Mohamed Hassan

Oil PVT characterisation using ensemble systems

In reservoir engineering, there is always a need to estimate crude oil Pressure, Volume and Temperature (PVT) properties for many critical calculations and decisions such as reserve estimate, material balance design and oil recovery strategy, among others. Empirical correlation are often used instead of costly laboratory experiments to estimate these properties. However, these correlations do not always give sufficient accuracy. This paper develops ensemble support vector regression and ensemble regression tree models to predict two important crude oil PVT properties: bubblepoint pressure and oil formation volume factor at bubblepoint. The developed ensemble models are compared with standalone support vector machine (SVM) and regression tree models, and commonly used empirical correlations. The ensemble models give better accuracy when compared to correlations from the literature and more consistent results than the standalone SVM and regression tree models.

Shale Gas Reservoirs—A Comparative Approach

Until recently, shales were regarded by the oil industry as a setback hindering the operations while drilling to target sandstone and limestone reservoirs. Also, it was considered as the source rock for hydrocarbons migrating into conventional reservoirs and as the seal for trapping oil and gas in underlying sediments. Thanks to the “shale gas revolution” in the USA, however, it is known today that shales, as the most abundant form of sedimentary rock on Earth, can form low-permeable reservoirs containing huge quantities of hydrocarbons. Unlike the conventional reservoirs, production from shale gas resources would not be economically feasible due to the very low rates of flow of natural gas from the formation of ultra-low permeability. Yet, this is only one of the characteristics of shale gas reservoir which makes them different from conventional resources. Outlining the fundamental differences between shale gas plays and conventional gas reservoirs along with a brief description of status of shale gas development is the subject of this chapter. Furthermore, the potential of shale gas in the UK and the problems associated with its development compared to the US shale gas are presented.

Numerical Study of Shale Gas Flow Behaviour in Reservoir and Hydraulic Fractures

Some of the inappropriate assumptions that are often made in the use of commercial simulators for shale gas simulations are discussed in this chapter. For shale gas reservoirs characterised by very small pore size network, these approximations could lead to serious errors. Modelling of the geological complexities of shale gas requires the use of appropriate grid structures within the simulator to handle these complexities. Also, implementation of appropriate numerical methods that can adequately solve the set of mathematical equations associated with the simulation of shale gas reservoirs is the key to obtain sensible simulation results. This chapter provides a review of these inherent challenges in shale gas modelling. The concept of instantaneous capillary equilibrium within the pore networks as well as the non-Darcy flow that occurs within the matrix of the pore network is critically reviewed while the existing theories for proppant transport within the fractures are examined.

Production Performance Analysis of Shale Gas Reservoirs

Prediction of production performance in shale gas reservoirs can be achieved through a number of techniques designed to produce information about the estimated ultimate recovery and also about the reservoir parameters such as permeability, skin and fracture properties. These methods sometimes involve obtaining well production data and making future predictions from them. This chapter looks at some of the analytical and semi-analytical models for production performance calculations associated with shale gas reservoirs. The techniques discussed in this chapter include decline curve analysis, pressure transient analysis and rate transient analysis.

Inherent Defying Features in Shale Gas Modelling

Accurate simulation and modelling of shale gas reservoirs are deemed crucial for efficient exploitation of these resources. Obtaining realistic results for resource estimation and performance predictions has a significant impact on the economics of the operating companies and all interested parties. Integrating all the unique characteristics of shale gas reservoirs within a single reservoir simulator for accurate predictions of future performance has become an increasingly intricate task. For many years now, various researchers have tried to tackle some of these challenges which include, but not limited to, how the natural fractures are simplified and represented in a simulator, the transport of gas within the matrix and fractures, adsorption and desorption phenomena within the shale gas system and also how the fractures are propagated within the shale formation upon hydraulic fracturing. This chapter provides an overview of the advances made in shale gas modelling and highlights the improved understanding conveyed by various researchers on the main defining characteristics of shale and the way these features of shale are modelled in numerical reservoir simulators.

Ensemble SVM for characterisation of crude oil viscosity

This paper develops ensemble machine learning model for the prediction of dead oil, saturated and undersaturated viscosities. Easily acquired field data have been used as the input parameters for the machine learning process. Different functional forms for each property have been considered in the simulation. Prediction performance of the ensemble model is better than the compared commonly used correlations based on the error statistical analysis. This work also gives insight into the reliability and performance of different functional forms that have been used in the literature to formulate these viscosities. As the improved predictions of viscosity are always craved for, the developed ensemble support vector regression models could potentially replace the empirical correlation for viscosity prediction.

Hybrid Functional Networks for Oil Reservoir PVT Characterisation

Abstract Predicting pressure-volume-temperature (PVT) properties of black oil is one of the key processes required in a successful oil exploration. As crude oils from different regions have different properties, some researchers have used API gravity, which is used to classify crude oils, to develop different empirical correlations for different classes of black oils. However, this manual grouping may not necessarily result in correlations that appropriately capture the uncertainties in the black oils. This paper proposes intelligent clustering to group black oils before passing the clusters as inputs to the functional networks for prediction. This hybrid process gives better performance than the empirical correlations, standalone functional networks and neural network predictions.