Research On ERT Image Reconstruction And Application In Concrete Damage Monitoring

As a new nondestructive testing technology,electrical resistance tomography(ERT)can monitor the distribution of medium in the object by a non-invasive method,which has been widely used in the fields of structural health monitoring,medical imaging,etc.In this paper,research will be carried out on two aspects of algorithm optimization and acquisition system construction.By applying excitation to a concrete block,the boundary voltage of the concrete is collected,and the electrical conductivity distribution image of the concrete’s internal medium is obtained by an image reconstruction algorithm.The main research questions are as follows:(1)The reconstruction images of the Tikhonov regularization algorithm under different defects are analyzed and compared.The non-negative constrained Tikhonov regularization algorithm is proposed for the ERT problem by literature investigation,which can effectively improve the "over-smoothing" of the reconstructed image through a large number of iterations.In order to improve the imaging speed,the adaptive idea is introduced,and the non-negative constrained ATikhonov regularization algorithm is proposed.By imposing more detailed penalties on the regularization matrix,the iterative efficiency of the algorithm is greatly improved.Compared with LBP,Newton,Landweber,Tikhonov,and CG algorithms,the results indicate that the relative error of the improved algorithm is far less than that of the above-mentioned traditional algorithms,and better imaging results can be obtained.(2)The NC-ATikhonov regularization algorithm needs more accurate prior information,but for the concrete environment with complex medium distribution and high impedance,the prior information provided by the empirical method cannot guarantee the quality of the reconstructed image.For this reason,the particle swarm optimization(PSO)is introduced to automatically optimize the prior information of the Tikhonov algorithm,the inertia weight of the particle swarm is adjusted,and the segmented function is combined with the adaptive algorithm.On the basis of improving the global search ability of the initial iteration of the PSO,the better local search ability is retained,and the search efficiency of the PSO is improved.The reconstructed images indicate that the algorithm can ensure imaging quality and enhance the applicability of the algorithm.(3)Building of ERT system.The ERT system is divided into a hardware acquisition system and an upper computer operating system.The NI Multisim is used to simulate each module of the hardware acquisition system to verify the feasibility of the system.Then,according to the characteristics of the high impedance of the concrete block,a potentiometer is introduced to adjust the output current based on the constant current source circuit to ensure the stability of the output excitation current.The host computer’s operating system is designed by APP Designer,which can not only control the communication of the lower computer but also reconstruct the image in the upper computer,greatly simplifying the operation process.(4)The channel consistency experiment is carried out in a saltwater environment,and the collected voltage curve showed a U-shaped distribution,which was in line with the experimental expectation.In order to improve the electrical conductivity of concrete,a certain amount of carbon fiber was added to the concrete test block,and the single-defect concrete test block was used as the experimental object.The reconstruction image effect of the DPSO-NC-ATikhonov algorithm is better,which could roughly reflect the size and location of the defect.