煤矿深部开采时围岩处于高应力状态，动力扰动可能导致巷道围岩失稳，甚至发生冲击地压；因此，探究动力扰动条件下巷道围岩变形的影响因素，对围岩控制具有重要意义。基于弹性力学和应力波理论，分析了巷道围岩失稳机理；采用正交试验法和FLAC^3D软件的动力学模块，探究了动力扰动条件下巷道围岩变形的影响因素。首先，根据极差、均值分析方法得到各因素对巷道围岩变形量的影响程度，由大到小依次为埋深、距离、扰动强度和扰动频率，埋深800 m、扰动强度20 MPa、距离9 m和频率5 Hz组合对巷道顶板及两帮变形量影响程度最大；埋深500 m、扰动强度5 MPa、距离18 m和频率20 Hz组合对巷道顶板及两帮变形量影响程度最小；其次，运用数值模拟方法分析了围岩位移、应力、塑性区和能量演化规律，验证了最大、最小组合的正确性；最终，通过多元线性回归数学模型对多因素作用下围岩顶板及两帮变形量进行预测，回归效果显著。

The surrounding rock is in a state of high stress during deep mining of coal mine, and the dynamic disturbance may lead to the instability of the surrounding rock and even rock burst. Therefore, it is of great significance to study the influencing factors of the deformation of roadway surrounding rock under dynamic disturbance. Based on elastic mechanics and stress wave theory, the instability mechanism of roadway surrounding rock is analyzed. By using the orthogonal experimental method and the dynamics module of FLAC^3D software, the influencing factors of roadway surrounding rock deformation under dynamic disturbance were investigated. Firstly, according to the method of range and mean value analysis, the influence degree of each factor on the deformation of roadway surrounding rock is obtained, from the largest to the smallest, it is buried depth, distance, disturbance intensity and disturbance frequency. The combination of buried depth of 800 m, disturbance intensity of 20 MPa, distance of 9 m and frequency of 5 Hz has the greatest influence on the deformation of the roof and two sides of the roadway. The combination of buried depth of 500 m, disturbance intensity of 5 MPa, distance of 18 m and frequency of 20 Hz has the least influence on the deformation of the roof and two sides of the roadway. Secondly, the displacement, stress, plastic zone and energy evolution of surrounding rock are analyzed by numerical simulation method, and the correctness of the maximum and minimum combination is verified. Finally, the multiple linear regression mathematical model is used to predict the deformation of surrounding rock roof and two sides under the action of multiple factors, and the regression effect is remarkable.