Better Defining the Uncertainties for the AGA-8 Equation
Eric W. Lemmon
In order to revise the AGA-8 documentation for publication in 2014, better knowledge of the 0.1% uncertainty level in the equation of state is needed. A 0.1% uncertainty level has been identified by an AGA-8 task group as the acceptable limit of error for density and compressibility factor calculations. With the availability of a new high-accuracy wide-range equation of state, it is now possible to locate regions within AGA-8 that are most likely outside the 0.1% uncertainty limit, which was not possible in 1992 due to a lack of experimental data. Code was written to locate the 0.1% error between the AGA-8 equation of state and the new equation. Compositions of each of the 21 components allowed in AGA-8 were varied, first for binary pairs of the component with methane, then as multi-component mixtures with compositions less than that determined for the binary pairs. The upper limit for each fluid was slowly decreased until a point was found that appeared to be the best compromise between the highest possible concentration for each fluid but with the lowest acceptable uncertainty. These limits are given in Table 3 of this document.