Application of supervised random forest paradigms based on optimization and post-hoc explanation in underground stope stability prediction

被引:3
作者
Li, Chuanqi [1 ]
Mei, Xiancheng [2 ]
Zhang, Jiamin [3 ]
机构
[1] Grenoble Alpes Univ, Lab 3SR, CNRS UMR 5521, F-38000 Grenoble, France
[2] Chinese Acad Sci, Inst Rock & Soil Mech, State Key Lab Geomech & Geotech Engn, Wuhan 430071, Peoples R China
[3] SINOPEC Res Inst Petr Engn, Beijing 100101, Peoples R China
关键词
Underground stope; Supervised random forest; Post -hoc explanation; Meta -heuristic optimization; GRAPH; MODEL;
D O I
10.1016/j.asoc.2024.111388
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This study applies a supervised random forest (SRF) paradigm to predict the underground stope stability. To improve the model prediction accuracy, 400 actual stopes with three main features (the optimized stability number (N), the hydraulic radius (HR), and a rock mass quality factor (Q)) are used to train the model, and nine meta -heuristic (MH) optimization algorithms are utilized to select the optimal hyperparameters of the SRF model. The prediction results show that the tiki taka algorithm -based SRF (TTA-SRF) model obtains the most satisfactory classification performance, with the smallest variance in the overfitting evaluation (Precision of 0.0201, Recall of 0.0257, Specificity of 0.0061, Accuracy of 0.0125, and F1 -measure of 0.0246). Furthermore, four post -hoc explainable techniques (i.e., model -independent) including the feature importance (FI), the Shapley additive explanations (SHAP), the partial dependence plot (PDP), and the individual conditional expectation (ICE) are adopted to explain the best prediction model. The results of the model interpretation show that all features are essential for predicting the stope stability and they have an opposite tendency on the stability prediction of stable and caved stopes. In particular, the limits at which HR, Q, and N do not contribute to the stope stability prediction are 18 m, 40, and 100, respectively. The results of optimization designs for a specific case stope based on the visualization program showed that the suggestions given by the prediction model are desirable. In conclusion, a high-performance and strongly explanatory prediction model is proposed in this study to facilitate the refinement of stope stability assessment in underground space.
引用
收藏
页数:21
相关论文
共 74 条
  • [1] Tiki-taka algorithm: a novel metaheuristic inspired by football playing style
    Ab Rashid, Mohd Fadzil Faisae
    [J]. ENGINEERING COMPUTATIONS, 2021, 38 (01) : 313 - 343
  • [2] Exponential distribution optimizer (EDO): a novel math-inspired algorithm for global optimization and engineering problems
    Abdel-Basset, Mohamed
    El-Shahat, Doaa
    Jameel, Mohammed
    Abouhawwash, Mohamed
    [J]. ARTIFICIAL INTELLIGENCE REVIEW, 2023, 56 (09) : 9329 - 9400
  • [3] Waterwheel Plant Algorithm: A Novel Metaheuristic Optimization Method
    Abdelhamid, Abdelaziz A.
    Towfek, S. K.
    Khodadadi, Nima
    Alhussan, Amel Ali
    Khafaga, Doaa Sami
    Eid, Marwa M.
    Ibrahim, Abdelhameed
    [J]. PROCESSES, 2023, 11 (05)
  • [4] The Arithmetic Optimization Algorithm
    Abualigah, Laith
    Diabat, Ali
    Mirjalili, Seyedali
    Elaziz, Mohamed Abd
    Gandomi, Amir H.
    [J]. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2021, 376
  • [5] Ant Lion Optimizer: A Comprehensive Survey of Its Variants and Applications
    Abualigah, Laith
    Shehab, Mohammad
    Alshinwan, Mohammad
    Mirjalili, Seyedali
    Abd Elaziz, Mohamed
    [J]. ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING, 2021, 28 (03) : 1397 - 1416
  • [6] Adoko A.C., 2023, P 57 US ROCK MECHANI
  • [7] A Feasibility Study on The Implementation of Neural Network Classifiers for Open Stope Design
    Adoko, Amoussou Coffi
    Saadaari, Festus
    Mireku-Gyimah, Daniel
    Imashev, Askar
    [J]. GEOTECHNICAL AND GEOLOGICAL ENGINEERING, 2022, 40 (02) : 677 - 696
  • [8] Machine learning-based models for predicting permeability impairment due to scale deposition
    Ahmadi, Mohammadali
    Chen, Zhangxin
    [J]. JOURNAL OF PETROLEUM EXPLORATION AND PRODUCTION TECHNOLOGY, 2020, 10 (07) : 2873 - 2884
  • [9] Bewick R., 2009, P 3 CANUS ROCK MECH, V1113
  • [10] Most Valuable Player Algorithm: a novel optimization algorithm inspired from sport
    Bouchekara, H. R. E. H.
    [J]. OPERATIONAL RESEARCH, 2020, 20 (01) : 139 - 195