CO2 obtained by capture process (such as post combustion, pre combustion and oxy-fuel combustion) is not 100% pure and may contain impurities such as H2, Ar, CO, H2S and water. The presence of such impurities in CO2 stream can lead to challenging flow assurance and processing issues. The gaseous CO2-rich stream is generally compressed to be transported as liquid in order to avoid two-phase flow and increase the density of the system. One aim of this work is to evaluate the effect of impurities on the physical properties of CO2 such as density, viscosity, speed of sound and on the phase behaviour of such systems. Speed of sound and isothermal compressibility of CO2/impurities mixtures were measured at condition above the saturation curve and temperature from 268.15 to 301.15 K. A new volume correction was implemented to the Peng-Robinson equation of state in order to minimise the error associated with the isothermal compressibility prediction. Moreover, density and viscosity are two of the most important properties in transport properties. Therefore, the effect of impurities on density and viscosity were experimentally and theoretically investigated in liquid CO2 and liquid CO2/impurities systems. The viscosity measurements were performed using in-house capillary tube apparatus in the range from 280 to 343.15 K and pressure up to 40 MPa. Two viscosity models, LBC and Pedersen, were modified in order to predict the viscosity of both pure and impure CO2. The density measurements were carried-out using an Anton Paar densitometer in both liquid and supercritical regions from 283.15 to 423.15 K. In order to improve the accuracy of EOSs in density of CO/impurities systems, a new modification was developed based on mixing the volume obtained from EOSs (SRK, PR and VPT) and the volume obtained from CO2-MBWR. The presence of water may result in ice and/or gas hydrate formation and cause blockage of pipelines. Several measurements were also conducted to evaluate the hydrate stability zone of pure and rich CO2 systems in free water. A thermodynamic model based on the VPT EOS was adopted to predict the hydrate phase of the systems. In addition, few saturation measurements of synthetic alkane mixture plus pure or impure CO2 were performed at 344.3 K in order to investigate the effect of impurities on the saturation pressure of CO2/alkane system. IFT and swelling factor properties on CO2/n-decane mixture were investigated at 310.95 K from ambient to near the minimum miscibility pressure of the mixture. The experiments were extended to cover the presence of impurities on the properties at the same range of pressure. Minimum miscibility pressure of the systems was estimated by both Vanishing Interfacial Tension method and multiple-mixing-cell calculation.

Oil companies propose polymer flooding techniques, but oftentimes find it difficult to convince asset teams to implement these. This is because it is much easier to estimate the return on investment from an infill well drilling programme, and the return is much quicker. On the other hand, there may be a delay of years before increased oil recovery is observed following implementation of polymer flooding process, and indeed, it may be difficult to ascertain just how much incremental oil has been recovered. The work developed in this thesis involved setting up a range of polymer flooding scenarios, performing analysis using both very detailed reservoir simulation calculations with a range of sensitivities, and also economic calculations, again testing a range of parameters, to ensure that a full range of possible outcomes is evaluated, and then making a comparison with infill drilling to maximise the value of mature assets. The method was first applied to a synthetic scenario with constant economic parameters, and was then applied and tested with varied operational and economic parameters. These sensitivity calculations have been performed by developing a computer program, coded in Java. Monte Carlo Simulation (MCS) is then performed to generate statistics from this method, and test economic uncertainties and the risks associated with implementation of polymer flooding. The method was then applied to a real field system where the choice of infill well drilling had previously been made by the operating company, to test the robustness of the analysis using polymer flooding against a conventional decision making process for which there is historical data. Finally, the approach was then used in an offshore field which has been undergoing waterflooding, but where the choice for further field development has yet to be made, with the operator considering polymer flooding as an alternative (or in addition) to infill well drilling. The thesis discusses the implications of using this newly developed methodology in identifying the risk of failure and in assisting in making an optimal choice based on technical and economic considerations in a fully integrated manner.

Advanced Well Completions (AWCs) employing Downhole Flow Control (DFC) technology such as Inflow Control Devices (ICDs), Interval Control Valves (ICVs),Autonomous Inflow Control Devices (AICDs) and/or Annular Flow Isolations (AFIs) provide a practical solution to the challenges normally encountered by conventional wells. Both oilfield operating companies and several researchers have developed workflows to identify the optimum well location and field development well configuration. However, all these approaches do not at present consider optimising advanced well completions employing DFCs. The objective of this thesis is to provide an automated, comprehensive workflow to identify the optimum advanced well completion design that ensures an optimum well performance throughout the well’s and field’s life. This study starts by describing the history of ICD, AICD, ICV and AFI development with emphasis on the (near and) fully commercially available types and their areas of application. The thesis then reviews the flow performance of available ICD, ICV and AICD types. It reviews the available advanced completion modelling techniques and their historical development. This allows provision of guidelines on how to model DFC technologies performance when combined with AFIs over the well’s life. It shows how the value of such well-construction options can be quantified using these tools. The thesis introduces a novel workflow outlining the process of designing ICD completions with or without AFIs for different well architectures applied in different reservoir types for production or injection purposes. The workflow incorporates: the ICD restriction sizing; the requirement for AFI, their frequency and distribution; the impact of ICD reliability throughout the life of the well, the effect of uncertainty on the design parameters, installation risks and the resulting economic value. This workflow is then extended to the design and evaluation of AICD completions, through identification of the optimum control of water and excess gas production. The value and applicability of the proposed workflow is verified using synthetic and real field case studies. The latter include three oil fields (H-Field, S-Field and U-Field), one thin oil column/gas condensate field (NH-Field) and a gas field (C-Field). These cases also illustrated the value which can be gained from the application of Downhole Flow Control technologies.

Most reservoirs, such as carbonate reservoirs not only have structural heterogeneities (e.g. complexly shaped geobodies or fractures). But they also have distributed wettabilities and are mixed- to oil-wet. The interplay of structural and wettability heterogeneities impacts sweep efficiency and oil recovery. Choosing the appropriate Improved Oil Recovery (IOR) or Enhanced Oil Recovery (EOR) technique based on adequate predictions of oil recovery requires a sound understanding of the fundamental controls on fluid flow in mixed- to oil-wet andstructurally complex rocks. The underlying multiphase flow processes are modelled with physically robust flow functions, i.e. relative permeability and capillary pressure functions. Obtaining these flow functions is a challenging task, especially when three fluid phases coexist, such as during Water-Alternating-Gas (WAG) injection. In this work we use pore-network modelling, a reliable and physically based simulation tool, to predict three-phase flow functions. We have developed a new three-phase flow pore-network model for rocks with arbitrary wettability, which allows us to analyse the fundamental multi-phase displacement processes. Unlike other models, our model combines three main features: (I) A novel thermodynamic criterion for formation and collapse of oil layers that strongly depends on the fluid spreading behaviour and the rock wettability. The model hence captures film/layer flow of oil accurately, which impacts, in particular, the oil relative permeability at low oil saturation and hence the accurate prediction of residual oil. (II) Multiple displacement chains, where injection of one phase at the inlet triggers a chain of interface displacements throughout the network. This allows accurate modelling of the mobilization of disconnected phase clusters that arise during higher order (WAG) floods. (III) The model takes as input realistic 3D pore-networks extracted from pore-space reconstruction methods and Computed Tomography (CT) images, preserving both topology and pore shape of the rock. The model comprises a constrained set of parameters that can be tuned to mimic the wetting state of a given reservoir. We have validated our model against available experimental data for a range of wettabilities. We demonstrate the importance of film and layer flow for the continuity of the various phases during subsequent WAG cycles and for the residual oil saturations. A sensitivity analysis has been carried out with the full 3D model to predict three-phase relative permeabilities and residual oil saturations for WAG cycles under various wetting conditions with different flood end-points and for different rock types. This revealed a wide range of three-phase relative permeabilities and residual saturations. The pore-scale generated three-phase flow functions have then been used in a heterogeneous reservoir model. Here we demonstrate their impact on the sweep efficiency after gas injection and WAG for a range of realistic wettability scenarios. We show that the uncertainty in flow functions can be as big as the geological uncertainty in a reservoir model that was history matched for an extended waterflood.

