Towards a Cloud-based Machine Learning for Health Monitoring and Fault Diagnosis
Towards a Cloud-based Machine Learning for Health Monitoring and Fault Diagnosis
复制标题
DOI：
10.36001/phmap.2017.v1i1.1843
复制
发表时间：
2017-07
期刊：
影响因子：
--
通讯作者：
Samir Khan;T. Yairi;Mariam Kiran
中科院分区：
文献类型：
--
作者：
Samir Khan;T. Yairi;Mariam Kiran
大型复杂的工程系统收集大量不同的数据集，使得通常难以处理和分析这些数据以诊断、隔离和预测操作期间的故障。要使用标准测试工具识别症状、推断潜在故障并最终诊断原因，需要持续的维护支持。这个问题在航空航天工业中尤为突出，在航空航天工业中，分析和维护资产以防止技术和人员的潜在故障或损失至关重要。最近使用的云计算提供了无限的计算资源，以快速处理和故障排除，减少了“修复时间”的问题。利用人工智能（AI）算法和云资源，可以帮助构建集成的故障诊断平台，为数据采集、处理和决策提供弹性和可扩展的资源。本文介绍了一个工业的角度和问题时，使用机器学习方法进行故障诊断，特别是在航空航天工业中使用云资源。特别注意的是支付的好处，与潜在的未来研究技术诊断被枚举。
Large complex engineered systems collect large amounts of varied data sets, making it often difficult to process and analyze these for diagnosing, isolating, and predicting faults during operation. To recognize symptoms with standard testing tools, infer potential faults and eventually diagnose causes needs constant maintenance support. This problem is particularly faced in the aerospace industry, where it is essential to analyze and maintain assets to prevent potential failures or loss both technological and human. Recent usage of Cloud computing provides infinite computing resources to quickly process and troubleshoot, reducing ‘time-to-fix’ problems. Exploiting artificial intelligence (AI) algorithms, with Cloud resources, can help build an integrated fault diagnostic platform to provide resilient and scalable resources for data acquisition, processing and decision making. This paper presents an industrial perspective and problems when using machine learning methods for fault diagnosis, particularly using Cloud resources in the aerospace industry. Special attention is paid to the benefits; with potential future research on technical diagnosis being enumerated.