A review of leakage detection methods for district heating networks

To investigate effective and low-cost leakage detection (LD) methods for urban district heating (DH) networks with an extensive literature review, this paper first classifies the existing methods based on their technology characteristics into three categories: physical model-based methods, data-driven methods and unmanned airborne infrared thermography (UAIT) methods, then analyzes and discusses these three methods in detail. Considering the similarity between oil/gas pipelines and water distribution networks to DH networks, the common LD methods for oil/gas and water applications are briefly summarized at the beginning. Key findings from this critical literature review include: (1) a need exists to explore new approaches for the physical model-based method to obtain a more accurate network model; (2) exploring new algorithms for the data-driven method to analyze network leakage through inadequate information is suggested; (3) optimizing the deployment of monitoring instruments is recommended, and (4) the UAIT method needs research into effective algorithms of classification and recognition, automatic post-processing, and obtaining high-quality images of varying site environments. Furthermore, a hybrid LD strategy is proposed to integrate the physical model-based and data-driven methods to adapt different characteristics of the primary network and secondary network with combining SCADA (Supervisory Control and Data Acquisition) monitoring data.