Proposal for AN INDUSTRY FUNDED PhD studentship at
Heriot-Watt Petroleum Engineering Institute
Modelling of fluid flow and seismic for improved reservoir description in stress-sensitive fractured reservoirs
Fractured reservoir description is widely accepted as being a major challenge for the oil industry. Due to strong heterogeneity in such reservoirs it is difficult to model and replicate their flow behaviour with any degree of precision, and hence forward predictions remain highly uncertain. Flow simulation is problematic because of the high permeability contrasts between the fracture pore space and the matrix, together with the lack of deterministic detail on the spatial and petrophysical characteristics of the fracture system. Some of the physics can be addressed to a certain extent by using dual porosity or dual permeability methods. However, such methods are not always satisfactory, as they do not explicitly consider the complex connectivities and intersections, the nature of the fracture faces and gouge material, and the overall response to reservoir stress changes.
Recent research at Heriot-Watt University in several areas has helped to define a framework for tackling a higher level of complexity than is adopted in current practice. Heriot-Watt has, for example, shown how the permeability of fractures depends critically on the balance between fluid pressure and stress, and how idealised vertical fracture permeability may be related to seismic and conditioned from seismic anisotropy attributes. The aim of the current proposal is to extend this avenue of research by investigating the relationship between azimuthal AVO and time-lapsed seismic data and fracture network properties in the context of stress sensitivity. This work is important as it represents a collaboration between geophysicists, reservoir engineers and structural geologists in the Institute.
We propose a project that will address the following areas:
(a) Creation of a workflow that links software from each domain to permit the integrated modelling of fluid flow, stress sensitivity and the seismic response for a multi-layered fractured reservoir.
(b) Modelling a number of geologically plausible elemental fracture-stress scenarios, to build up an understanding of the relationships between the fracture properties, fluid flow, and production behaviour. Here, models are guided by fracture patterns derived from outcrop studies. The type of rock, and fractures at specific length scales will be considered. The properties of fractures at each scale are to be effectively upscaled to the reservoir and seismic scale.
(c) We aim to evolve fracture models towards greater and greater realism: they will become more connected, have rougher walls, and intersecting geometries. Rock and fluid physics developments are required to link seismic and flow simulation by a common model.
(d) Of particular interest is the change in seismic response and fluid flow due to reservoir depletion and injection. Here, the full benefits of using 4D seismic versus well data are explored by modelling the pre- and post-stack time-lapsed seismic response.
(e) Examination of the influence of different loads and fluid pressure changes on the two-phase fluid flow through a realistic reservoir model. This will involve coupling between the effective stress tensor, permeability tensor and elastic tensor.
(f) The modelled seismic properties help towards forming an understanding as to what information can reasonably be expected to contribute to the reservoir description. The uncertainty involved in this process will be assessed, and the additional information required to quantify the seismic response identified.
An extension of this project is the application of the technology to a fractured carbonate reservoir in which a complete integrated dataset is available.
Tensorial permeability - at Heriot-Watt University, we have gained experience in calculating effective permeability tensors in models of sedimentary structures and fracture networks, and in using the full permeability tensor when simulating flow at a larger scale. This project would extend our work on tensor permeabilities to investigate the coupling between the effective stress tensor and the effective permeability tensor.
Stress sensitive flow simulation - will be carried out using software that has been developed recently at Heriot-Watt University (in conjunction with the Department of Civil Engineering at Glasgow University). In this software, the matrix is modelled using polygonal blocks, which are bounded by fractures. The blocks can respond to stress by moving relative to each other, thus opening or closing the fractures. The model is triangulated using a fine mesh (with cells smaller than the fractures), and flow simulation is performed using a finite element method.
Seismic modelling software – simulator to seismic code developed in-house at Heriot-Watt will be used to determine the 4D and anisotropic seismic response. Our experience at interpreting real field data will be of benefit in ensuring a practical interpretation of the results.
This project will require a Ph.D. student to work for three years. The student should have a background in numerical modelling, and an interest in earth sciences. The project will be jointly supervised by Gary Couples (structural geologist), Colin MacBeth (reservoir geophysicist) and Gillian Pickup (reservoir engineer). We are seeking a total funding of £90K spread over three years, from one or several industry sponsors.