Nuclear Magnetic Resonance Two-phase Flow Measurement System And Two-phase Flow Response Characteristics Research

On-line monitoring of two-phase flow is an important guarantee for crude oil production and reservoir management.Previous studies have shown the complexity of two-phase flow components and the variability of flow processes,which has limitations for traditional measurement techniques.Nuclear magnetic resonance technology is a non-contact,accurate and efficient measurement method which can accurately measure fluid information.In this paper,a two-phase flow measurement system based on nuclear magnetic resonance technology is designed to measure the two-phase flow.In order to meet the requirements of the two-phase flow measurement system with FPGA and DSP as the control center,the designed system mainly completes the acquisition,storage and transmission of nuclear magnetic resonance signals.In addition,the feature analysis and selection were completed by using NMR echo signal and T2 spectrum,and the corresponding mathematical model was established to predict the parameters such as water content and flow rate of two-phase flow.The main work and innovations of this paper are as follows :(1)We completed the design of NMR two-phase flow hardware circuit The whole hardware circuit system is mainly based on FPGA and DSP as the control center,including AD acquisition module,FPGA and peripheral circuit module,DSP and peripheral circuit module and system power supply circuit module.The circuit design mainly completes the functions of analog signal acquisition,digital signal acquisition,and communication between FPGA and DSP.Through the test,the maximum measurement of the AD module is 2m V.Different signals are measured in the system joint debugging,and FPGA and DSP can complete the accurate measurement of the signal.(2)The software design of the measurement system controlled by FPGA and DSP is completed.In the FPGA project,the design of the special circuit is completed through the modular design,including the data acquisition and control module,the data storage module,and the data transmission module.In the DSP project,the design of the data receiving function and the signal denoising design are mainly completed.The signal-to-noise ratio of the denoised signal is increased by 34.6 %,and the design of the upper computer software is finally completed.Through the simulation test of each module,the functional requirements of the module have been realized.(3)We analyzed and selected the characteristics of echo signal and T2 spectrum,and then screened the characteristics of two-phase flow by Pearson correlation filtering method.By establishing BP neural network(BPNN)model and least squares support vector machine(LSSVM)model,the water content and velocity of oil-water two-phase flow were predicted and analyzed.The linearity of water content prediction results of LSSVM model and BPNN model is99.962 % and 99.898 % respectively,and the linearity of flow rate prediction results is 99.999 %and 99.976 % respectively.LSSVM shows good linearity in the prediction results of two-phase flow parameters.The two-phase flow measurement system designed in this paper can well measure the two-phase flow fluid and analyze the response characteristics of the two-phase flow.