Wednesday, October 09
His Excellency Shaikh Mohammed bin Khalifa Al Khalifa
Minister of Oil
Kingdom of Bahrain
Khalid A. Zainalabedin
Manager, Reservoir Description and Simulation Department
The development and management of naturally fractured reservoirs (NFRs) is a challenging task for asset teams due to the complexity of these reservoirs. The challenges are observed from the initial exploration phase and continue up to the field development stage. Placing production and injection wells in NFRs pose serious challenges. In most cases, highly permeable fractures are encountered during drilling leading to substantial loss of drilling fluids and extensive use of loss circulation materials. Thus, asset teams need to proactively predict location of these highly conductive fractures since they act as channels for rapid water or gas movement leading to early breakthrough and poor volumetric sweep efficiency.
Therefore, designing optimum well placement, well control, and well completion strategies are essential to reducing the short-circuiting effect of highly conductive fractures and allowing injected fluids contact more oil. Monitoring NFRs through surveillance plans become vital to understand fluid movements and make timely corrective actions. These remedial actions can be water shut-off, recompletion with in-flow control devices, or recompletion with smart completions. This session is a platform for sharing experiences and best practices in developing and managing NFRs.
The session covers the development life cycle including well placement, well completion, well control, reservoir monitoring and surveillance, and well intervention.
Integration remains a key theme in the characterisation of fracture-matrix systems. The reservoir envelope within which fractures develop is required to be defined by geological, geomechanical, and geophysical constraints, but the most appropriate data sets of each are typically acquired over differing length scales. The architecture of fracture sets has strong spatial heterogeneity and typically, those parameters that have the greatest impact on the required ‘effective’ fracture properties, such as fracture aperture and connectivity, are poorly defined. Fracture hierarchies may also exist, in which dynamic properties are dominated by just a small sub-set of all the realised fracture elements.
Data acquisition strategies focus on capturing individual fracture characterisation elements that are key to reducing uncertainty. Both consistent and novel interpretation techniques may be required. Careful monitoring of the temporal behaviour of the reservoir under production, of saturation and pressure change, in turn can be the only way to distinguish which of many possible fracture concepts (modelled or expected) is being realised.
This session will explore those practices proven successful in the evaluation and characterisation of naturally fractured carbonate reservoirs and offer some of the, often unexpected, lessons learnt.
Thorough characterisation and modelling of naturally fractured carbonate reservoirs is a key step in the effective management of such challenging reservoir systems. Indeed, it helps to limit the risk of drilling dry wells, to estimate the oil reserves, to prevent or mitigate early breakthrough of the injected fluids, and to understand reservoir compartmentalisation or complex production mechanisms in order to improve the recovery factor.
Building fractured reservoir models incorporates both well and inter-well data, such as advanced cores and well logs analysis, seismic attributes interpretations, geo-mechanical analysis, that together capture the spatial complexity of fracture-matrix properties. When differing geological concepts (fracture drivers) are possible, separate (multi-scenario) modelling workflows may be required to allow differences in reservoir behaviour to be distinguished. Modelled fracture networks may be discrete of continuous and are further calibrated with dynamic analyses but can also be further degraded through upscaling requirements.
The purpose of this session is to present best practices in data acquisition and interpretation associated to the most appropriate workflows intended to build reliable static models. The latter enable better understanding of the fracture network, translating in better management of real world naturally fractured carbonate reservoirs.
Thursday, October 10
Identifying the dominant flow media existing in naturally fractured reservoirs, their distribution within the reservoir complex, and impact on the multiphase fluid flow mechanism has been very challenging. Consequently, dynamic models, which are end-results of such characterisations, are associated with substantial uncertainties. Such models typically exhibit low confidence in their predictability despite the enormous resources and time spent in developing them.
However, with recent advances in dynamic data acquisition and computational technologies, there has been a growing expectation to improving the value proposition of simulation models. Naturally fractured reservoirs are not only comprised of matrix and fracture media but could also include other porosity systems such as connected vugs, as observed in karst reservoirs. All these porosity media systems should be accurately characterised, upscaled, and incorporated in dynamic models.
Furthermore, characterising the wettability of the reservoir rock is critical in maximising the sweep efficiency and recovery factors. Though it is generally accepted that most carbonate reservoirs are oil-wet, there have been instances of mixed wettability. Therefore, efforts should be exerted to understand the rock-fluid interaction and ensure that they are accurately represented in the dynamic models.
In this session, recent advances in dynamic reservoir characterisation and simulation will be presented. Also, application of 4D seismic data to enhance predictability of dynamic models will be highlighted.
In most NFCRs, the high level of connectivity of the fracture network soon requires the application of IOR and/or EOR methods to unlock production from the low permeability matrix blocks, which generally bear the most of the oil in such reservoirs. Even when based on sound static and dynamic characterisations, production optimisation remains challenging: classical flooding mainly produces the oil contained in the fractures during the early stages of the field development. The high connectivity of the different levels of the fracture network along with the low permeability of the oil-bearing matrix blocks, combined with the often oil-wet nature of carbonate formations, are the three main factors at play preventing the production of the remaining oil from the matrix blocks.
This session will aim to present and discuss the EOR methods applied on the field, including the following technologies: gas injection, thermal- and solvent-assisted production, and chemical EOR. The latter being seen as a way to improve the kinetic of oil recovery from the matrix.
Oil and gas is amongst the top three industries, which produces huge amounts of data. However, this data is severely underutilised with one study reporting a utilisation of less than 2%. With the ongoing digital transformation and IR 4.0 accelerating across the E&P value chain, promising opportunities emerged to harness the power of data and unlock its value. In the Middle East, 80% of oil and gas is in carbonate reservoirs. Carbonate reservoir characterisation is challenging due to geological heterogeneity. The presence of natural fractures interconnectivity in these reservoirs improves production but adds another dimension to this heterogeneity.
The session objective is to explore new digital methodologies, leveraging big data, machine learning, and artificial intelligence and analytics technologies, to decipher naturally fractured carbonate assets through the various development stages from characterisation to production. In addition, the session provides a platform to discuss hybrid solutions combining scientific methods and data-driven techniques to enhance exploration and production operations.