History matching is a very important activity during the continued development and management of petroleum reservoirs. Time-lapse (4D) seismic data provide information on the dynamics of fluids in reservoirs, relating variations of seismic signal to saturation and pressure changes. This information can be integrated with history matching to improve convergence towards a simulation model that predicts available data. The main aim of this thesis is to develop a method to speed up the convergence rate of assisted seismic history matching using proxy derived gradient method. Stochastic inversion algorithms often rely on simple assumptions for selecting new models by random processes. In this work, we improve the way that such approaches learn about the system they are searching and thus operate more efficiently. To this end, a new method has been developed called NA with Proxy derived Gradients (NAPG). To improve convergence, we use a proxy model to understand how parameters control the misfit and then use a global stochastic method with these sensitivities to optimise the search of the parameter space. This leads to an improved set of final reservoir models. These in turn can be used more effectively in reservoir management decisions. To validate the proposed approach, we applied the new approach on a number of analytical functions and synthetic cases. In addition, we demonstrate the proposed method by applying it to the UKCS Schiehallion field. The results show that the new method speeds up the rate of convergence by a factor of two to three generally. The performance of NAPG is much improved by updating the regression equation coefficients instead of keeping it fixed. In addition, we found that the initial number of models to start NAPG or NA could be reduced by using Experimental Design instead of using random initialization. Ultimately, with all of these approaches combined, the number of models required to find a good match reduced by an order of magnitude. We have investigated the criteria for stopping the SHM loop, particularly the use of a proxy model to help. More research is needed to complete this work but the approach is promising. Quantifying parameter uncertainty using NA and NAPG was studied using the NA-Bayes approach (NAB). We found that NAB is very sensitive to misfit magnitude but otherwise NA and NAPG produce similar uncertainty measures.

The formation of mineral scales such as barium sulphate and calcium carbonate remains an issue for the oil industry, after many years of oil exploration. In the last 10 years, the difficulty in dealing with scale deposition has been accentuated by the appearance of more complex conditions, involving complicated well completions for deepwater or long sub-sea tiebacks. If scale control measures fail in these situations then long distances between the scale deposits and the production platform are present. Intervention into such systems is either impossible or extremely expensive. To combat such problems, the front end engineering design stage (FEED) now attempts to bring together multidisciplinary teams to provide a full risk assessment of all areas in which production chemistry problems might arise. Hence, benefits come from each discipline team having as much knowledge as possible available to them. This thesis aims to fuel this knowledge by developing a fundamental understanding of how various factors, conditions or environmental, impact scale inhibitor mechanisms, so that the results can be incorporated into the FEED process. Key areas affecting scale inhibitor operation were investigated. From these studies, a number of important findings can be highlighted. The presence of calcium was found to improve scale inhibitor (SI) performance, especially phosphonate types, whilst magnesium ions had little effect on polymeric performances and detrimentally affected the phosphonates’ inhibition efficiency (IE). These trends were related to the SI affinity for the divalent ions – polymer PPCA binds to calcium but shows incompatibility at [Ca2+] > 1000ppm - observed as low IE, whilst the phosphonate DETPMP binds with either ion but prefers calcium. Two inhibition mechanisms - nucleation and crystal growth blocking - were identified for different types of SI species and were illustrated using static IE tests relating IE to [SI] left in solution. High IE corresponds to high [SI] and similarly low IE with low [SI]. These initial results have since been investigated further in a additional study. An extensive range of phosphonate and polymeric scale inhibitor species can now be classified as i. either Type 1 or 2 (based on IE, Ca2+ and Mg2+ sensitivity ration and SI consumption tests) or ii. either Type A or B (based on compatibility/incompatibility with [Ca2+]= ~1000-2000ppm+). A requirement for both homogeneous and heterogeneous nucleation to be investigated for a scaling system was identified, as deposition kinetics can vary requiring different ii levels of SI. A [SI] falling below minimum inhibitor concentration (MIC), can promote surface scaling. Hence, scaling systems should be studied experimentally over a range of temperatures, to represent the conditions from sub-sea tiebacks to the production well. A model was developed from experimental data enabling the prediction of safe sulphate levels and mass of barite deposited. This model can be applied to un-seeded and seeded tests where, as expected, the foreign particles accelerated the reaction to equilibrium with the greatest deposition rate for barite over sand and for a higher surface area over a lower one. Both theoretical and experimental confirmation of each retention mechanism occurring in a porous medium was achieved. This adsorption/precipitation model has been incorporated into Squeeze VII, an in-house squeeze design software, to allow a better physical description of a squeeze treatment. The predictions of Squeeze VII have also been improved by using the more accurate data for the scale inhibitor return concentrations from core floods due to the better developed analysis techniques. The direct value of these improvements to industry is significant. These advances reduce OPEX costs and deferred oil production whilst giving the industry the opportunity of improved future lifetime predictions and operations.

