Electrostatic Tomography System Based On Spatiotemporal Information

Gas-solid two-phase flow is widely existed in industrial production processes,such as pneumatic conveying of particles and emission process of smoke.Due to the complexity of gas-solid two-phase flow and the non-uniformity of flow distribution parameters,the measurement of the gas-solid two-phase flow distribution parameters has been an internationally recognized problem.The electrostatic tomography(EST)technology is suitable for the measurement of charge distribution and velocity distribution.It has the advantages of non-invasiveness,visibility and low cost,which provides an effective method for measuring the distribution parameters of gas-solid two-phase flow.Since the electrostatic method is a passive measurement method,the independent measurement information and the ability to characterize the depth field of the object field are insufficient,compared with the excitation electrical tomography methods.Aiming at this problem,this paper proposes a direct velocity distribution reconstruction method based on the cross-correlation spatiotemporal model and builds a spatiotemporal EST system.The main works are summarized as follows:1.The structural parameters of electrostatic sensor array(ESA)directly affect the time-frequency domain characteristics of the measurement signal and the number of independent measurement information.In the paper,three optimization criterions including sensitivity field strength and uniformity,singular value analysis and reconstructed image correlation coefficient are selected.Based on COMSOL finite element simulation platform,the electrode axial length and electrode number of the inner flush-mounted ESA are optimized.2.As a velocity distribution reconstruction method commonly used in electrical tomography,gray correlation method has limitations in EST system.Aiming at this problem,the cross-correlation spatiotemporal model is proposed.Then,a direct velocity distribution reconstruction method is presented and effectively reduces the underdetermination and ill-posedness of the system.Furthermore,the positive problem of the spatiotemporal model is validated on the belt electrostatic induction experimental device.3.The EST system was built and the image reconstruction algorithm was selected considering the characteristics of the cross-correlation sensitivity matrix.Based on the belt electrostatic induction experimental device and the vertical pipe particle gravity experimental device,the direct velocity distribution reconstruction method and the gray correlation velocity distribution reconstruction method under different particle volume distributions and solid phase volume content were compared and analyzed.4.Since the measurement signal of the inner flush-mounted ESA contains both the induced charge signal and the transfer charge signal,which represent the flow field information from different angles,the paper analyzes the generation mechanism and signal characteristics of the induced charge signal and the transfer charge signal.Then,a direct velocity distribution reconstruction method based on the discrete wavelet transform(DWT)filtering is proposed,and different velocity measurement methods,including ring-shaped electrode correlation velocity measurement method and the arc-shaped ipsilateral electrode correlation velocity measurement method were compared on the pneumatic conveying experimental platform.