Report on Visit to North China Electric Power University

Following the approval of my application for the UKCCSRC International Research Collaboration Fund, I visited North China Electric Power University (NCEPU) over the period from 1 September 2016 to 30 April 2017. The following is a report on the activities that were undertaken during my visit. 

Experimental Tests

The main activity of my visit was to undertake an experimental programme on the CO2 two-phase flow test rig at NCEPU. The first part of the experimental programme was to observe the transient behaviors of CO2 under variable process conditions. A range of sensors and flow instruments were installed on the horizontal test section of the rig. These include a Coriolis mass flowmeter manufactured by KROHNE Ltd, a high-speed camera and a differential-pressure transducer. Due to the scale of the test work and manpower required for rapid changes in the flow conditions and associated data logging, several researchers from NCEPU supported my work throughout the experimental programme.  Figure 1 shows the experimental set-up. The characteristics of CO2 flow in the test section were observed under variable load, startup and shutdown conditions. When the load changed, a step change in the mass flow rate of liquid-phase or gaseous CO2 was observed. A piston pump in the liquid-phase loop was used to create the startup and shutdown conditions. As expected, transient CO2 two-phase flow regimes were observed. Images of the flow regimes and corresponding mass flow rate, density, pressure and temperature data of CO2 under a range of test conditions were recorded for further analysis.

In order to visualize the CO2 flow in the pipeline, a transparent acrylic spool piece was installed in the test section. The acrylic section was specially designed to withstand the high pressure in the two-phase flow rig. The high-speed camera was placed at the same height as the pipe axis along with a continuous white light source for illumination of the CO2 flow. Figure 2 shows a typical liquid CO2 slug flow during a start-up experiment.

The second part of the experimental programme was the measurement of the mass flow rate of two-phase CO2 and identification of CO2 flow regimes. The experimental set-up is shown in Figure 3. The maximum range of Coriolis flowmeters that can achieve the accuracy of ±1.5% (uncertainty set in the EU Emission Trading Scheme) and typical CO2 flow regimes under CCS conditions were investigated. The mass flow rate of liquid CO2 on the test section was from 150 kg/h to 3200 kg/h while the range of gaseous CO2 flow rate was 0-300 kg/h. The gas volume fraction of up to 75% was achieved during the tests. The maximum range of the developed multi-modal sensing system that can achieve the accuracy of ±1.5% is from 800 kg/h to 3200 kg/h. When the flow is between 150 kg/h and 800 kg/h, the error is within ±2% on the vertical section and ±5% on the horizontal section.

An electrical capacitance tomography (ECT) system was specially designed and constructed by Beihang and Tianjin Universities to visualize the cross sectional distribution of CO2 two-phase flow in the pipeline. The differences between the dielectric constants of the liquid and gaseous CO2 result in small changes in the measured capacitances between different electrodes in the ECT system. Figure 4 shows reconstructed stratified flow regimes for different gas volume fractions when the liquid mass flow rate was 500 kg/h. However, since the spatial resolution of the ECT system is very limited, small bubbles in the bubbly flow regime cannot be reconstructed. Attempts are being made to use the capacitance signals between different electrode pairs in the ECT to identify the flow regime.

Effect of Impurity Gases

The presence of impurity gasses in the CO2 stream could affect the flow characteristics and hence the performance of the flow metering system. As preliminary experimental observations, Nitrogen was diluted as a contaminant into the pure CO2 gas flow for various mass fractions to create an adulterated mixture of the CO2 gas. The results indicate that no significant changes in flow characteristics were observed and the Coriolis flowmeters still performed well in the presence of the impurity gas under single-phase and two-phase flow conditions. Unfortunately, we were unable to conduct the planned tests with other impurity gases as significant changes were to be made to the test rig in order to control the quantity and flow rate of impurity gases. This area of work will be undertaken in the near future.


Concluding Remarks

The phase behaviors of CO2 under CCS conditions are different from those of other chemicals in transportation pipelines. Difficult flow conditions impose significant challenges in the metering and characterisation of CO2 flows. We have combined Coriolis mass flowmeters, high-speed imaging equipment and an ECT system to measure and observe the CO2 flow under steady and transient conditions. In addition to the experimental programme, several discussion meetings were held between the researchers from the University of Kent, NCEPU, Beihang and Tianjin Universities on topics concerning the improvement of the CO2 two-phase flow rig, measurement of two-phase CO2 flow and identification of CO2 flow regimes. At the time of writing, the joint research team is still processing the data from the experimental programme. It is planned that such data will be published in the near future. 


Y. Yan