论文部分内容阅读
以矿山台阶爆破块度分布分析为例,建立了爆破块度预测的SVM模型。预测模型与实测结果相比,平均相对误差为6.67%,与BP神经网络模型、R-R分布及G-G-S经验模型相比,SVM模型预测爆破块度具有明显的优越性和可靠性。
Taking the analysis of block blasting block distribution as an example, a SVM model of block prediction is set up. Compared with the measured results, the average relative error is 6.67%. Compared with the BP neural network model, the R-R distribution and the G-G-S empirical model, the SVM model has obvious superiority and reliability in predicting the blasting degree.