Study On The Non-destructive Testing And Identification Method For Typical Diseases Of Tunnel Lining

The nondestructive testing technology represented by ground penetrating radar(GPR)is widely used in tunnel lining quality inspection because of its convenient operation and strong anti-interference ability,but it also limits the development of ground-penetrating radar technology because of the subjective strong subjectivity and low identification accuracy of image interpretation.Therefore,it is urgent to carry out the research on identification methods of typical tunnel lining diseases by non-destructive testing,to achieve the objective and accurate identification of tunnel lining diseases,and to provide research ideas and technical support for the future research on image interpretation.In this paper,the automatic identification of typical lining diseases is realized by means of theoretical analysis,physical test and field measurement,and the main work and research results are as follows:(1)The characteristics and causes of typical lining diseases are systematically analyzed.Through consulting literature and field investigation,this paper makes a systematic study on the characteristics and causes of two typical diseases of tunnel lining cavity and non-compaction,and in combination with engineering practice,gives the accurate definition of lining cavity and non-dense disease,and analyzes and explains the relationship between the two.The above analysis and research provide a preliminary theoretical basis for the construction of physical model test and automatic identification of lining diseases in this paper.(2)Based on the physical model test,the optimum time for detecting typical lining diseases is obtained.Based on the systematic study of typical disease characteristics and genesis of tunnel lining,the physical model test is constructed by combining Maxwell's classical electromagnetic theory.By comparing the test block radar images in the same age period,it is found that the radar images with a maintenance time of 17 days are the clearest,so the best time to detect the typical diseases of lining is the 17th day.(3)The characteristics of radar images under different conditions of typical lining diseases are obtained.By comparing the radar images of two shielding antennas,it is found that the 800MHz shielding antennas have better identification effect on the typical diseases of lining.Based on the study of radar images obtained from 800MHz shielded antennas,the image characteristics of hollow and non-dense diseases in many cases are obtained,which provides theoretical guidance and data support for the subsequent automatic identification of typical diseases.(4)An automatic identification method for detecting typical diseases of tunnel lining by ground penetrating radar is proposed.Based on the experimental acquisition of radar data,the characteristics of signal are extracted in time domain and frequency domain respectively,and three identification features,such as maximum amplitude,energy and spectrum standard,are obtained.On this basis,a two classification model is established by combining Support Vector Machine theory to distinguish disease signals.An automatic identification method for detecting typical diseases of tunnel lining by ground-penetrating radar is proposed.The results show that the identification accuracy of this set of two diseases is more than 85%,which can provide reference for the quality inspection of tunnel lining in the future.(5)Application of engineering examplesBased on the project of the Honghualiang Tunnel in Yanchong Expressway in the Winter Olympics,nondestructive testing research is carried out on the basis of the obtained automatic identification method.After verification,it is found that the test results are basically consistent with the tunnel site situation,which shows that the automatic identification method of typical disease of tunnel lining proposed in this paper has high reliability and can be applied to the field of tunnel lining quality inspection in the future.