A hybrid artificial bee colony algorithm and support vector machine for predicting blast-induced ground vibration

作     者：Zhu Chun Xu Yingze Wu Yongxin He Manchao Zhu Chuanqi Meng Qingxiang Lin Yun Zhu Chun;Xu Yingze;Wu Yongxin;He Manchao;Zhu Chuanqi;Meng Qingxiang;Lin Yun

作者机构：School of Earth Sciences and EngineeringHohai UniversityNanjing 210098China State Key Laboratory for Geomechanics&Deep Underground EngineeringChina University of Mining andTechnology(Beijing)Beijing 100083China State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal MinesAnhuiUniversity of Science and TechnologyHuainan 232001China College of Petroleum EngineeringXi’an Shiyou UniversityXi’an 710065China School of Resources and Safety EngineeringCentral South UniversityChangsha 410083China 

出 版 物：《Earthquake Engineering and Engineering Vibration》 (地震工程与工程振动（英文刊）)

年 卷 期：2022年第21卷第4期

页      面：861-876页

核心收录：

学科分类：08[工学] 081402[工学-结构工程] 081304[工学-建筑技术科学] 0813[工学-建筑学] 0814[工学-土木工程] 

基　　金：National Natural Science Foundation of China(NSFC)under Grant Nos.52104125 and 52104109 the Fundamental Research Funds for the Central Universities under Grant No.B220202056 the Opening Fund of State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines under Grant No.SKLMRDPC21KF04 the Natural Science Basic Research Plan in Shaanxi Province of China(2022JQ-304) the Fund of Young Elite Scientists Sponsorship Program by CAST under Grant No.2021QNRC001 

主　　题：recycled concrete frame-shear wall concealed bracings shaking table test nonlinear time-history response analysis 

摘      要：Because nearby construction has harmful effects,precisely predicting blast-induced ground vibration is *** this paper,a hybrid artificial bee colony(ABC)and support vector machine(SVM)model was proposed for predicting the value of peak particle velocity(PPV),which is used to describe blast-induced ground *** construct the model,5 potentially relevant factors,including controllable and uncontrollable parameters,were considered as input parameters,and PPV was set as the output ***-five samples were recorded from the Hongling lead-zinc *** ABC-SVM model was developed and trained on 35 samples via 5-fold cross-validation(CV).A testing set(10 samples)was used to evaluate the prediction performance of the ABC-SVM *** and four empirical models(United States Bureau of Mines(USBM),Amraseys-Hendron(A-H),Langefors-Kihstrom(L-K),and Central Mining Research Institute(CMRI))also were introduced for ***,the performances of the models were analyzed by using 3 statistical parameters:the correlation coefficient(R2),root-mean-square error(RMSE),and variance accounted for(VAF).ABC-SVM had the highest R2 and VAF values followed by the SVM,A-H,USBM,CMRI,and L-K *** results demonstrated that ABC-SVM outperformed SVM and the empirical predictors for predicting ***,the best results from the R2,RMSE,and VAF indices were 0.9628,0.2737,and 96.05%for the ABC-SVM *** sensitivities of the parameters also were investigated,and the height difference between the blast point and the monitoring station was found to be the parameter that had the most influence on PPV.