台阶爆破参数的PCA与遗传算法比较寻优

Compared Optimization on Bench Blasting Parameters by PCA and Genetic Algorithm

针对大黑山钼矿过去台阶爆破综合效果较差的问题,进行了39组不同爆破参数组合的生产试验,将所取得结果进行线性插值综合评分,并且与主成分分析法的评价结果进行对比。通过遗传算法反向寻优,得到适应度最大的一组参数组合:底盘抵抗线为3.7m;孔距和排距分别为4.0,3.6m;孔距米延时取为10 ms/m;排距米延时取为16 ms/m。实际生产验证爆破效果综合得分为92.6分,取得了非常理想的效果,证实了优选参数的合理性。

For the poor comprehensive effect of bench blasting in Daheishan Mine, 39 groups of production tests with dif- ferent parameters were carried out. The results were scored comprehensibly by linear interpolation and compared with the evaluation results of PCA. Through reversed optimization of genetic algorithm, a group of parameters with the best fitness was achieved, including the chassis resistance line of 3.7 m, the pitch of 4.0 m, the row spacing of 3.6 m, the pitch delay of 10 ms/m and the row spacing delay of 16 ms/m. And in the actual production, the blasting effect, with good results, scored 92.6 points, which verified the ra- tionality of preferred parameters.