Using Measurement Uncertainty in Production Allocation (UPM 2016)

This paper shows how production measurement uncertainties can significantly affect allocation results under proportional and by-difference methods. This paper evaluates these impacts and proposes an uncertainty-based allocation using data validation and reconciliation (DVR). DVR incorporates measurement and fluid-property uncertainties, balances the full production system from subsea to custody transfer under physical constraints, and estimates the most likely sales quantities. Applied for the first time to a multi-tier allocation system, the method quantifies both measured and unmeasured variables with their uncertainties and provides diagnostics that support condition-based measurement maintenance, leading to fairer and more defensible allocations.

Download paper

 

Authors

Amin Amin – Upstream Production Measurement (UPM) Forum – Houston TX, – 24-25 February 2016.

 

Abstract

The effect of production measurements uncertainties is evaluated on different oil and gas production allocation methods, proportional, by difference, and uncertainty based. The scope is further expanded by examining and proposing an allocation method where measured and allocated quantities uncertainties are considered to achieve fair and equitable production allocation results.

The proposed approach uses data validation and reconciliation (DVR) methodology (governed by establish industry standards) to account for the measurement and fluids properties uncertainties rarely considered in traditional production allocation schemes. The DVR approach has the advantage to balance a complete production system, from subsea to custody transfer point, by estimating the most likely flow quantities available to sales within the known accuracies of the measurement devices and by obeying the physical constraints/laws of the producing network.

The DVR methodology is used for the first time in a multi-tier allocation system. It considers and quantifies measured and unmeasured parameters including their uncertainties in an all-encompassing production allocation scheme. The methodology also allows for measurement error diagnostics that can be integrated in a condition-based measurement maintenance program.