KAUST DepartmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Permanent link to this recordhttp://hdl.handle.net/10754/667740
MetadataShow full item record
AbstractAddressing anomaly detection and attribution is essential to promptly detect abnormalities, and it aids the decision-making of operators, allowing them to better optimize performance, take corrective actions, and maintain downstream processes. Recently, deep learning models have developed rapidly, especially in terms of their learning capabilities. In this chapter, we propose a novel hybrid deep-learning-based anomaly detection method. In particular, we focus on the benefits of deep learning models due to their greedy learning features and the sensitivity of clustering approaches to reveal anomalies in the monitoring process. In this chapter, we discuss and present applications of some deep-learning-based monitoring methods. We apply the developed approaches to monitor many processes, such as detection of obstacles in driving environments for autonomous robots and vehicles, monitoring of wastewater treatment plants, and detection of ozone pollution.
CitationHarrou, F., Sun, Y., Hering, A. S., Madakyaru, M., & Dairi, A. (2021). Case studies. Statistical Process Monitoring Using Advanced Data-Driven and Deep Learning Approaches, 255–303. doi:10.1016/b978-0-12-819365-5.00014-0
RelationsIs Part Of: