A study on a real-time leak detection method for pressurized liquid refrigerant pipeline based on pressure and flow rate

Abstract

Leakage from pressurized liquid ammonia pipeline has been a serious problem in large commercial cold storages because it might release large amount of liquid ammonia and without safety supervision in daily operations. The present paper shows a detection method for a pressurized liquid ammonia pipeline with a leak. The variations of pressure, flow rate and pressure difference profile are studied. A leak indicator (σ), proposed with the one-dimensional steady-state flow model, is used to detect the leak occurrence by comparing it with a threshold value (σLe). A locating algorithm based on pressure difference profile along the pipeline is also proposed, which has considered the effect of the static pressure increase at the leak point. Experiments on different leak positions and ratios from liquid R22 and ammonia pipelines are carried out to validate this method. It is found that, with a relatively low false alarm rate (as three percent), the minimum detectable leak ratio reached 1% for the R22 pipeline and 4% for the ammonia pipeline. The locating errors are between −27% ~ 17% for R22 pipeline and −27% ~ 27% for ammonia pipeline.