| In recent years,China has made great achievements in tunnel engineering construction.Tunnel boring machine(TBM)is a key equipment in the process of tunnel engineering construction.The parameter adjustment in its working process is closely related to the physical and mechanical parameters of the rock.If the physical and mechanical parameters of rock cannot be sensed in time,the working parameters of TBM cannot be adjusted in time,which will affect the tunneling efficiency and may cause adverse consequences.The current measurement methods of physical and mechanical parameters of rock have disadvantages such as poor timeliness and great environmental restrictions,which can not meet the needs of real-time detection.Through the establishment of TBM side block parameter automatic detection system can solve the above problems.By utilizing robotic vision system for automatic identification and detection of rock blocks,the limitations of harsh environments can be overcome,and real-time detection of rock block parameters can be achieved.Based on the above ideas,this thesis proposes an automatic rock block parameter detection method based on machine vision.The main research is as follows:(1)The overall framework of the automatic detection system for rock parameters was designed,dividing the entire system into three parts:rock acquisition and transportation,rock recognition and size detection,and rock point load testing.(2)For the rock recognition and size detection process,a rock recognition method that integrates target detection and semantic segmentation was proposed.Together with the deep camera and hand-eye calibration method,accurate positioning of the rock and measurement of key external dimensions were achieved.(3)For the rock point load testing process,a multi-position detection device was designed to improve the efficiency of rock detection.Through data statistics and analysis,the conversion formula for point load strength and uniaxial compressive strength of rock was fitted,and the prediction of uniaxial compressive strength was achieved through the point load strength.(4)To meet the linkage control requirements of the entire system,a control system composition scheme was designed.The selection and calculation analysis of key devices and equipment were carried out,the communication scheme of the entire system was designed,and the software system was also developed.(5)To verify the effectiveness of the design scheme,an experimental platform was built,and dynamic suction experiments with rocks,rock recognition and suction experiments,and point load testing experiments were designed and completed.The experimental results showed that each module could complete the expected function,which demonstrated the feasibility of the proposed automatic detection method for rock parameters in this thesis.With the rock parameters detection during TBM construction as the background,this thesis constructed an automatic detection system for rock parameters that integrates robotic vision system.The key technologies for rock recognition and size detection,as well as rock point load testing,were studied,providing a solution for the automatic detection of rock parameters during TBM construction.. |
