Uncertainty Estimation for the Liquid Hydrocarbons Measurement in Static and Dynamic Measurement Systems: A Colombian Case Study

Abstract

Summary To evaluate the level of accuracy of more than 500 measurement systems of liquid hydrocarbons in Colombia, a computer tool is currently used to calculate the uncertainty of measurement, applying the Guide to the Expression of Uncertainty in Measurement (GUM) method described in JCGM100 (2008). Taking into account that the selected method presents limitations by assuming that the output variable always has a Gaussian probability distribution and that the model that represents the measurand should preferably be linear, there is a risk that some of these conditions will not be met for the current application. Therefore, this present work describes a comparison between the uncertainty assessment performed according to the GUM uncertainty framework and the Monte Carlo method (MCM) to confirm that the GUM method is valid for the uncertainty measurement evaluation of the measurements under study. This comparison was made from actual data sets associated with two selected examples: liquid hydrocarbon measurement with a vertical fixed-top tank and dynamic measurement with a positive displacement meter. As a result, the maximum difference between the GUM and MCM uncertainties obtained during the tests was 0.003% relative to the volume measured in the vertical tank example. In the case of the dynamic measurement with a positive displacement meter, it was 0.0004% relative to the volume. These differences were less than the numerical tolerance established according to the Monte Carlo adaptive procedure; therefore, it was possible to conclude that the application of GUM is adequate for the examples evaluated and others with similar operating conditions.