系统的数据测量和收集技术，而不是使用先前Engage赠款中所述的历史工厂。该项目的目的是提供一种比现有商业产品更具成本效益的方法，用于更大规模的状态监测。这将通过在给定约束条件下对从实验室设备收集的几个电特征实施数据分析技术来完成。这项工作的结果将为开发定制的低成本数据分析收集系统以及开发人工智能监测技术提供基础。希望本课题所开发的硬件和软件能够为低成本的大规模状态监测技术提供解决方案。

3-phase induction motors are a critical part of many Alberta industries as they are a cost effective tool for manyindustrial processes, especially relevant in Alberta because of the oil and gas, forestry and petro-chemicalindustries. Typically, 60-80% of electricity in the industrial sector is consumed by induction motors. Hence,energy efficiency reliable operation of induction motors are a major interest from both economic andenvironmental perspectives. Motors failures incur significant costs by decreasing production time. Therefore,the potential benefits in Alberta industries for large-scale, continuous monitoring of equipment integrity aresignificant, and including: increased efficiency in the operating process; reduction of un-planned downtimes;increased useful lifetime of equipment; enhanced understanding of failure causes. This project is a continuationof a previous Engage grant on the same topic with a new focus on data measurement and collection techniquessuitable for custom low cost data acquisition system rather than using historical plant as stated in the previousEngage grant. The intention of this project is to provide a more cost effective means, than existing commercialproducts, for condition monitoring on a larger scale. This will be accomplished by implementing data analysistechniques on several electrical signatures collected from laboratory equipment under given constraints. Theresults of this work will provide the foundation for both the development of a custom low cost data analysiscollection system, as well as the development of Artificial Intelligence monitoring techniques. It is the hopethat the hardware and software developed in this project will provide solutions to the problem of low cost largescale condition monitoring techniques.