Leakage identification and numerical simulation of embankment dams based on infrared thermal imaging

Timely and accurate identification of leakage is vital for ensuring the safe operation of embankment projects. However, few studies have considered utilizing information related to the temperature-field anomalies on dam surfaces obtained through infrared thermal imaging in the leakage detection process. In this context, to explore the feasibility and characteristics of infrared thermal imaging in detecting leakage in embankment dams under natural environmental conditions, this study fully considered the correlation between leakage and temperature, and on-site inspection tests were conducted for dam leakage detection based on infrared thermal imaging. By studying the coupling theory of the seepage and temperature fields, a numerical simulation study was designed to investigate the effects of water–air temperature difference and seepage-channel permeability coefficient on the surface temperature field of a downstream dam. The results showed that the infrared images captured by the imager onboard our unmanned aerial vehicle could vividly and intuitively present temperature anomalies caused by the leakage, with the lowest temperature located at the leakage outlet point. The temperature development law and infrared image characteristics of the leakage outlet of the earthen dam were consistent with the numerical simulation results. A temperature difference rate index was proposed using model calculations: the higher the temperature difference rate, the more evident the temperature anomaly in the downstream leakage area of the dam. It is advisable to choose a temperature difference value greater than 5°C between the air temperature and the reservoir water for testing while characterizing the degree of anomaly in the temperature field on the surface of small reservoirs. These findings can provide a scientific basis for leakage detection based on infrared thermal imaging using unmanned aerial vehicle platforms.