Submitted by admin on Mon, 08/08/2022 - 00:07

Leakages leading to safety hazards and negative environmental impact

Sequence Number 2 Industry Energy Banner Leakages leading to safety hazards and negative environmental impact How 5G enabled

Real-time transfers of high-definition imagery data and asset maintenance data can help identify potential leaks/emissions in the infrastructure

Data Flows
Title Devices Icon Devices Description
  • Cameras (sensors) collect data (e.g., images, video, corrosion, damage, leakage, emission [GHG])
  • Cameras: Fixed; Robotics; Drones; etc.
Title Connectivity Icon Connectivity Description
  • Mainly imagery and time series data transport
  • Asset maintenance data
Title Edge Compute Icon Edge Compute Description
  • Not time critical at present, however this could change to real-time in case of new emission laws
Title Cloud Compute & Storage Icon Cloud Compute & Storage Description
  • Historical video real-time sensor data from multiple cameras
  • Loaded and kept in Enterprise storage
Title Applications & Services Icon Applications & Services Description
  • Machine Vision (MV) to identify potential leaks, on-site safety hazards and security threats
  • All detection is fully automatic
Title Inform Decision Makers
Icon Inform Decision Makers Description
  • Alarm-based reporting when leaks, emissions, hazards, etc. are detected
Title Support Decision Making Icon Support Decision Making Description
  • End of process
Application Logic
Description
  • Industrial assets such as civil structures (e.g., oil rigs, drilling equipment, etc.) retrofitted with IoT Camera sensors to watch critical areas and gather image data
  • Dependent on the scope of predictive maintenance, multiple cameras will be fitted (fixed/drones/robotics/etc.) to collect all necessary data. It will be an iterative process to get cameras in the right positions.
  • If there is a lack of data, look at options of creating and using synthetic data
Description
  • Image data collected by the cameras will be stored long-term in Enterprise storage
  • AI (ML) application and AI (MV) application requests the data it needs for model development from Enterprise storage to copy in its own storage
  • Development of the ML model is done through an iterative process, and a quality ML model (fully data-driven) will require multiple steps to detect anomalies/potential failures
  • Model will be stored and maintained by AI application
Description
  • Process from data collection to execution of models is fully automated
  • Only alerts operations when anomalies are detected in results (leaks/emissions/spillage)
  • It will take multiple steps to get to the proper recognition of problems
  • SME input is required to identify leakages, emissions, etc.
  • Various options for visualisation of entire operations (e.g., XR options on OpenXR platform)
Expected benefits Key value created