Research On ERT Image Reconstruction Algorithm And Design Of The Application Software System

Electrical Resistance Tomography (ERT) technique is a high-new detection technique that has been developed from the eighties last century, it can be applied to measure the material parameters of multi-phase flow in a closed pipe or a process device, and it has a good foreground.Image reconstruction is the last step to realize visualization for ERT technique, and the reconstruction speed and the quality of the reconstructed images are two key factors. This paper is based on the principle of ERT technique, and carefully research has been made on image reconstruction algorithm, better reconstruction algorithms have been presented, and ERT simulation software kit has been designed.The main contribution and works are as follow:(1) For the forward problem of ERT, Finite Element Method is used to solve it. Data storage optimization method and field division optimization method are presented. The field is divided, and on the basic of the field division, the forward problem of ERT is solved.(2) For the inverse problem of ERT, that is, image reconstruction, research is done from two aspects: non-iterative image reconstruction algorithm and iterative image reconstruction algorithm. For non-iterative image reconstruction algorithm, regularization image reconstruction algorithm is revised and revised regularization image reconstruction algorithm is presented; for iterative image reconstruction algorithm, combined image reconstruction algorithm is presented.(3). A new evaluation standard for image reconstruction is presented, that is relative resistivity error to evaluate the image reconstruction effect.(4). An ERT application software system is developed with C++ Builder software development system and OpenGL development tool. The software system can solve the forward problem and the inverse problem according to different given parameter. With different given parameters, the field can be divide differently, and the marginal voltage can be calculated; with different image reconstruction algorithms, the image can be reconstruction differently; and the data can be stored and displayed.