One of the main objectives of this work was to investigate the applicability and accuracy of artificial neural networks for estimating special core analysis (SCAL) parameters from minimal core training data and wireline logs. The SCAL data was obtained from measurements on core plugs undertaken at the Libyan Petroleum Institute (L.P.I). Previous neural network studies have attempted to predict routine core analysis parameters, such as permeability, but not SCAL parameters such as true formation resistivity (Rt), resistivity index (RI), water saturation (Sw), saturation exponent (n) and Amott-Harvey Wettability Index (IA/H). Different combinations of wireline logs were used to train a variety of neural network predictors. Some of the predictors were trained using a large dataset from the entire cored interval of the training well. Other genetically focused neural network (GFNN) predictors were trained just from one short representative genetic unit (RGU) in the training well. The predictors were then tested in an adjacent well in the same oil field and also in another well in a different oil field. Significantly the performance of the GFNN predictors was as good (and in most cases better) than the predictors trained on the much larger dataset. This demonstrated the useful of the GFNN approach, which is very cost effective in terms of the minimal core that is required, and the reduced computer processing time. Moreover, this is the first time that these GFNN predictors have been used to predict SCAL parameters in the studied area, the Nubian Sandstone Formation in North Africa. These neural network predictors are particularly useful in this area due to the limited amount of SCAL data that is currently available. Quantitative statistical measures of heterogeneity were also examined on the reservoir samples, followed by a comparative analysis of hydraulic units (HUs) with a newer approach of global hydraulic elements (GHEs) to characterize the reservoir units in the studied area. The GHEs were then applied to select minimal representative core training data to train the genetically focused neural networks (GFNNs) to predict the SCAL parameters. The thesis also describes the factors affecting SCAL resistivity parameters. Laboratory measurements on the Nubian Sandstone reservoir rock samples showed changes in the formation resistivity factor (F) and cementation exponent (m) between ambient conditions and at overburden pressures. Changes were also observed in the saturation exponent (n) before and after wettability measurement. The experimental results also showed that there was a good relation between resistivity and the type of pore system which is consistent with study result from Swanson (1985) confirming earlier work.

Gas hydrates can cause serious economic/safety concerns in oil and gas production operations. Recently, low dosage polymeric Kinetic Hydrate Inhibitors (KHIs) have seen increasing industry use as alternatives to traditional thermodynamic inhibitors (e.g. methanol, glycol). To date, KHIs have been primarily understood to work by delaying/interfering with the hydrate nucleation process, inhibiting the onset of hydrate growth for a significant ‘induction time’ (ti) period. If the induction time exceeds fluid residence time in the hydrate region, then hydrate formation/plugging is avoided. However, due to nucleation being probabilistic, induction time data measured in standard laboratory KHI evaluation studies are often highly stochastic, making KHI assessment problematic and time-consuming To address this problem, the primary aim of this project was to develop a crystal growth inhibition (CGI) based approach to KHI evaluation. In this technique gas hydrate growth and dissociation patterns in the presence of KHI polymers were carefully inspected to evaluate repeatability of features and the existence of any consistency between runs and transferability between set ups within KHI systems. Extensive studies using this method show that KHIs - rather than being solely ‘nucleation delayers’ - induce a number of highly repeatable, well-defined hydrate crystal growth inhibition regions as a function of subcooling, ranging from complete inhibition, through reduced growth rates to ultimate failure with increasing subcooling. These crystal growth inhibition properties, in addition to offering further protection against hydrate formation/plugging (e.g. if hydrate nucleation does occur), provide a means to evaluate formulations much more rapidly and reliably. These measured CGI regions have shown good correlation with traditional induction time data, meaning CGI methods can be used to both rapidly approximate ti patterns and support/confirm ti test results, speeding up the KHI evaluation process while giving greatly increased operator confidence in inhibitor performance. Furthermore in this project, the new approach has been applied for evaluating the performance of different types of kinetic hydrate inhibitors as well as assessing the influence of various other components (e.g. liquid hydrocarbons, salts and thermodynamic inhibitors) on KHI performance. Moreover, studies have been conducted on KHI evaluation in different hydrate structure systems (i.e., Structure I, structure II and structure H) systems in the presence of several different single, binary and multi-component gases. For this purpose in all experiments undertaken throughout this thesis, with the application of the newly developed CGI technique, crystal growth inhibition regions have been measured for different systems and from the extent of these regions, hydrate inhibition properties of each KHI system has been evaluated and analysed. Results of these studies proved that the pendant group of a polymer plays a major role on the KHI inhibition properties. Also investigations of different guest gas/hydrate structure systems using the new CGI technique indicated that guest/cage occupancy plays an important role in hydrate inhibition and different hydrate structure systems (e.g. s-I, s-II and s-H) are inhibited differently by the same KHI. For instance, PVCap performance was considerably superior in s-II and s-H forming systems compared to s-I forming systems (e.g. methane), supporting stronger polymer adsorption on s-II or s-H hydrate crystal surfaces. Also through the newly developed CGI studies, it was found that while the presence of NaCl enhances PVCap methane hydrate inhibition, a carbonate salt like K2CO3 can have a generally negative effect on PVCap performance. In addition to that, test on liquid hydrocarbons proved that the presence of these compounds can slightly deteriorate PVCap performance. Moreover, results indicated that the combination of thermodynamic inhibitors and PVCap show better performance than thermodynamic inhibitors alone although glycols generally acted as ‘top-up’ thermodynamic inhibitor with PVCap which was a much better compared to the performance of alcohols with PVCap.

