Development And Application Of Surrounding Rock Classification System For Hydraulic Tunnel Based On Cloud Computing Technology

Surrounding rock classification is an important and key basic work in the design and construction of hydraulic tunnels,and it is an important basis for determining the stability of surrounding rock and its supporting methods.However,the existing classification methods of hydraulic tunnel surrounding rock are difficult to obtain classification indexes,long time period,and vulnerable to human factors.Therefore,in order to establish an accurate and efficient evaluation method of surrounding rock quality,this paper carried out the research on the cloud classification method of surrounding rock of hydraulic tunnel.In this paper,cloud computing technology,transfer learning technology and digital image processing technology are used to process and analyze the face image of hydraulic tunnel to determine the lithology and fracture coefficient of surrounding rock,so as to obtain the strength and integrity of surrounding rock,and the surrounding rock classification is carried out based on HC method.Finally,the Android Studio platform and Java language are used to develop the cloud classification system of surrounding rock of hydraulic tunnel,and the cloud classification results are compared with the actual engineering classification results.The main research results of this paper are as follows:(1)A lithology identification migration learning model based on Iception-ResNet-V2 network is constructed.A lithology identification model based on Iception-ResNet-V2 network is constructed by using migration learning technology and sub-image method.The accuracy of the model for lithology identification of granite,limestone,basalt and shale can reach more than 90%.Due to the unique structure,structure,texture and color of granite,the accuracy of the model for granite identification can reach more than 95%.(2)An automatic extraction algorithm of rock joint and fissure based on Canny operator and Otsu method is developed.Firstly,the face image is grayed and denoised.Secondly,the Canny operator is used to detect the edge of the image,and then the Otsu method is used to realize the segmentation of the main crack.Finally,the cloud calculation method is used to count the area and length of the surrounding rock joint cracks.(3)The classification method of hydraulic tunnel surrounding rock based on HC method is proposed.The correlation between lithology and rock strength,the area and length of joint fissures and rock mass fragmentation coefficient is established successively,so as to obtain the two main classification indexes of surrounding rock strength and integrity.Combined with the artificially obtained groundwater state,structural plane state and main structural plane occurrence correction indexes,the accurate classification of surrounding rock of hydraulic tunnel is realized based on HC classification method.(4)The surrounding rock classification system of hydraulic tunnel based on cloud computing technology was developed.The Android Studio platform and Java language are used to establish the client of surrounding rock classification of hydraulic tunnels.The lithology identification model and the automatic extraction algorithm of rock joints and fractures are deployed in the cloud server,and the interactive connection between the client and the cloud is realized.The cloud classification results were compared with the actual engineering classification results,and the classification accuracy was 88%,which verified the classification accuracy of the hydraulic tunnel cloud classification system.