Real-time perception of rock–machine interaction information in TBM tunnelling using muck image analysis

Abstract

Tunnel boring machine (TBM) is frequently used in tunnel construction due to its advantages in safety, high efficiency, environmental friendliness. However, the tunnelling efficiency of a TBM heavily depends on the rock–machine interaction conditions. To realize the real-time perception of rock–machine interaction information during the TBM tunnelling, an automatic acquisition and analysis system for rock muck is developed. Firstly, a framework for using the system to perceive the rock mass and TBM conditions was constructed. Then the method to quantitatively evaluate the geometric properties of rock muck is elaborated. Finally, the system was applied to a TBM-excavated tunnel to verify the functions of rock muck in the real-time perception of rock–machine interaction information. The results show that the geometric properties of rock muck can indicate both the rock mass conditions and the TBM performance. In rock mass condition perception: Rock blocks trend to be large in size and elongated in shape as the uniaxial compressive strength (UCS) of the surrounding rock decrease. The blocks produced from a relatively intact rock mass trend to be small in size and round in shape. Larger or smaller blocks with flat planes are frequently produced when TBM drives in the jointed rock masses. In TBM performance perception: The TBM utilization can be estimated through the ratio of the number of images with muck to the total number of captured images. A high value of the field penetration index (FPI) corresponds to blocks with large sizes and elongated shapes (or small sizes and round shapes). Elongated blocks are produced when TBM drives at a low FPI. These results are significant for the safe and efficient TBM tunnelling.