Mass Flowrate Measurement And Flow Condition Monitoring Of CO₂ Under CCS Conditions

It is a major challenge for the world to deal with global warming and reduce greenhouse gas emissions.Carbon capture and storage(CCS),as a key technology to reduce CO2 emissions from industries,makes fossil fuels into clean energy with zero carbon emissions.CO2 is expected to be captured and transported to a specific geological storage site in CCS processes.CO2 transportation is the middle part linking the capture part and storage part.Pipeline networks are considered to be the most economical and safest way for CCS transportation with the advantages of large transportation volume,low cost and high safety.Accurate mass flowrate metering of CO2 in transportation pipelines provides reliable measurement data to enable effective carbon trading,and is also essential for monitoring the operation and evaluating the economy of CCS projects.However,it is challenging to achieve the accurate mass flowrate measurement due to complex and variability transport conditions of CO2 flow in CCS pipelines.In this study,methods for mass flowrate measurement of CO2 in transportation pipelines under CCS conditions are proposed.The novelties and research contents in this study include the following aspects:(1)A method for mass flowrate measurement of gas-liquid two-phase CO2 flow under CCS conditions is proposed based on a Coriolis mass flowmeter,a capacitive sensor and other conventional instruments.The reasons for the varying characteristics of measurement errors of CO2 mass flowrate under CCS conditions are identified through the investigation of the sensing mechanism of Coriolis mass flowmeters.A mechanism-data-driven model is employed to correct the measurement errors of Coriolis mass flowmeters.(2)A gas-liquid two-phase CO2 flow rig is designed and constructed,including the design and implementation of the control system and the reference system and establishment of relevant health and safety regulations.The range of the operation temperature of the rig is 15～30℃,while the range of the operation pressure is 5.0～7.2 MPa.The inner pipe diameter of test sections is 25 mm.The test rig is capable of providing standards for the mass flowrate measurement of gas CO2 flow,liquid CO2 flow,and gas-liquid two-phase CO2 flow.Moreover,impurity gases such as?????can also be injected into gas-liquid two-phase CO2 flow.Experimental investigations were conducted on the gas-liquid two-phase CO2 flow rig under conditions of pure CO2 flow and CO2 flow with impurities.(3)A high-pressure capacitive sensor with four electrodes is designed and used to achieve the flow condition monitoring and gas void fraction measurement of gas-liquid two-phase CO2 flow.The relationship between capacitance signals and gas void fraction under different flow patterns is investigated to establish the flow pattern identification model based on the unsupervised fuzzy C-means clustering.Then,the gas void fraction measurement models under different flow patterns are established.The experimental results show that the proposed flow pattern identification model can achieve the monitoring of CO2 flow state under CCS conditions.The relative error of gas void fraction measurement is within ±5.9%while the absolute error is within ±2.0%with the gas void fraction ranging from 5%to 80%.(4)In this study,the method of mass flowrate measurememnt under steady-state gas-liquid two-phase CO2 flow conditions is proposed.Mechanism-data-driven models incorporating the deep neural network(DNN)and least squares support vector machine(LSSVM)are established to measure the mass flowrate of pure CO2 two-phase flow.Experimental results demonstrate that the mechanism-data-driven model based on DNN yields most relative errors within±1.5%and outperforms the LSSVM model for the mass flowrate ranging from 300 kg/h to 3000 kg/h and the gas void fraction from 0%to 80%.Without changing the structure and parameters,two mechanism-data-driven models are applied to measure the mass flowrate of CO2 two-phase flow with impurities,including N2,O2,Ar,to trial the generalization performance.Experimental results demonstrate that the mechanism-data-driven model based on LSSVM shows better generalization performance with a relative error within ±2%.(5)A mechanism-data-driven model is established to measure the mass flowrate of gas-liquid two-phase CO2 flow under dynamic flow conditions.Experimental observations of the CO2 transient behaviours were conducted on the gas-liquid two-phase CO2 flow rig.Due to the rapid change of mass flowrate and gas void fraction,the vibration damping of the measuring tube becomes larger.It is difficult for the Coriolis mass flowmeter to follow the rapid changes.The varying characteristics of measurement errors of Coriolis mass flowmeter are analyzed under dynamic flow conditions.A mechanism-data-driven model based on long short-term memory(LSTM)algorithm is established to track the rapid change of mass flowrate.Results show that the proposed model significantly improves the measurement accuracy of Coriolis mass flowmeter under dynamic flow conditions,which yields relative errors in horizontal and vertical pipelines within ±4.6%and ±3.7%,respectively.