Coal and gas outbursts are complex dynamic disasters that occur near coal mining faces. In recent years, coal and gas outburst accidents induced by external vibrations have become increasingly frequent. Understanding the dynamic changes in ground stress and the characteristics of gas migration near the working face during various coal mining operations is essential for predicting coal and gas outbursts and providing early warnings of potential dynamic hazards. Although various geological prediction methods, such as direct current detection, ground-penetrating radar, seismic exploration, and transient electromagnetic methods, are widely used, they still have limitations in the real-time monitoring of changes in the surrounding rock stress. To explore the qualitative and quantitative relationships between artificial acoustic signals generated by mechanical vibrations and the corresponding stress, strain, and gas pressure in coal, as well as to enhance students’ practical scientific research skills, a teaching experiment on artificial acoustic signals induced by mechanical vibrations was conducted.

A self-developed device for testing artificial acoustic signals based on mechanical vibrations was used to conduct acoustic signal excitation tests under varying axial loading stress and gas pressure values. The device comprised five main units: gas charging and exhaust units, a mechanical vibration unit, a vibration-force monitoring unit, an axial-pressure loading unit, and an artificial acoustic signal monitoring and acquisition unit. Standard coal samples with a diameter of 50 mm and a height of 100 mm were prepared, and uniaxial compression tests were conducted to determine their basic mechanical parameters. During the experiment, a pendulum was set to swing freely from a fixed angle of 20°, and acoustic signals were collected every 10 s under different axial stresses (10~40 kN) and gas pressures (0~1.2 MPa). The relationship between the spectral characteristics of the artificial acoustic signals and the axial loading stress and gas pressure was analyzed and fitted using the introduced relative stress coefficient K.

Before the axial stress reached the uniaxial compressive strength of the coal sample without gas, the sample underwent compaction and elastic stages. The original cracks closed, and a few new microcracks were generated. No macroscopic cracks appeared, and the failure was not obvious. During this phase, the K value gradually increased with increasing axial loading stress, and the increasing trend gradually slowed down. Once the axial stress exceeded the peak value, cracks in the coal sample propagated, macroscopic cracks developed, and the K value began to decrease. When the gas pressure was 0.9 and 1.2 MPa, a different trend was observed: before the axial loading stress reached the uniaxial compressive strength of the gas-free coal sample, the K value showed a decreasing trend. The coefficient of determination (R²) for the fitting function relating K to the axial stress before coal body failure exceeded 98.53%. Before the instability and failure of the coal sample, as the axial stress gradually increased, the primary cracks inside the coal body closed, and only some microcracks appeared. The signal source propagated well within the coal body, eventually leading to a gradual increase in the K value with increasing axial stress. The R² of the fitting function relating K to the gas pressure before coal body failure exceeded 98.94%.

By examining the relationship between the spectral characteristics of artificial acoustic signals and the parameters of coal and gas outbursts, this experiment enhances students’ autonomous learning and problem-solving abilities, stimulates academic interest, cultivates academic thinking, strengthens team cohesion, and establishes a solid foundation for future research.