Accurate prediction of compressive strength of rocks relies on the rate-dependent behaviors of rocks,and correlation among the geometrical,physical,and mechanical properties of rocks.However,these prop-erties may not be easy to control in laboratory experiments,particularly in dynamic compression ex-periments.By training three machine learning models based on the support vector machine(SVM),back-propagation neural network(BPNN),and random forest(RF)algorithms,we isolated different input parameters,such as static compressive strength,P-wave velocity,specimen dimension,grain size,bulk density,and strain rate,to identify their importance in the strength prediction.Our results demonstrated that the RF algorithm shows a better performance than the other two algorithms.The strain rate is a key input parameter influencing the performance of these models,while the others(e.g.static compressive strength and P-wave velocity)are less important as their roles can be compensated by alternative pa-rameters.The results also revealed that the effect of specimen dimension on the rock strength can be overshadowed at high strain rates,while the effect on the dynamic increase factor(i.e.the ratio of dy-namic to static compressive strength)becomes significant.The dynamic increase factors for different specimen dimensions bifurcate when the strain rate reaches a relatively high value,a clue to improve our understanding of the transitional behaviors of rocks from low to high strain rates.

Mingdong Wei;Wenzhao Meng;Feng Dai;Wei Wu

School of Civil and Environmental Engineering,Nanyang Technological University,Singapore;State Key Laboratory of Hydraulics and Mountain River Engineering,College of Water Resource and Hydropower,Sichuan University,Chengdu,610065,China

岩石力学与岩土工程学报（英文版）

[1]Mingdong Wei;Wenzhao Meng;Feng Dai;Wei Wu-.Application of machine learning in predicting the rate-dependent compressive strength of rocks)[J].岩石力学与岩土工程学报（英文版）,2022(05):1356-1365

bifurcate

Application,machine,learning,predicting,dependent,compressive,strength,rocks,Accurate,prediction,relies,behaviors,correlation,among,geometrical,physical,mechanical,properties,However,these,may,not,easy,control,laboratory,experiments,particularly,dynamic,compression,By,training,three,models,support,vector,back,propagation,neural,network,BPNN,random,forest,RF,algorithms,isolated,different,input,parameters,such,static,wave,velocity,specimen,grain,size,bulk,density,strain,identify,their,importance,Our,results,demonstrated,that,shows,better,performance,than,key,influencing,while,others,are,less,important,roles,compensated,by,alternative,also,revealed,effect,overshadowed,high,rates,increase,ratio,becomes,significant,factors,dimensions,when,reaches,relatively,value,clue,improve,our,understanding,transitional,from,low

0.486721