A Smart Anomaly Detection Method in Cyber Physical Systems Using Machine Learning

Abstract

Cyber-Physical systems (CPS) are recognized by a broad range of complicated multi-tasking fixings with good interaction which results in combining cyber areas within the actual physical world. Thinking about the substantial development of cyber physical methods as well as a result of the prevalent utilization of sensible communication and features equipment, brand new issues have emerged. With this regard, a brand new model of CPSs for an intelligent power grid are confronting various vulnerabilities and lots of attacks and threats. Anomaly detection is a crucial information evaluation undertaking as among the techniques for CPSs protection. As various anomaly detection techniques are provided, it's tough to evaluate the pros and cons of the strategies. In this particular chapter machine learning (ML) techniques for detection of anomalies are provided by way of a situation learning that shows the usefulness of machine learning methods at classifying false data injection (FDI) strikes.