HeatTrack: Enhanced Reliability, Monitoring and Diagnostics of Complex Cooling Systems through Advanced Thermal Management



Start date:


End date:


This project will focus on studying and developing optimization, monitoring and diagnostics methods and tools dedicated for complex industrial cooling systems. A healthy cooling system will strengthen the resilience of a complex energy system. However, a defective cooling system can dramatically reduce the overall system resilience.
To respond to the high cooling demand in industry e.g., electronic systems or energy systems, new complex cooling solutions are being developed every day to ensure that all components of a given industrial installation should not operate beyond the acceptable temperature range.
Tough requirements regarding temperature limits are particularly imposed on sensitive components as electronics. To ensure that the temperature remains within the desired limits, the designers often consider the worst case and consequently over dimension the cooling system to be able to provide the required cooling capacity when the thermal load is at its highest value. Control systems have been successfully added to reduce the energy consumption and CO2 emissions which are already dramatically high in our planet. These developments are still not sufficient to reach the ambitious targets of carbon neutrality by 2050.
Non controlled temperature levels could lead to faults with dramatic situations e.g., unavailability of the electric grid creating a local or entire city black-out with consequent economic, social and environmental damage. A higher reliability and efficiency of our energy systems cannot be reached without highly reliable and efficient cooling solution.
This is why we propose to do much more to enhance the performance of the complex cooling systems in industry and make them highly reliable.
This will be done through a total integration of the cooling in the plant details by considering all interacting internal and external/ambient parameters including all transient behavior due to thermal load change, due to ambient conditions or due to any other internal or external variation.
To reach this target, this project will: i) Collect the available measurement and simulation data, introduce new sensing technologies as infrared (IR) camera sensing to provide input for monitoring and diagnostics. ii) Implement high fidelity modeling coupled with continuous and on-demand measurements through a proper integration of Artificial Intelligence (AI) technologies as Physics Informed Neural Networks (PINNs). iii) Research and development of adaptive methods and tools based on machine learning (ML) and deep learning (DL) through information fusion. This will guarantee a continuous assessment of the leakage status and other key important events as well as a review of the remaining useful life (RUL) of the sensitive cooling system components.
Early prediction of the defects in the cooling system allows quick and easy fix and avoids operations shutdown. This maintains the system resilience capability and extends the lifetime of the industrial process.
The scientific and technical community will strongly benefit from the methods and inventive steps developed within the project. The industrial partners will use the project results and guidelines to enhance their system resilience through integrating and improving their cooling solutions to the overall process.

Hitachi Energy Industrial
LedAi AB Industrial

Shahina Begum, Professor

Email: shahina.begum@mdu.se
Room: U1-089
Phone: +46-21-107370