基于SOA-RF智能算法的深部巷道支护参数优化研究

Optimization of support parameters for deep roadway based on SOA-RF intelligent algorithm

强开采扰动作用下深部沿空巷道围岩稳定性控制是一项技术难题。为了研究深部沿空巷道在动载荷作用下位移及应力场的演化特征,以高家堡煤矿103工作面为背景建立数值模型。然后,利用FLAC 3D软件分析了动载作用下深部沿空巷道位移及应力场演化特征。最后,通过机器学习的方法,进行了巷道响应参数敏感性分析。结果表明:根据重要性得分,不同动荷载因素作用下巷道围岩对动荷载幅值响应最为明显;不同动载幅值作用下巷道最大变形速率达到17.5 mm/MPa,两帮最大变形速率达到16.1 mm/MPa,塑性发育最大速率达到3552.5 m^(2)/MPa。利用回归分析方法得到了动载荷条件下的经验方程。在此基础上,提出了“让压-支护”协同控制技术,以缓解沿空巷道围岩变形。数值模拟与现场监测结果验证了协同控制技术的可行性。

Stability control of the surrounding rock of gob-side entry in deep mine under strong mining disturbance is a technical challenge.To study the evolution characteristics of the displacement and stress field in deep roadway along gob under the dynamic load,the numerical model was established based on No.103 working face of Gaojiabao Coal Mine.Then,the displacement and stress field evolution characteristics in deep roadway along gob under dynamic load were analyzed by FLAC 3D software.Finally,the sensitivity analysis of the roadway response parameters was carried out by the machine learning method.The results show that,according to the importance ranking,the roadway surrounding rock has the most obvious response to the dynamic loading amplitude under different dynamic loading factors;under different dynamic loading amplitudes,the maximum deformation rate of the roadway reaches 17.5 mm/MPa,the maximum deformation rate of the two sides reaches 16.1 mm/MPa,and the maximum rate of plasticity development reaches 3552.5 m^(2)/MPa.By means of regression analysis method,the empirical equations under different dynamic loading conditions were obtained.On this basis,the cooperative control technology of“yield pressure-support”was proposed to alleviate the deformation of the surrounding rock in roadways by gob.The results of numerical simulation and field monitoring verified the feasibility of the cooperative control technology.