长江科学院院报 ›› 2024, Vol. 41 ›› Issue (2): 135-141.DOI: 10.11988/ckyyb.20221277

• 岩土工程 • 上一篇    下一篇

基于MARS的岩石抗拉强度预测模型

徐国权1, 王鑫瑀2   

  1. 1.东华理工大学 地球科学学院, 南昌 330000;
    2.河北钢铁集团矿业有限公司,河北 唐山 063000
  • 收稿日期:2022-09-26 修回日期:2022-11-30 出版日期:2024-02-01 发布日期:2024-02-04
  • 作者简介:徐国权(1983-),男,辽宁鞍山人,讲师,博士,主要从事爆破智能化、采矿工程方面的研究工作。E-mail:xgq2017@gmail.com
  • 基金资助:
    国家自然科学基金青年基金项目(52008080)

Prediction Model for Tensile Strength of Rock Based on Multivariate Adaptive Regression Splines

XU Guo-quan1, WANG Xin-yu2   

  1. 1. School of Earth Science, East China University of Technology, Nanchang 330000, China;
    2. Hebei Iron & Steel Group Mining Co., Ltd., Tangshan 063000, China
  • Received:2022-09-26 Revised:2022-11-30 Online:2024-02-01 Published:2024-02-04

摘要: 将无损检测技术与机器学习相结合,通过建立预测模型来快速确定岩石抗拉强度已经成为热门研究方向之一。为了建立预测模型,提出一种基于多元自适应回归样条(MARS)的数据驱动建模技术,用于岩石抗拉强度预测。共收集了80组试验数据,包括施密特回弹数、干密度、点荷载强度指数以及巴西抗拉强度。所有数据被随机分为2个部分,其中70%的数据用于训练模型,剩余30%的数据用于测试模型性能。同时开发了人工神经网络、支持向量机和决策树3种数据驱动模型。选择了4种常用的模型性能评价指标,分别为均方根误差、平均绝对误差、相关系数和决定系数,以此来对所开发模型的预测性能进行比较。结果表明:所开发的智能模型均能够提供较高的预测精度,其中MARS模型性能优于其他3种模型,支持向量机和人工神经网络模型次之,决策树模型相对较差。值得一提的是,MARS模型能够通过方差分析来评估每个变量的相对重要性。研究成果有助于快速确定岩石抗拉强度。

关键词: 岩石力学, 抗拉强度, 多元自适应回归样条(MARS), 机器学习, 预测模型

Abstract: The prediction model which integrates non-destructive testing and machine learning has emerged as a hotspot for predicting tensile strength of rock. This paper presents a data-driven modeling approach for predicting rock’s tensile strength based on Multivariate Adaptive Regression Splines (MARS). An experimental dataset comprising 80 data sets, including Schmidt hammer rebound number, dry density, point load strength index, and Brazilian tensile strength, was collected. The dataset was randomly divided into training (70%) and testing (30%) sets. Additionally, artificial neural network (ANN), support vector machine (SVM), and decision tree (DT) techniques were utilized to develop machine learning models. The performance of the MARS model was compared with those of the DT, ANN, and SVM models. The predictive accuracy of the developed models was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MSE), coefficient of correlation, and coefficient of determination. The results manifested the satisfactory predictability of the machine learning models, with the MARS model exhibiting optimal performance, followed by SVM model, ANN model, and DT model in descending order. Notably, MARS was able to assess the relative importance of each variable through ANOVA decomposition. The model presented in this paper is conducive to rapidly obtaining the tensile strength of rock.

Key words: rock mechanics, tensile strength of rock, Multivariate Adaptive Regression Splines(MARS), machine learning, prediction model

中图分类号: