Deep Recurrent Graph Convolutional Architecture for Sensor Fault Detection, Isolation, and Accommodation in Digital Twins

The rapid adoption of Internet-of-Things (IoT) and digital twins (DTs) technologies within industrial environments has highlighted diverse critical issues related to safety and security. Sensor failure is one of the major threats compromising DTs operations. In this article, for the first time, we address the problem of sensor fault detection, isolation, and accommodation (SFDIA) in large-size networked systems. Current available machine-learning solutions are either based on shallow networks unable to capture complex features from input graph data or on deep networks with overshooting complexity in the case of large number of sensors. To overcome these challenges, we propose a new framework for sensor validation based on a deep recurrent graph convolutional architecture which jointly learns a graph structure and models spatio-temporal interdependencies. More specifically, the proposed two-block architecture 1) constructs the virtual sensors in the first block to refurbish anomalous (i.e., faulty) behavior of unreliable sensors and to accommodate the isolated faulty sensors and 2) performs the detection and isolation tasks in the second block by means of a classifier. Extensive analysis on two publicly available datasets demonstrates the superiority of the proposed architecture over existing state-of-the-art solutions.

I. Introduction

Under the umbrella of Industry 4.0, digital twins (DTs) have garnered striking interest over the last few years through the process of industry digital transformation [1]. Fundamentally, a DT can be defined as a digital profile that mirrors a physical object or process, i.e., the physical twin (PT), and provides a bidirectional interaction between the physical and digital parts. Leveraging DTs, operators can simulate complex systems behavior, test/predict asset changes in specific scenarios, and remotely control/monitor/steer systems.