This Thesis investigates the two main limitations of high temperature gas chromatography (HTGC) in the analysis of heavy n-alkanes: pyrolysis inside the GC column and incomplete elution. The former is studied by developing and reducing a radical pyrolysis model (7055 reactions) into a molecular pyrolysis model (127 reactions) capable of predicting low conversions of (nC14H30-nC80H162) at temperatures up to 430°C. Validation of predicted conversion with literature data for nC14H30, nC16H34 and nC25H52 yielded an error lower than 5.4%. The latter is addressed by developing an analytical model which solves recursively the diffusion and convection phenomena separately. The model is capable of predicting the position and molar distribution of components, using as main input the analytes’ distribution factors and yielded an error lower than 4.4% in the prediction of retention times. This thesis provides an extension of the data set of distribution factors of (nC12H26– nC98H198) in a SGE HT5 GC capillary column, based on isothermal GC measurements at both constant inlet pressure and flow rate. Finally, the above two models were coupled, yielding a maximum mass lost of 1.3 % in the case of nC80H162 due to pyrolysis and complete elution up to nC70H142, in a 12 m HT5 column.

The uncertainty in the produced oil volume can be minimised by substituting intelligent wells (IWs) for conventional wells. A previous study showed that IWs reduce the impact of geological uncertainty on the production forecast (Birchenko, Demyanov et al. 2008). This investigation has now been extended to the “dynamic” parameters (fluid contacts, relative permeabilities, aquifer strength and zonal skin). The efficiency of the IWs in reducing the total production uncertainty due to the reservoir’s dynamic parameters was found to be comparable to that reported for the static parameters. However, this later study identified that the result was strongly dependent on the strategy employed to optimise the field’s performance. Experience has shown that challenges arise while using commercial software for optimisation of a typical, modern field with multiple reservoirs and a complex surface production network. Inclusion of the optimisation algorithm dramatically increases the calculation time in addition to showing stability and convergence problems. This thesis describes the development of a novel method of a reactive control strategy for ICVs that is both robust and computationally fast. The developed method identifies the critical water cut threshold at which a well will operate optimally when on/off valves are used. This method is not affected by the convergence problems which have lead to many of the difficulties associated with previous efforts to solve our non-linear optimisation problem. Run times similar to the (non-optimised) base case are now potentially possible and, equally importantly, the optimal value calculated is similar to the result from the various optimisation software referred to above. The approach is particularly valuable when analysing the impact of uncertainty on the reservoir’s dynamic and static parameters, the method being convergent and independent of the point used to initiate the optimization process. “Tuning” the algorithm’s optimisation parameters in the middle of the calculation is no longer required; thus ensuring the results from the many realisations are comparable.