Research On Optimization Of ECT Image Reconstruction Algorithm

Electrical Capacitance Tomography(ECT)technology is a visual measurement method that can be used for two-phase flow or multi-phase flow condition monitoring.In this technology,the capacitance signal collected by the sensor is processed by mathematical reconstruction algorithm,and finally the distribution of dielectric constant in the measured area is obtained.ECT is widely used in parameter detection of two-phase flow or multiphase flow due to its advantages of non-invasiveness,non-damage,high safety and fast imaging speed.ECT system is mainly composed of capacitive sensor,acquisition and transmission system and image reconstruction software module.The image reconstruction algorithm has great influence on the final image reconstruction quality of ECT and directly determines the performance of the whole ECT system to some extent.Due to the serious nonlinearity,underdetermination and ill-condition in the solution process of ECT inverse problem,it has become a main factor restricting the high-quality reconstruction results of ECT.In this paper,the ECT image reconstruction algorithm is studied,and the algorithm is optimized from two different perspectives.In order to improve the solution accuracy and reduce the artifacts and distortion in the reconstruction results,two improvement schemes are proposed as follows:(1)Based on the idea of regularization,the mathematical model of ECT image reconstruction has been optimized and improved.Considering the sparsity,stability and robustness of the solution,the weighted L1-L2-norm is firstly introduced as a data fidelity item to enhance the anti-noise ability of solution.Secondly,considering the multiple characteristics of the imaging target,L1-Lp-L2-norm is introduced as a hybrid regularization term;at the same time,an algorithm based on the Alternating Direction Method of Multipliers(ADMM)algorithm framework is designed.The solution idea is to decompose the complex objective optimization problem into several sub-problems,and then use the Iterative Shrinkage Thresholding Algorithm(ISTA)and Iterative p-shrinkage algorithm(IPSA)to solve the sub-problems one by one.In order to verify the effectiveness of the proposed method,simulations of different flow patterns are carried out.Through the comparative analysis of the reconstruction results of different algorithms,it is verified that the algorithm can effectively reduce the artifacts in the reconstructed images and obtain images with clear edges and less distortion,and has better robustness to noise.(2)According to the intelligent optimization theory,an ECT image reconstruction method based on the Chaotic Simulated Annealing Fruit Fly Optimization Algorithm(CSAFOA)is proposed.The method first updates the odor concentration in the standard fruit fly algorithm to avoid the problem that the individual can only move around the origin,and the solution cannot be a negative value,which expands the range of motion of the fruit fly;at the same time,the dynamic step size is used instead of the fixed step size,so that the algorithm can maintain a larger search range in the early stage and a smaller search radius in the later stage,so as to avoid missing the optimal solution and improve the search efficiency of the algorithm.Secondly,in the process of fruit fly optimization,individuals are prone to fall into the local optimal solutions.The chaotic optimization algorithm and simulated annealing algorithm are used together,which can enhance its global optimization ability and further improve the solution accuracy.Finally,through the analysis and comparison of the numerical simulation results of different flow patterns,the effectiveness of the algorithm in optimizing the reconstructed image quality is verified.