ThermoSense

Project Description

The ThermoSense project aims to develop and demonstrate an intelligent maintenance solution for district heating pipelines using advanced sensors, drone-based thermal imaging, and artificial intelligence. The system will predict the Remaining Useful Life of pipeline sections and detect early signs of damage—helping utilities avoid both premature replacements and large-scale, costly repairs within short timeframes.

The project will:

• Develop enhanced clamp-on sensors for monitoring flow, pressure events (e.g., water hammer), and anomalies in DH pipes.
• Build and fine-tune a multimodal Large Language Model (LLM) to interpret thermal images and pipeline data, enabling predictive maintenance through natural language interfaces.
• Validate the system through field tests in collaboration with Fjernvarme Fyn, Energi Ikast Varme, and Fjernvarme Horsens, and integrate it into existing software platforms from Danish companies ReMoni and Drone Systems.

Expected Green Outcomes:

ThermoSense supports the green transition by reducing unnecessary pipe replacements, detecting leaks early, and lowering heat loss in DH networks. This results in energy savings, reduced CO₂ emissions, and longer asset lifetimes, contributing to more sustainable and cost-effective DH infrastructure

Project Summary