Study On The Forward And Inverse Problem Of Two Phase Flow Measurement Based On The ERT Technology

The two-phase flow phenomenon exists widely in the industrial production process.As a complex fluid flow phenomenon,it may induce safety problems and even affect the stable and reliable operation of the whole system or equipment,such as the cryogenic sealing and bearing system in the high-speed turbopump of liquid rocket engine.The acquisition of its physical properties is the core of the attention of the industry and the science and technology circles.In view of the complicated mechanism of its generation,the measurement of relevant properties by experimental method is the most common method.Based on the comprehensive analysis of the current two-phase flow property measurement,this paper proposes and carries out the measurement of the two-phase flow pattern based on the Electrical Resistance Tomography(ERT)technology,which can further obtain other twophase flow parameters,such as phase holdup,flow rate,flow rate,etc.The research process constructs the positive problem theory model to realize the simulation and prediction of twophase flow pattern;Developed the flow pattern recognition technology based on neural network model for inverse problem,and carried out experiments to verify the model.The specific work is as follows: 1.The forward problem simulation model of two-phase flow based on ERT technology is proposed,and the analytical method and numerical method are used to solve the model,and the boundary potential of different flow patterns is obtained.The positive problem is the conductivity of the medium in the known sensitive field.The simulation model is built based on ERT,and the boundary potential is obtained by solving the simplified analytical solution and numerical solution.Through the comparison with the results of the existing literature,the average relative error is within 5%,which verifies the correctness of the model.At the same time,the effects of current intensity,electrode number and electrode width on the boundary potential are simulated.The results show that the relationship between current intensity and boundary potential is linear and proportional.The number of electrodes should not be too many or too few,and the electrode width should not be too wide.It provides theoretical support for subsequent physical design.2.A two-phase flow pattern recognition method combining Radial Basis Function(RBF)and ERT technology is developed,which broadens the inverse problem solving idea of identifying two-phase flow pattern by boundary potential.By knowing the potential of the sensitive field boundary,the conductivity distribution of the medium in the sensitive field can be obtained,and then the visualization of the flow pattern of the two-phase flow can be realized.COMSOL simulation is used to generate core flow and circulation boundary potential data samples,and RBF neural network is used for flow pattern image reconstruction.The reconstructed visual image is formed by triangle subdivision unit and gray value filling method to display the visual recognition results.The results show that the fidelity of core flow and circulation flow pattern recognition is close to 90%,and the reconstructed flow pattern image quality is better.3.A two-phase flow measurement experiment platform based on ERT technology is designed and built,relevant experiments are carried out and the experimental results are analyzed.The designed two-phase flow experiment platform includes sensor module,data acquisition module and image reconstruction module.A group of experimental test results show that the system can achieve the established measurement objectives and the results are reasonable.The results of theoretical and experimental studies have important guiding value for the measurement of flow pattern attributes of two-phase flow and can also provide important reference value for the acquisition of other two-phase flow parameters.