基于GRNN的金山店铁矿爆破振动峰值速度预测

发布时间：2018-04-26 01:04

本文选题：金山店铁矿 + GRNN　；参考：《爆破》2017年02期

【摘要】：为研究爆破振动对金山店铁矿地表构筑物和井下巷道的影响,引入广义回归神经网络(GRNN)的方法,分别以地表、井下部分振动监测数据为学习样本对GRNN进行训练,构建地表、井下爆破振动峰值速度的GRNN预测模型,以剩余振动监测数据为检测样本对地表和井下GRNN预测模型进行检验,并将GRNN模型的预测结果与BPNN、基函数回归法和经验公式法作对比。同时,将地表、井下GRNN模型的预测结果与以地表和井下综合训练数据为学习样本构建的综合GRNN预测模型进行对比。研究结果表明:对于地表监测点,四种方法的预测误差率依次为12.1%、18.9%、30.3%、43.7%;对于井下监测点,四种方法的预测误差率依次为14.0%、16.2%、19.9%、23.0%。GRNN的预测精度最高,其为爆破振动峰值速度的预测提供了一种新方法,且采用GRNN对地表、井下质点爆破振动峰值速度分别进行预测更加合理。
[Abstract]:In order to study the influence of blasting vibration on the surface structures and underground tunnels of jinjindian iron mine, the generalized regression neural network (GRNN) was introduced to train the GRNN in the ground surface and the underground part of the vibration monitoring data. The GRNN prediction model of the ground surface and the blasting vibration peak velocity of the underground mine was constructed, and the residual vibration monitoring data were taken as the data. The test samples are tested on the ground surface and the downhole GRNN prediction model, and the prediction results of the GRNN model are compared with the BPNN, the base function regression method and the empirical formula method. At the same time, the prediction results of the ground surface and the downhole GRNN model are compared with the comprehensive GRNN prediction model constructed with the ground surface and the underground comprehensive training data for the learning samples. The results show that the prediction error rates of the four methods are 12.1%, 18.9%, 30.3% and 43.7% for the surface monitoring points. For the underground monitoring points, the prediction error rates of the four methods are 14%, 16.2% and 19.9%, and the 23.0%.GRNN has the highest prediction accuracy. It provides a new method for the prediction of the peak velocity of blasting vibration, and uses GRNN to the surface and underground. It is more reasonable to predict the peak velocity of particle blasting vibration respectively.

【作者单位】：武汉科技大学资源与环境工程学院;
【基金】：国家自然科学基金面上项目(编号:51074115);国家自然科学基金青年项目(51204127) 湖北省自然科学基金重点项目(2015CFA142)
【分类号】：TD235.1

【相似文献】