Anomaly Detection and Diagnosis in IIoT Systems: A Review of Techniques and Applications.

Abstract

Real-time monitoring and control of industrial systems are now feasible thanks to the IIoT. As a result, massive amounts of data have been produced, which may be used to better understand how the systems work. However, it may be challenging to find and analyse problems in data acquired by IIoT due to the richness and variety of the data. The continuing safety, dependability, and efficiency of IIoT systems rely on their capacity to identify and diagnose anomalies. Complex methods like machine learning algorithms, deep learning models, or hybrid strategies are needed for this purpose. In this research, we discuss the advantages and disadvantages of the various methods and approaches used for anomaly detection and diagnosis in IIoT systems. We also discuss the challenges and potential of this area of study, as well as their consequences. We also assess commonly used datasets and benchmarking systems, as well as real-world applications in a wide range of sectors, to gauge the efficacy of anomaly detection strategies. The purpose of this research is to offer a comprehensive overview of the essential themes in order to facilitate the development and deployment of effective and efficient anomaly detection and diagnostic systems in industrial settings.