| Two phase flow widely exists in power,environmental protection,chemical industry,metallurgy,grain processing and other production fields,and plays an important role in the national economy.For example,in the process of grain transportation in closed pipelines,cement production and transportation,and chemical plant sewage discharge,it is necessary to monitor the two-phase flow in pipelines,so as to avoid pipeline blockage in the transmission process,So as to cause the occurrence of adverse events.Therefore,in the process of production and transportation,the detection and monitoring of two-phase flow has important practical significance for safety production.Electrical capacitance tomography(ECT)technology is to place the copper capacitance plate on the outer wall of the closed pipe,take the obtained capacitance value as the projection data,and reverse the medium distribution inside the pipe through an appropriate algorithm,so as to realize the visualization of the internal information of the closed pipe.This technology has the characteristics of low cost,good real-time performance,non-invasive,no radiation and so on.It is a kind of process tomography,and two-phase flow parameter detection and monitoring is a typical application direction of this technology.At present,cross-correlation method is commonly used to detect the time delay of the upstream and downstream image signals collected by ECT system,and then obtain the average flow velocity.However,the time window selection ability of the cross-correlation function itself is limited,and the noise in the detection environment and the error of the detection instrument will reduce the time detection accuracy.In addition,the traditional fast reconstruction algorithm has fast imaging speed,but the reconstructed image effect is poor,which is not suitable for velocity detection.Therefore,this paper mainly focuses on the measurement of solid medium flow velocity and reconstruction image optimization processing in the process of gas / solid two-phase flow transmission.The main work and achievements of this paper are as follows:Firstly,the structure and parameters of ECT sensor are studied,and four typical ECT sensors are simulated.The reconstruction images and errors of center flow,circulation and laminar flow are given.Through the simulation results,the ECT sensor structure and corresponding parameters suitable for the solid medium flow rate measurement in this paper are finally determined.Secondly,this paper uses learning based super-resolution image reconstruction algorithm to visualize the reconstructed image.The learning based super-resolution image reconstruction algorithm is used to reverse the LBP reconstructed image,that is,the fuzzy processing method of inputting high-definition image and outputting fuzzy image is introduced;By adding an image segmentation module in front of the input layer of the super-resolution image reconstruction algorithm,the high-definition image is scaled and divided as the input image,and then the super-resolution image is processed.After processing,the reconstructed image is closer to the original medium distribution.Thirdly,the improved bi-directional Q algorithm is used to optimize the time window selection in cross-correlation method,that is,the time window of cross-correlation function calculation is calibrated,so as to eliminate the useless signals outside the time window,so as to improve the calculation accuracy of cross-correlation function.Through the dynamic experiment of gas-solid two-phase flow,the velocity and error obtained by the cross-correlation algorithm optimized by the improved bi-directional Q algorithm and the traditional cross-correlation algorithm are compared and analyzed.The experimental results show that the proposed optimized cross-correlation algorithm is superior to the traditional cross-correlation algorithm. |
PDF Full Text Request