A Novel Infrared Absorption Technique for Measuring Flare-Gas Flow

摘要

1. The modified IR4MKII meter was tested on air over the range 0.4 - 67 m/s with an atomised water jet located at 14D and 24D upstream of its inlet proving a traceable feature. This extended the limit of previous applications of 10:1, to about 165:1. 2. The meter error varied from about 9% to 18% across the velocity range tested. A moderate level of repeatability was achieved over the velocity range. The error is positive because the IR4MKII meter takes its signal from the dominant flow region close to the pipe axis. These biases can be removed by further calibration and type testing over a range of line sizes and velocities 3. The behaviour of the meter was similar with the injector at 14D and 24D upstream of the meter, although there did appear to be a fairly weak influence from m_f on meter reading. 4. The IR4MKII meter was seen to work with variations in flow of only 0.001% by mass, which was much lower than the calculated detection limit. Additional testing showed that the reason for this was probably due to the reflection and refraction of the infrared by the water droplets, whereas it was anticipated prior to testing that absorption would be the dominant measurable feature. 5. It was calculated that only a 0.1% variation in flare gas density would be required to get a similar level of signal from the meter as seen in the current tests. 6. It appears that the IR4MKII meter can operate at velocities as low as 0.4 m/s, provided there is enough variation in the signal due to turbulence etc. It is anticipated that the meter could not operate at much below 0.4 m/s in an 8-inch line as the flow would be tending to be laminar in nature. The low velocity cut-off point would increase with line size for a given gas mixture, flow rate and line conditions. 7. Even if it was found that there is not enough variation in a flare-gas line at normal, quiescent velocities (e.g. 0.1 - 0.5 m/s), the IR4MKII meter could be still be used to measure the moderate-to-high velocity range that occurs during depressurisation processes where other technologies are known to struggle.