Experiments for the Measurement of LNG Mass Burning Rates

Abstract

Liquefied Natural Gas (LNG) is a commonly used flammable fuel that has safety concerns associated with vapor dispersion and radiation emitted from pool fires. The main objective of this effort is to advance the knowledge of pool fires and to expand the data that is commonly used to validate semi-empirical models. This includes evaluation of the methods that are utilized to obtain experimental values of mass burning rates, which are used in models where semi-empirical correlations cannot be applied.

A total of three small-size experiments designed to study the radiative characteristics of LNG pool fires were carried out at Texas A & M University's Brayton Fire Training Field (BFTF). This set of experiments was designed to study how the heat feedback from the fire to the pool surface is subsequently distributed through the liquid volume and the validity of different methods for measuring burning rates. In this work, a number of semi-empirical correlations were used to predict the characteristics of the flame and examine the predictive accuracy of these correlations when compared to the values obtained experimentally. In addition, the heat transferred from the energy received at the pool's surface to the surroundings was investigated. Finally, the parameters that influenced the measurement of radiative head feedback to the liquid pool were analyzed to investigate potential causes of calibration drift in the instrumentation.

The results of this work provided information regarding the validity of certain techniques for the measurement of mass burning rates and the use of correlations to predict the characteristics of an LNG pool fire on a small-scale. The findings from this work indicate that the energy received at the liquid surface was used entirely for evaporation and no indications of transmission to the surroundings were observed. Lastly, it was found that during the experiments, the sink temperature of the sensor was not constant, and therefore, the readings of the radiative heat were unreliable. This was due to the insufficient cooling effect of the water circulated. It was later shown in the laboratory that through a series of qualitative tests, a change of 20°C in the cooling water resulted in a calibration drift.