Real-Time Permittivity Extraction Based on Ultra-Wideband Impulse Radar and Genetic Algorithm for Tunnel Construction Operations

Tunnel construction with tunnel boring machi- nes (TBMs) is frequently hampered by hidden complexities in the tunnel face. This article investigates the feasibility of utilizing impulse ultra-wideband (UWB) radar technology to sense and probe the tunnel face in real-time and alert the operator of potential dangers. This study aims to identify buried objects, such as water mains and gas pipes, concrete structures, and naturally occurring objects like boulders present in the tunnel path. In this study, we considered different test objects extracted from tunnel construction projects in Edmonton and extracted their complex permittivity. A genetic algorithm (GA) is employed to retrieve the buried object’s electrical and physical parameters for near-real-time operations. It is shown that the dielectric constant and the thickness of the buried object are obtained within a maximum error of 14% and 2.5 mm absolute error, respectively. A more complex scenario involving hidden buried oil and sewer pipes in the tunnel face is evaluated. To assess various scenarios, buried pipes containing different fluids are considered, and measured their electrical properties within few seconds. Using the UWB radar, the fluids’ relative permittivity and loss tangents are estimated, along with the pipe’s physical dimensions and dielectric constant. These measurements distinguish the sewer pipe from the oil pipe. This study confirms the capability of UWB impulse radar technology for acquiring real-time geological conditions in tunnel construction.