摘要
对光面爆破效果的精准预测可以减小爆破对围岩造成的损伤,提高围岩的稳定性和爆破开挖后的安全性。针对传统预测算法计算代价大、易陷入局部最优解等缺点,依托承平高速南火道特长隧道工程,选取围岩单轴抗压强度、埋深、节理裂隙发育程度等作为算法中光爆效果预测的输入影响参数,拱顶下沉、最大线性超挖量和超欠挖面积为输出参数,对光面爆破效果进行预测,通过PSO-LSSVM、GWO-LSSVM和PSO-BP预测模型预测参数,利用GA算法嵌入到PSO算法中,建立粒子群和遗传算法优化最小二乘支持向量机(GA-PSO-LSSVM)进行对比。研究结果表明:通过训练得到的GA-PSO-LSSVM光爆效果预测模型预测得到的拱顶下沉、最大线性超挖和超欠挖面积相对误差的平均值分别为0.38%、0.67%和1.61%,GA-PSO-LSSVM预测模型在对拱顶下沉、最大线性超挖和超欠挖面积等3个效果参数的预测上较其他算法具有更显著的优越性,是最佳的光面爆破效果预测模型。
In order to achieve accurate prediction of the smooth blasting effect,reduce the damage caused by blasting to the surrounding rock,and improve the stability of the surrounding rock and the safety after blasting and excavation,in view of the shortcomings of the traditional prediction algorithm,such as high calculation cost and easy to fall into the local optimal solution,relying on the Nanhuo Tunnel project of the Chengde-Pinggu Expressway,the uniaxial compressive strength,buried depth,and joint fracture development degree of the surrounding rock are selected as the input influencing parameters for the prediction of the light explosion effect in the algorithm.The maximum linear overmining amount and over-undermining area are used as output parameters,and the GA algorithm is embedded into the PSO algorithm to establish particle swarm and genetic algorithm to optimize the least squares support vector machine(GA-PSO-LSSVM)and predict the smooth blasting effect,and compared with the PSO-LSSVM,GWO-LSSVM and PSO-BP prediction models.The results show that the average relative errors of the vault subsidence,maximum linear over-excavation and over-under-excavation area predicted by the trained GA-PSO-LSSVM light burst effect prediction model are0.38%,0.67%and 1.61%,and the GA-PSOLSSVM prediction model has significant superiority over other algorithms in predicting the three effect parameters of vault subsidence,maximum linear over-excavation and over-under-excavation area,and is the best prediction model for smooth blasting effect.
作者
殷胜光
高成梁
高鹏
王举阳
费春伟
YIN Sheng-guang;GAO Cheng-liang;GAO Peng;WANG Ju-yang;FEI Chun-wei(Road&.Bridge International North China Engineering Co.Ltd.,Beijing 10oo24,China)
出处
《公路》
北大核心
2025年第11期58-64,共7页
Highway
关键词
隧道工程
钻爆法
隧道围岩
光面爆破
预测模型
效果参数
tunnel engineering
drilling and blasting method
surrounding rock of tunnel
smooth surface blasting
prediction model
effect parameters
