煤炭工程 ›› 2024, Vol. 56 ›› Issue (1): 63-69.doi: 10.11799/ce202401010

• 施工技术 • 上一篇    下一篇

灰岩含水层定向钻孔注浆效果智能评价方法研究

李文昕,贾东秀,陈建刚,等   

  1. 1. 山东科技大学
    2. 山东能源新汶矿业集团有限责任公司 榆树井煤矿
    3. 青岛市黄岛区发展研究中心
    4. 邱集煤矿
    5. 山东省邱集煤矿有限公司
    6. 山东能源临沂矿业集团有限责任公司
  • 收稿日期:2022-10-08 修回日期:2023-03-03 出版日期:2024-01-20 发布日期:2024-01-29
  • 通讯作者: 李文昕 E-mail:1114369158@qq.com

Intelligent evaluation method of directional drilling grouting effect in limestone aquifer

  • Received:2022-10-08 Revised:2023-03-03 Online:2024-01-20 Published:2024-01-29

摘要: 基于黄河北煤田邱集煤矿现场注浆工程,通过XGboost、支持向量机、K近邻和BP人工神经网络四种算法,对灰岩定向钻孔的注浆效果进行科学分析。研究表明,XGboost、支持向量机、K近邻的模型精度均达不到09,而BP人工神经网络测试的拟合程度达093,使用现场数据测试准确度达09,证明了BP人工神经网络评价注浆效果的可行性和准确性|最后运用MATLAB提出了一种基于BP人工神经网络的注浆效果智能化评价方法,并根据模型制作了简单的演示平台,实现了注浆效果评价的智能化与快捷化。

关键词: 定向注浆, 效果评价, 机器学习, BP人工神经网络, 智能评价

Abstract: Grouting water plugging is one of the main means to prevent mine water disaster. Based on the grouting project of Qiuji Coal Mine in Huanghebei Coalfield, this paper scientifically analyzed the grouting effect of limestone directional drilling by XGboost, support vector machine, K-nearest neighbor and BP artificial neural network. The research shows that the model accuracy of XGboost, support vector machine and K nearest neighbor is less than 0.9, while the fitting degree of BP artificial neural network test is 0.93. The accuracy of field data test is 0.9, which proves the feasibility and accuracy of BP artificial neural network in evaluating grouting effect. Finally, an intelligent evaluation method of grouting effect based on BP artificial neural network is proposed by MATLAB, and a simple demonstration platform is made according to the model, which realizes the intelligent and rapid evaluation of grouting effect, and has a scientific guiding role for the development of mine grouting effect evaluation theory and coal safe mining.