Research On Image Reconstruction Algorithm Of Electrical Capacitance Tomography

Electrical capacitance tomography(ECT)is one of electrical process tomography technologies.This technology obtains the cross-sectional image of the fluid in the measured field based on the measured capacitance value.Because ECT technology is characterized by non-radioactive,non-invasive,lower costs and visualization,etc.It has been widely concerned and studied by scholars at home and abroad.At present,the research on ECT technology mainly focuses on the design of capacitance sensor,the optimization of data acquisition system and image reconstruction algorithms,etc.In this thesis,the research object is the electrical capacitance tomography system of 8-electrode and the image reconstruction algorithms.By comparing and analyzing various image reconstruction algorithms,aiming at the typical problems of Landweber algorithm in ECT image reconstruction,an improved gradient accelerated Landweber algorithm is proposed.In order to obtain high-quality reconstruction images,particle swarm optimization algorithm is used in ECT image reconstruction.The main contents of this thesis include:(1)The principles,advantages and disadvantages of several traditional image reconstruction algorithms are compared and analyzed.The calculation formula of the traditional algorithm is derived based on the principle model of ECT image reconstruction.The performance of traditional algorithms is compared and analyzed through numerical simulation experiments.The analysis results show that Landweber algorithm is a good tradeoff between the speed and accuracy of image reconstruction.(2)Aiming at the problem of slow and unstable convergence of Landweber algorithm.Firstly,this thesis has analyzed the performance of several accelerated Landweber algorithms,and then has deeply studied the gradient accelerated Landweber algorithm with better performance according to the series theory.A new iterative formula is obtained by constructing residual matrix and adding constraint factors.Finally,an improved gradient accelerated Landweber algorithm is proposed and applied.Numerical simulation results show that the proposed algorithm can obtain stable convergence characteristics and improve the speed and accuracy of reconstructed image.(3)In order to obtain high-quality reconstructed images,particle swarm optimization algorithm is used in ECT image reconstruction.Firstly,based on particle swarm optimization algorithm,nonlinear convergence strategy of inertia weight factor is adopted to adjust particle speed,and nonlinear particle swarm optimization algorithm is proposed.Then,the nonlinear particle swarm optimization algorithm is improved by using the search strategy of sine cosine optimization algorithm.A nonlinear particle swarm optimization algorithm based on sine cosine search strategy is proposed and applied to ECT image reconstruction.Finally,the effectiveness of the proposed algorithm is verified by numerical simulation.