Swift and precise classification of rock masses significantly enhances the prudent and efficient advancement of rock mass engineering. To expeditiously and accurately determine the engineering classification of slope rock masses, this paper introduces a hybrid Data Comprehensive Weight Analysis (DCWA)-Equilibrium Optimizer (EO)-AdaBoost ensemble prediction model. This model is formulated to prognosticate revised Basic Quality value ([BQ]) and the engineering classification of slope rock masses, utilizing the Basic Quality (BQ) classification method. Six elements have been chosen to formulate the prognostic index, and a dataset comprising information on the slope rock mass of 266 groups has been assembled for the purpose of training and evaluating the predictive accuracy of the established model. Two slope sections in DEZIWA open pit mine is chosen to validate the correctness of established DCWA-EO-AdaBoost model in the field through numerical simulation. In comparison to AdaBoost, DCWA-AdaBoost, DCWA-EO-AdaBoost, K-Nearest Neighbor (KNN), Back Propagation (BP), Decision Tree (DT), and Random Forest (RF) models, it is discerned that the DCWA-EO-AdaBoost hybrid model exhibits elevated coefficient of determination (R2, 0.986), variance accounted for (VA, 98.64%), prediction accuracy (A, 92.31%), and kappa coefficient (KC, 88.87%). Conversely, the mean absolute error (E, 5.97%) and root mean square error (ER, 38.37) are diminished, affirming its reliability and superiority. Field validation reveals that, combine the DCWA-EO-AdaBoost model and BQ classification method, mechanical parameters used for slope stability analysis can be obtained accurately, and can obtains slope stability analysis results basically consistent with the engineering practice through numerical simulation, which can prove the correctness of the established prediction model and clarify the use of the established model in the process of slope stability analysis. This attests that the established predictive model holds substantial merit, offering a noteworthy reference for the expeditious and precise acquisition of [BQ] and the classification of slope rock mass based on the BQ method, and can be adopted in the process of slope stability analysis.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Hybrid Prediction Model of Engineering Classification of Slope Rock Mass Based on DCWA-EO-AdaBoost Model and BQ Method


    Additional title:

    KSCE J Civ Eng


    Contributors:
    Wang, Han (author) / Gao, Yongtao (author) / Xie, Yongsheng (author) / Wu, Shunchuan (author) / Sun, Junlong (author) / Zhou, Yu (author) / Xiong, Peng (author)

    Published in:

    Publication date :

    2024-09-01


    Size :

    19 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English





    A prediction model of short-term ionospheric foF2 based on AdaBoost

    Zhao, Xiukuan / Ning, Baiqi / Liu, Libo et al. | Elsevier | 2013


    Driver braking intention prediction method based on AdaBoost integrated learning

    LIU LIFENG / LI ZHICHENG | European Patent Office | 2023

    Free access

    Vehicle detection combining gradient analysis and AdaBoost classification

    Khammari, A. / Nashashibi, F. / Abramson, Y. et al. | IEEE | 2005