Matches in SemOpenAlex for { <https://semopenalex.org/work/W4380368500> ?p ?o ?g. }
Showing items 1 to 62 of
62
with 100 items per page.
- W4380368500 endingPage "1686" @default.
- W4380368500 startingPage "1686" @default.
- W4380368500 abstract "摘要: 岩石抗压强度是评估岩体工程稳定性的重要力学参数,传统统计回归方法对于岩石抗压强度预测存在一定的局限性.为此,提出了一种利用简单岩石力学参数实现岩石抗压强度智能预测的方法,首先收集了620组含不同类型岩石的三轴试验数据,然后分别采用随机森林(Random Forest,RF)、极限梯度提升树(XGBoost,XGB)和轻量梯度提升机(LightGBM,LGB)3种主流的集成学习算法建立了岩石抗压强度预测模型,使用贝叶斯优化算法在模型训练过程中进行超参数优化,最后利用决定系数(R2)、平均绝对百分比误差(MAPE)和均方根误差(RMSE)对优化后模型的泛化能力进行了综合评估和对比分析.此外,利用LGB模型对输入特征进行重要性分析,以评估不同输入特征对模型泛化性能的影响重要程度.研究结果表明:所建立的3种模型对岩石抗压强度均取得了较好的预测结果,其中LGB模型泛化性能优于另外两种模型(R2=0.978,RMSE=5.58,MAPE=9.70%),且运行耗时相对最少.弹性模量(E)、围压(σ3)和密度(ρ)对模型的泛化性能影响较大,泊松比(v)影响较小.提出的预测模型对于岩石抗压强度预测有良好的适用性,为机器学习与岩土工程的结合提供了新的思路. 关键词: 岩石强度 / 集成学习 / 贝叶斯优化 / 随机森林 / 极限梯度提升树 / 轻量梯度提升机 / 工程地质" @default.
- W4380368500 created "2023-06-13" @default.
- W4380368500 creator A5003699837 @default.
- W4380368500 creator A5006247366 @default.
- W4380368500 creator A5031172404 @default.
- W4380368500 creator A5033897388 @default.
- W4380368500 creator A5052590910 @default.
- W4380368500 creator A5057202277 @default.
- W4380368500 creator A5060186258 @default.
- W4380368500 creator A5086855227 @default.
- W4380368500 date "2023-01-01" @default.
- W4380368500 modified "2023-09-24" @default.
- W4380368500 title "基于集成学习与贝叶斯优化的岩石抗压强度预测" @default.
- W4380368500 cites W1157421975 @default.
- W4380368500 cites W1985207154 @default.
- W4380368500 cites W2020391390 @default.
- W4380368500 cites W2021205107 @default.
- W4380368500 cites W2061536117 @default.
- W4380368500 cites W2067395819 @default.
- W4380368500 cites W2094199595 @default.
- W4380368500 cites W2116940203 @default.
- W4380368500 cites W236012661 @default.
- W4380368500 cites W2567139121 @default.
- W4380368500 cites W2789758093 @default.
- W4380368500 cites W2911964244 @default.
- W4380368500 cites W2972899043 @default.
- W4380368500 cites W2974668914 @default.
- W4380368500 cites W2981931085 @default.
- W4380368500 cites W3036047683 @default.
- W4380368500 cites W3037314178 @default.
- W4380368500 cites W3049652382 @default.
- W4380368500 cites W3154176209 @default.
- W4380368500 cites W4206256427 @default.
- W4380368500 cites W4206367093 @default.
- W4380368500 cites W4230428500 @default.
- W4380368500 cites W4248497251 @default.
- W4380368500 doi "https://doi.org/10.3799/dqkx.2023.029" @default.
- W4380368500 hasPublicationYear "2023" @default.
- W4380368500 type Work @default.
- W4380368500 citedByCount "0" @default.
- W4380368500 crossrefType "journal-article" @default.
- W4380368500 hasAuthorship W4380368500A5003699837 @default.
- W4380368500 hasAuthorship W4380368500A5006247366 @default.
- W4380368500 hasAuthorship W4380368500A5031172404 @default.
- W4380368500 hasAuthorship W4380368500A5033897388 @default.
- W4380368500 hasAuthorship W4380368500A5052590910 @default.
- W4380368500 hasAuthorship W4380368500A5057202277 @default.
- W4380368500 hasAuthorship W4380368500A5060186258 @default.
- W4380368500 hasAuthorship W4380368500A5086855227 @default.
- W4380368500 hasBestOaLocation W43803685001 @default.
- W4380368500 hasConcept C192562407 @default.
- W4380368500 hasConceptScore W4380368500C192562407 @default.
- W4380368500 hasIssue "5" @default.
- W4380368500 hasLocation W43803685001 @default.
- W4380368500 hasOpenAccess W4380368500 @default.
- W4380368500 hasPrimaryLocation W43803685001 @default.
- W4380368500 hasVolume "48" @default.
- W4380368500 isParatext "false" @default.
- W4380368500 isRetracted "false" @default.
- W4380368500 workType "article" @default.