Drones-based Thermography of Heat Distribution Networks

Figure 1: Thermal orthophoto. Blue lines represent heating network. Red marks represent detected thermal anomalies, which are in close proximity to the heating network.
Team:  A. Sledz, J. Unger, C. Heipke
Year:  2018
Funding:  AiF (Forschungsnetzwerk Mittelstand) / BMWi (Bundeministerium für Wirtschaft und Energie)
Duration:  2018 - 2019

In cooperation with: Fernwärme Forschungsinstitut Hannover (FFI) and industry partners like Enercity.

Project description

District heating systems distribute heat, which often is generated as waste heat during the production of electricity. For the integration of renewable heat sources, district heating networks are required to have a high degree of flexibility, especially in case of low-temperature heat distribution. The high technological relevance is faced by high costs for the construction, maintenance and repair of heat distribution networks.

Today’s maintenance strategy is based on a statistical assessment of the state of damage of the networks. Relevant pipeline sections must be gradually taken out of operation and emptied. This can be remedied by airborne thermal infrared imaging of heating networks that avoids interference with the operating process. This technology makes it possible to visualize temperature differences of a surface. Temperature anomalies indicate leaking district heating water and therefore damage to the district heating system.

Currently, airborne thermal flights can only be carried out efficiently over a large area with airplanes or helicopters and are limited due to cost considerations and high planning effort in particular for small-scale networks.

In this context, the use of Unmanned Aerial Systems (UAS) as a flexible and low-cost platform equipped with a thermal sensor is under investigation to detect the condition and further ageing of these networks. Thermal data acquisition using the UAS is followed by automated thermal mapping. This task is split into three steps: (i) the photogrammetric processing of the images, (ii) the description of the state of the heat supply network including possible changes and identification of anomalies by means of image analysis methods and (iii) anomalies classification to achieve false alarm rate as low as possible.