Oil-water two-phase flow exists widely in the process of oil exploitation and transportation,and its flow state is complex and changeable.Electrical tomography has attracted extensive attention,especially in the field of two-phase flow parameter detection,because of its advantages of non-invasive,low cost,simple structure,high reliability and no radiation.The traditional electrical tomography acquisition system can only establish the relationship between a single medium parameter(conductivity or permittivity)and the measured signal when detecting multiphase flow.For example,the mixed fluid in the sensitive field is equivalent to resistance in resistance tomography,and the mixed fluid in the sensitive field is equivalent to capacitance in capacitance tomography.However,due to the characteristics of oil reservoir distribution,oil-water two-phase flow has the characteristics of high salinity,and the nonlinear problem of traditional electrical chromatography will be aggravated.On the basis of the above electrical tomography studies,this paper studies a non-contact electrical impedance tomography system.By extracting the real,imaginary and phase components of complex impedance information in the measurement area and combining with image fusion technology,the system fuses the reconstructed images of the three components at pixel level,which improves the nonlinearity of electrical tomography and improves the accuracy of image reconstruction.A cross-section image reflecting the distribution of media is obtained.The main research work and stage achievements of this paper are as follows: The mathematical physical model and equivalent circuit model of the non-contact electrical impedance tomography system were studied and analyzed.The simulation model of the array sensor was constructed by COMSOL finite element analysis software.Under the conditions of different flow patterns,conductivity and excitation frequency,the imaging effects of the real part of the impedance,the virtual part of the impedance and the phase component were compared comprehensively,and the image reconstruction algorithm and excitation frequency were optimized.In order to further improve the imaging accuracy and generalization ability of the system,an electrical impedance tomography method based on the modified logic filter fusion algorithm with variable threshold is proposed.The method integrates the real part,the imaginary part and the phase information to realize the reconstruction of the medium distribution of oilwater two-phase flow.A non-contact impedance tomography system is designed,which consists of array electrode,analog array switch,impedance measurement circuit and host computer software.Among them,the upper computer software uses Lab VIEW to realize the functions of image reconstruction algorithm,image fusion algorithm and image display.Finally,a noncontact impedance tomography system test platform for oil-water two-phase flow was built,and the above method was verified by experiments under different media flow patterns and water conductivity.The experimental results show that the fusion algorithm proposed in this paper can effectively improve the image reconstruction accuracy of oil-water two-phase flow compared with the traditional imaging effect of a single medium. |