Intelligent marine engines health assessment system based on digital twins and data driven models

Project: Research

Project Details


The i-HEATS project aims to develop an intelligent condition monitoring and
diagnostics system for marine engines based on first-principles digital twins
and data-driven models. The project has direct application to ship and land
power plants, endeavouring to provide a game-change in the condition
monitoring/diagnostics of internal combustion engines. I4.0 technologies will be
employed leading to the development of intelligent tools, including digital twins,
deep learning, sensors fusion, and cloud computing. The project key objectives
1. Installation of a novel, custom made, data acquisition system for measuring the
instantaneous shaft torque, integrated with the existing monitoring systems to
acquire the required performance parameters.
2. Storage and analysis of acquired data locally and on cloud (edge/cloud
3. Development of thoroughly validated engine first-principles digital twins,
integrating thermodynamic models with crankshaft/shafting system dynamics
4. Development of data-driven models based on deep learning techniques,
complementing the digital twins to offer real-time predictive capabilities.
5. Use of the developed data-driven model to identify inaccuracies of critical
sensors of the measured parameters and rectify those as needed.
6. Development of intelligent algorithms to monitor engine condition and provide
diagnostics based on the measured instantaneous torque and other critical
7. Develop the prototype i-HEATS system by integrating the above tools with
appropriate hardware and user interfaces
8. Extensive testing and verification of the prototype i-HEATS system in lab and
full-scale conditions onboard two ships and improvement of its functionality
leading to a pre-commercial version.
Short titlei-HEATS
Effective start/end date1/10/2131/07/23


  • marine engines
  • diagnostics system
  • first-principles digital twins
  • Machine Learning
  • artificial intelligence
  • data driven models
  • Health management


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.