A Smart Condition Monitoring System for HV Networks with Artificial Intelligence, Augmented Reality and Virtual Reality: Copyright Material IEEE, Paper No. PCIC-2018-37

The authors present a conceptual design for a SMART asset monitoring solution for high voltage (HV) networks in the petrochemical industry. The paper discusses the potential for incorporating artificial intelligence (AI), augmented reality (AR) and virtual reality (VR) into an application of the Industrial Internet of Things (IIoT) for condition monitoring. The paper is a continuation of the work presented by the authors at the IEEE-PCIC 2017 conference in Calgary. The proposed asset management system analyses condition monitoring (CM) data and assesses the risk of failure data across complete HV networks. Knowledge of deteriorating asset condition provides the operator with an advanced, early warning of incipient mechanical and electrical faults. With knowledge of the severity and source of such faults, pinpointed preventative maintenance interventions can then made during planned maintenance outages. The complete HV network asset monitoring solution described includes permanent sensors and monitoring nodes deployed at strategic locations across the network. Processed data is passed via a local area network to local servers and then via secure data cloud transmission to a centralized monitoring server located at the CM headquarters. This central server operates a CM database that logs, displays, benchmarks and trends the condition data with comparison to a statistically-significant database of measurements. It is proposed in the IIoT solution proposed that this database will be downloadable to a smartphone/tablet for use by the field engineer. The monitoring technology will likely also incorporate a number of AI machine learning software modules for the de-noising of raw signals and the diagnosis of different types of defects within different types of HV plant items. The proposed SMART CM system includes an advanced graphical user interface (GUI) for viewing HV asset CM data along with operational and maintenance (O&M) data. The GUI will also be able to display both condition criticality and operational criticality (on a color-coded range of 0-100%) for individual HV plant items on a digitized mimic of the HV network's single-line diagram (SLD). This could also be combined with geometric positioning data of assets across the facility (including HV cable routes and lengths) to provide a fully digitized SMART network diagram for use in the IIoT asset management solution. Asset management data, combined with the application of the developing techniques of AI, AR and VR, will greatly help the user to visualize the plant items in 3-D, their position within the network, their condition and operational criticality along with all related asset management information together on one dashboard screen, downloaded onto smartphone/tablet. The paper concludes with a case study showing the development of a specification for a SMART IIoT asset condition monitoring solution suitable for a large petrochemical refining facility.