Combining Testing-by-Difference, Geochemical Fingerprinting, and Data-Driven Models: An Integrated Solution to Production Allocation in a Long Subsea Tieback

Abstract

Abstract Well-by-well production allocation in a subsea production cluster is often very challenging without a separate test line. This is recognized in field development plans and can be mitigated by installing subsea and downhole measurement devices, such as downhole gauges and/or flowmeters. Long-term reliability of these devices, however, remains an issue. In the subsea cluster described by this paper, most of the measurement equipment indeed failed or became erroneous within limited timeframe after installation in 2002. As reliable information became scarcer, the uncertainties in the production allocation increased with a direct negative impact on field management. The solution that was implemented in this Shell UK Limited operated North Sea cluster consists of an integrated application of periodic testing-by-difference and data-driven modelling. Data-driven models have the potential to act as continuous virtual flow meters, relating pressure and temperature changes per well to variations in the well production rates. Data-driven models for real-time monitoring of well-by-well oil and gas production have been widely deployed in the Shell Group in the form of the FieldWare Production Universe (FWPU) application. For these data driven models to be sufficiently accurate, they need to be calibrated with actual production information. In a subsea environment without testing facilities, this information can be derived from testing the wells by-difference and also from short and medium term production data. This paper shows that fairly satisfactory production rate estimates can be obtained with FWPU, even when the subsea tieback to the production and export facilities is more than 50 km long. Accuracy of the production rates derived from testing-by-difference can be impacted by interference between the wells, especially with a long tieback. Carefully designing and implementing the test and quality checking the derived rates with well flow models can ensure the usefulness of the data. In this paper we will also show that geochemical fingerprinting of fluid samples taken during the test provides valuable information about the quality of the estimates and the behaviour of the fluids in the flowline. The aim of this paper is not to justify the omission of physical measurement devices or well testing facilities in challenging subsea environments, but to promote the described techniques as valuable additions and possible contingencies.