Abstract: In-seam seismic exploration has been widely used in underground geophysical exploration of coal mines because of its high resolution and long detection distance. This technology can detect the location and trend of faults, but it cannot realize the quantitative detection of fault displacement. In order to realize the quantitative identification of fault throw, from the perspective of theoretical derivation and numerical simulation, the influence of faults with different fault throws on the dispersion characteristics and energy distribution mode of Love trough wave is analyzed, and the influencing factors of energy distribution are analyzed. By quantitatively calculating the amplitude distribution curve and dispersion curve of Love wave and comparing with the theoretical data, the sensitive parameter of equivalent reflection coefficient ( the ratio of reflected in-seam wave energy to the total energy of reflected and transmitted in-seam wave ) is proposed, and the quantitative relationship between fault throw and the change rate of equivalent reflection coefficient with wavelength is derived. The adaptive range of the quantitative relationship is analyzed by the in-seam wave data of numerical simulation. The verification of the numerical model shows that the parameter has good detection accuracy when the fault distance is greater than 1 m. When the fault distance is less than 1 m, the detection accuracy may decrease due to the error in energy picking and calculation. Finally, through the test of measured in-seam wave data, it is found that the equivalent reflection coefficient is not only affected by the fault displacement, but also by the fault dip angle and other factors. The absolute error between the fault displacement predicted by the quantitative relationship and the actual exposed fault displacement is 0.58 m, and the relative error is 12.9%. The predicted fault distance of the measured data is basically consistent with the exposed fault distance, which verifies the effectiveness of the method. This method can provide technical support for safe and efficient production of working face and accurate geological modeling of subsequent coal mining face. However, this method is limited by fault dip angle and observation system in application. In the case of large fault dip angle and difficulty in receiving transmitted channel wave, the accuracy will be further reduced. The next step will study fault dip angle detection and reduce the dependence of this method on observation system.