Page 48: of Offshore Engineer Magazine (Nov/Dec 2014)

47 49

Production Increasing the accuracy of uncertainty data hile fow meters are cali- brated under ideal laboratory

W conditions, the environments into which they are installed vary greatly.

Uncertainty analyses are therefore essen- tial to determine whether measurement systems, once installed, are capable of meeting required performance targets.

However, when developing uncertainty budgets for new measurement systems, it is often diffcult to obtain reliable data to provide such evidence. Two main meth- ods are used, the analytical method and the Monte Carlo method.

The analytical method

Calculating uncertainty using the analytical method is described in detail in the Guide to the expression of Uncertainty of measurement (GUM), produced by Working Group One of

Photo from NEL.

the Joint Committee for Guides in

Metrology.

Sharing information

The technique involves a series of about uncertainty steps: 1. Defne the relationship between all the budgets for fow inputs to the measurement and the fnal meters could lead to a result.

2. For each input, draw up a list of all the better understanding factors that contribute to the uncertainty 7. Express the overall uncertainty of measurement in that input. as the interval about the measured systems, says NEL’s 3. For each of the uncertainty sources value, within which the true value is make an estimate of the magnitude of the expected to lie with the required level of

Callum Hardie. uncertainty. confdence.

4. Convert the uncertainties to standard The uncertainty budgets created using uncertainties by assigning a probability the analytical method are very use- distribution to each uncertainty source. ful tools for optimizing measurement 5. From the relationship defned in step systems, as the effect of changes in input 1, estimate the effect that each input has uncertainties on the output uncertainty on the measured result. This is usu- can be seen very quickly. The input ally achieved by calculating sensitivity uncertainty sources can also be ranked coeffcients. to determine which sources have the 6. Combine all the input uncertainties most signifcant effect on the overall using the root sum squared technique to uncertainty. The process of develop- obtain the overall uncertainty in the fnal ing uncertainty budgets can also be result. benefcial in that it helps to gain a full

November 2014 | OE oedigital.com 50 050_OE1014_ProdOps1_NEL.indd 50 10/21/14 5:28 PM