Experimental Research On Rock Strength Acquisition Based On Real-Time Monitoring Of Mechanical Parameters Of Rock Drill

With the development of tunnel construction towards less manpower and mechanization,large-scale mechanized supporting construction represented by drilling jumbo gradually replaces traditional manual construction.The drilling jumbo is equipped with a large number of sensors and data acquisition systems,which can realize construction informatization,monitor and record working parameters such as percussion pressure,rotation speed,penetration rate,etc.In order to make effective use of a large amount of construction data and play its active role in the intelligent construction of tunnels,This paper relies on the on the construction in progress of Chi-Huang high speed railway Ling-shang Village Tunnel,and comprehensively uses the research methods such as literature research,field test,statistical analysis,etc.,to carry out the experimental research on rock strength acquisition based on real-time monitoring of the drilling parameters of the rock drill.The main work and conclusions are as follows:(1)Combined with the theoretical analysis and investigation of the working principle and actual situation of the rock drill,percussion pressure,thrust pressure,rotation pressure and penetration rate are selected as drilling parameters related to rock strength.It is found that percussion pressure,thrust pressure,rotation pressure are all positively correlated with rock strength,while penetration rate is negatively correlated with rock strength;(2)Combined with literature research,rock rebound test and point load test are determined to quickly obtain the uniaxial saturated compressive strength of rock on site.A total of 15 sets of comparative tests of rebound method,point load method and uniaxial saturated compressive failure are completed.It is found that both the rebound value and point load strength have a good correlation with the uniaxial saturated compressive strength,and the accuracy of the point load estimation of the uniaxial saturated compressive strength is slightly higher than that of the rebound method;the accuracy of the point load method for estimating the uniaxial saturated compressive strength is slightly higher than that of the rebound method;(3)Drilling tests are carried out on rock-like materials and tunnel face,46 sets of sample libraries including drilling parameters and rock strength are obtained.Based on the traditional hard computing regression analysis method and the soft computing LSSVM machine learning algorithm,various rock strength prediction models are established respectively.It is found that when only a single drilling parameter is considered,the prediction accuracy of the drilling rate model is higher,and when multiple drilling parameters are considered,the prediction accuracy of the models are further improved,indicating that considering multiple drilling parameters can effectively improve the rock strength prediction accuracy.Both the multivariate nonlinear regression model and the LSSVM algorithm model based on multiple drilling parameters can be applied to predict rock strength in actual construction;(4)Three application ideas of rock strength prediction model are put forward.It is found that the uniaxial saturated compressive strength distribution of the rock mass on tunnel face basically conforms to the normal distribution law.Surrounding rock intelligent classification program based on real-time acquisition of drilling parameters of rock drill is assisted in the development,and it is applied to engineering.