Quantifying near-wellbore permeability heterogeneity using wellbore flow modeling and fiber optics distributed pressure sensor

Abstract

In the oil and gas industry, there is a need to accurately obtain downhole pressure to determine the distribution of flow rates at perforations in the wellbore. Furthermore, these flow rates can be used to quantify near-wellbore reservoir heterogeneity such as a major fracture conduit. Fiber-optic sensing has gained more popularity due to many advantages over traditional electrical sensors, such as long-term durability and distributed sensing in harsh conditions. One of the most valuable parameters that fiber-optic sensing could provide is a near-continuous recording of fluid pressure along the wellbore via distributed pressure sensing (DPS). In addition, fiber optics are used in distributed acoustic sensing (DAS), distributed temperature sensing (DTS), and distributed strain sensing (DSS). For instance, DAS data can be used to detect seismic and microseismic events; DTS can be used to monitor steam injection applications; DSS can be used to detect pipe leakages. However, we focus on DPS in this thesis and propose a modeling method to quantify major permeability spikes from such measurements. In theory, measured pressure data along the wellbore from fiber-optic sensing can be used to calculate specific productivity index variations, which, in turn, quantifies permeability spikes along the wellbore, such as high-permeability fractures and leaking faults. Because of the lack of field data, a proxy numerical model was developed to generate flowing pressure data in a horizontal wellbore. The flow in a reservoir is governed by the conservation of mass and momentum and Darcy's law, whereas the flow in a wellbore is governed by the conservation of mass and energy. We then used a semi-analytical model to determine the permeability variations along the wellbore that yield the closest wellbore pressure profile to the simulated measured pressure. From a numerical model, a build-up test was conducted to determine the apparent permeability and the average reservoir pressure. The apparent permeability provided an initial estimate of the permeability for a semi-analytical model. Then, the `interior-point barrier method' algorithm was used to minimize the squared difference between the fluid pressures from the two models and obtain permeability along the wellbore. Several numerical examples, such as a single-phase flow and a water-oil flow in a single-porosity reservoir, were examined. Knowing the productivity of the reservoir along the wellbore allows operators to make quality decisions, leading to optimal reservoir development. In addition, one can generate a flow rate curve along the wellbore as a proxy to a production logging measurement. Fiber-optic pressure sensing provides a continuous stream of valuable information, leading to reservoir description on a real-time basis.