Research On AECVT 3D Imaging Technology Based On Differential Evolution Algorithm

Adaptive Electrical Capacitance Volume Tomography(AECVT)is a new,non-intrusive Electrical Capacitance Tomography(ECT)3D imaging technology.AECVT is based on a composite plate composed of small plates along the 2D boundary of the imaging area to measure the capacitance data and restore the distribution of the insulating material in the 3D domain.Because of its nondestructive properties and the advantages of variable capacitor plates,it has gradually become one of the research hotspots in the field of industrial inspection and process imaging.This thesis focuses on the in-depth study of the sensitive field distribution of AECVT,especially the inverse problem imaging algorithm.Based on previous research results,the following tasks have been completed: 1.The development status and basic principles of 3D ECT imaging technology at home and abroad are reviewed,and the advantages of AECVT sensors are highlighted.2.This thesis analyzes the sensitive field distribution of AECVT under different voltage packages.The mathematical model of AECVT sensitive field was deduced.Using the fast calculation method of sensitivity,combined with the design advantages of AECVT small plates,the influence of different package excitations on the distribution of AECVT sensitive field was analyzed by changing the amplitude,quantity and distribution of voltage packages.The simulation shows that the sensitive field distribution of AECVT has nothing to do with the voltage amplitude under a single voltage excitation;the multiple voltage excitations can improve the sensitive field distribution and lay the foundation for the subsequent inverse problem research.3.The method of preprocessing using the AECVT three-dimensional sensitivity of Laplace transform method is discussed.Simulation results show that under the same simulation conditions,the Laplace sensitivity preprocessing method can reduce the relative modeling error from 0.9474 before processing to 0.8682 after processing,which improves the distribution of sensitive fields.4.Applying Differential Evolution(DE)to AECVT image reconstruction and verifying its feasibility in simple and complex media distribution models.Based on this,in view of the limitation of the imaging accuracy of the basic DE algorithm for the complex medium distribution model,a gray-scale preprocessing adaptive selection mutation strategy differential evolution algorithm(PADE)is proposed.The effect of voltage encapsulation on the performance of AECVT image reconstruction based on PADE is further studied.The simulation data show that the PADE algorithm can break through the local optimal solution during the evolution process,the image reconstruction error is small,and it can ensure a fast convergence speed,and it has good robustness to different media distribution models.