Abstract:
Unfavorable geological conditions such as karst cave often occur in the construction process of city subway and other railways, which seriously affect the construction progress and the staff’s safety. Therefore, in order to guarantee the normal construction, it is necessary to accurately detect the development features such as the location, size and filling properties of underground hidden karst caves. In this paper, the methods of forward and inversion numerical simulation on the basis of both finite element code and least square were used to construct the geo-electric model, by using three variables of underground caves, such as pitch, electrode spacing and relative hole spacing. Meanwhile, it shows the simulation results in a way of combining two-dimensional section with one-dimensional sounding curve, analyzes and summarizes the response characteristics and cross-hole resistivity CT for gas filled, water filled and partially water filled karst caves. The results indicate that the interface between water and gas in some water-filled karst caves is obvious, and the positions of the low-resistivity area and the high-resistivity area are the same as those of the water-filled and gas-filled parts of the karst caves. In addition, with the increase of filling water volume in the karst caves, the low-resistivity response of the karst caves increases, and the range of low-resistivity anomalies expands towards the top of the karst caves. The resistivity response characteristics of karst caves gradually decrease with the increase of electrode spacing and hole spacing, and the response attenuation of the hole spacing from 6 m to 16 m decreases more obviously than that from 16 m to 25 m. The response to electrode distance from 1 m to 2 m decreases less obviously than that from 0.5 m to 1 m. Meanwhile, the response of partially water filled karst caves decreases more obviously than that of single filled karst caves. When the electrode spacing and hole spacing increase to a certain range, the detection method can’t effectively identify complex karst caves and single filled karst caves; In the process that the karst cave model gradually deviates from the central position of the measured section, the response of the abnormal area weakens when it deviates up and down in the depth direction, and increases when it deviates towards left and right in the transverse direction. Various numerical models indicate that, when the ratio of hole spacing to hole depth and electrode spacing is appropriate, the cross-hole resistivity CT method can effectively identify the filling characteristics of the karst caves, and accurately determine the location and size of the karst caves, which provides a theoretical basis for the cross-hole exploration site and the interpretation of results.