Hybrid Estimation of Inflow Multiphase Production Rates Using a Dynamic Wellbore Flow Model

This paper considers the problem of estimating the quantitative parameters of a two-phase fluid flow in a well based on the dynamic physical flow model. This is a challenging problem in the oil and gas industry, where the knowledge of multiphase production rates plays an important role during reservoir characterization, production optimization and reservoir management. As the direct measurement of these rates is not easily available, they can be inferred from conventional sensors (e.g., pressure gauges) in combination with a dynamic multiphase flow model. The methodology proposed in this work uses inverse modeling concepts to estimate flow rates that are not measured directly. The mismatch between the available data and model prediction is numerically minimized, leading to the optimal set of dynamic flow variables characterizing the flow. Two different scenarios are considered: firstly, when the well has only a flow meter located at the wellhead (minimum amount of available information), and when the well has distributed pressure sensors in addition to the topside flow meter (maximum amount of information). The feasibility of the proposed concept is assessed via several simulation-based case studies.