基于潜孔钻机钻进参数的岩石硬度预测方法研究
发布时间：2019-02-24 19:49
【摘要】：岩石硬度是矿山建设与生产的重要参数之一,对矿山开采工艺选择、设备选型以及爆破参数设计起到重要作用。利用潜孔钻机钻进参数快速预测岩石硬度,明确沿炮孔轴向的岩石硬度分布情况,对炸药单耗量、装药间隔位置等爆破参数选择具有指导作用,进而改善爆破效果,降低生产成本,同时也充实矿山地质信息,对智能化钻机的研制以及矿山开采的数字化、智能化建设具有重要意义。通过对KQG150Y潜孔冲击钻机的结构与其工作方式的分析得出与岩石硬度密切相关的钻进参数指标。运用设计的钻进参数采集系统对大明山石灰石矿潜孔钻机的钻进参数进行样本数据采集,采用最小二乘曲线拟合、逐步回归法以及BP神经网络对岩石硬度预测进行分析研究并建立五个岩石硬度预测模型,并对模型的拟合结果与预测效果进行分析比较。研究表明：在岩石硬度预测模型中,以冲击频率为自变量的岩石硬度预测模型拟合优度很高但预测能力很弱。以回转扭矩、凿岩速度为特征向量的BP神经网络模型具有良好的泛化能力,预测效果能够达到预期要求。运用BP神经网络方法能够实现岩石硬度快速和准确的预测,为大明山石灰石矿爆破设计提供基础数据,为露天开采中的岩石硬度预测提供方法借鉴。
[Abstract]:Rock hardness is one of the important parameters in mine construction and production, which plays an important role in mining process selection, equipment selection and blasting parameter design. The drilling parameters of submersible drilling rig are used to predict rock hardness quickly and determine the distribution of rock hardness along the axial direction of the hole, which can guide the selection of blasting parameters such as explosive unit consumption, charge interval and so on, and then improve the blasting effect. It is of great significance to reduce the production cost and enrich the geological information of the mine, which is of great significance to the development of intelligent drilling rig and the digitization and intelligent construction of mining. By analyzing the structure and working mode of KQG150Y drilling rig, the drilling parameters related to rock hardness are obtained. The drilling parameters of downhole drilling rig in Daming Mountain Limestone Mine were collected by the designed drilling parameter acquisition system, and the least square curve was used to fit the sample data. The method of stepwise regression and BP neural network were used to analyze and study the prediction of rock hardness. Five prediction models of rock hardness were established, and the fitting results of the model were compared with the results of prediction. The results show that in the rock hardness prediction model, the rock hardness prediction model with impact frequency as the independent variable has a high goodness of fit but a weak prediction ability. The BP neural network model with rotating torque and rock drilling velocity as the characteristic vector has good generalization ability and the prediction effect can meet the expected requirements. The BP neural network method can be used to predict rock hardness quickly and accurately, to provide basic data for blasting design of Daming Shan limestone mine, and to provide reference for prediction of rock hardness in open-pit mining.
【学位授予单位】：辽宁工程技术大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD422.1
本文编号：2429875
[Abstract]:Rock hardness is one of the important parameters in mine construction and production, which plays an important role in mining process selection, equipment selection and blasting parameter design. The drilling parameters of submersible drilling rig are used to predict rock hardness quickly and determine the distribution of rock hardness along the axial direction of the hole, which can guide the selection of blasting parameters such as explosive unit consumption, charge interval and so on, and then improve the blasting effect. It is of great significance to reduce the production cost and enrich the geological information of the mine, which is of great significance to the development of intelligent drilling rig and the digitization and intelligent construction of mining. By analyzing the structure and working mode of KQG150Y drilling rig, the drilling parameters related to rock hardness are obtained. The drilling parameters of downhole drilling rig in Daming Mountain Limestone Mine were collected by the designed drilling parameter acquisition system, and the least square curve was used to fit the sample data. The method of stepwise regression and BP neural network were used to analyze and study the prediction of rock hardness. Five prediction models of rock hardness were established, and the fitting results of the model were compared with the results of prediction. The results show that in the rock hardness prediction model, the rock hardness prediction model with impact frequency as the independent variable has a high goodness of fit but a weak prediction ability. The BP neural network model with rotating torque and rock drilling velocity as the characteristic vector has good generalization ability and the prediction effect can meet the expected requirements. The BP neural network method can be used to predict rock hardness quickly and accurately, to provide basic data for blasting design of Daming Shan limestone mine, and to provide reference for prediction of rock hardness in open-pit mining.
【学位授予单位】：辽宁工程技术大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD422.1
【参考文献】
中国期刊全文数据库 前10条
1 贺建平;鲁娜;;浅谈国内外牙轮钻机的现状及发展趋势[J];机械管理开发;2012年06期
2 何俊海;赵希春;朱金堂;黄少华;宿连军;;全自动岩石可钻性测试仪的研制[J];石油矿场机械;2011年08期
3 周杨洲;王佰顺;;气动潜孔锤硬质砂岩钻进技术研究[J];煤炭技术;2011年05期
4 张思渊;刘皓;袁春娥;;钻井参数判断岩石可钻性问题的研究与讨论[J];山西建筑;2010年23期
5 寻知锋;余继峰;;聚类和判别分析在测井岩性识别中的应用[J];山东科技大学学报(自然科学版);2008年05期
6 王克雄;郭清;朱益;;测井资料在岩石硬度预测中的应用[J];煤田地质与勘探;2006年05期
7 张作清,韩克宁,于代国,张振城;应用支持向量机方法预测储层敏感性[J];测井技术;2005年04期
8 ;An optimizing selection model in preparation of powdery emulsion explosives[J];Journal of University of Science and Technology Beijing(English Edition);2004年01期
9 曹正林,周丽清,王志章,张祥忠;利用钻井参数进行随钻岩性预测[J];天然气工业;2000年01期
10 赵秀菊,李春成,李祖奎,燕静,翟应虎,王克雄;利用声波速度预测地层岩石硬度的试验研究[J];西南石油学院学报;1999年03期
，本文编号：2429875
本文链接：https://www.wllwen.com/kejilunwen/kuangye/2429875.html
