Abstract: In recent years, in-seam wave technology was often used to detect hidden structures in coal seams. However, suitable excitation parameters have a great influence on the detection accuracy and effect. Taking the transportation along the 6-7 working face of the E1N4 section of Liujia coal mine as the research object, the optimal collection parameters of the area were selected through the multi-parameter excitation test and the reception effect test. Then, on this basis, a 3D model with faults was established, and 3D forward modeling was carried out to check the suitability of theoptimal parameters. Finally, the detection application was carried out on the 6-7 working face of E1N4 section. The results shown that: for a hole depth of 3m and an excitation dose of 300g in the study area, a better reflection trough wave could be obtained. The receiving method based on the anchor geophone was the best choice for receiving trough wave data. The forward modeling results were in good agreement with the model fault locations, which demonstrates the suitability of the optimal parameters. According to the optimized parameters, the detection practice was carried out, the existence of 2 faults was identified, the early fault information was corrected, and a new fault CF2 was discovered, which provided a reliable geological guarantee for safe production in the later stage, and provided a parameter basis for the through wave detection.