Agbami is a deepwater oil field located offshore Nigeria in the Gulf of Guinea. The field has been online since July 2008 and has produced over 275 MMBO to date. The Turbidite depositional setting has lead to a complex stratigraphic architecture with reservoirs being broken into distinct subunits. Although initially in thermodynamic equilibrium they posed dynamic challenges of conformance control and differential depletion. For this reason, Intelligent Well Completions (IWC) were installed throughout the field in both injectors and producers. The field development currently consists of 39 completions across 21 wellbores which are individually isolated and controlled via Interval Control Valves (ICV). Combined with multiple downhole pressure/temperature sensors and inline flowmeters the Agbami IWC provides the real time surveillance and control necessary for optimization of field performance and recovery. During development planning, identifying the potential value from the installation of IWCs and deciding on or recognizing their possible applications is often not a trivial one. The value can be hard to quantify where the information impacts multiple decisions, there is limited in-house experience or where robust forecasts are difficult to generate. The objective of this paper is to bridge some of these gaps through providing an overview of the application of the IWC within Agbami from the workstation of the Production Engineer. Although often overlooked at the early stages of field development, IWCs can provide significant value to the Production Engineer who focuses on the short to midterm field performance and long-term completion reliability. The paper will cover specific examples of value added in relation to production performance over the last three years of field history. IWCs are used actively by the production engineering team with the objective of prudent reservoir management through integration of their capabilities into surveillance plans, production management practices and production optimization efforts. The expanding employment of IWCs for existing or new field developments warrants further literature which shares detailed examples of IWC practices and benefits. The intention of this paper being to focus on the production engineering aspects and values of the IWC as it has been utilized within Agbami at the operations level to improve reservoir management.

Coiled tubing fracturing has been successfully applied in multi-stage vertical well stimulation in the Belridge diatomite in the Lost Hills field. This same methodology was used to complete two northwesttrending horizontal wells drilled on the northeast flank of the Lost Hills anticlinal structure that targeted thinner higher oil-saturation strata, separated by thicker low oil-saturation intervals. The target reservoir is comprised of high porosity, low matrix permeability Opal A diatomite.The perforations were jetted by pumping sand slurry down the coiled tubing and the frac job was pumped down the annulus. The stages were isolated by setting sand plugs. Nine and twelve stages were pumped in the two wells respectively. The perforation locations for different stages were selected in areas with: 1) high resistivity and inferred high oil saturations, 2) absence of hydraulic fractures from nearby wells, 3) excellent cement bonding, and 4) low intensity of natural fractures. These assessments followed logging while drilling (LWD) gamma ray, induction resistivity and azimuthally focused resistivity (image) logs and cased-hole ultrasonic image tool (USIT) run with the aid of a tractor. The hydraulic fractures were monitored using surface tiltmeter sensors. Oil and water soluble tracers were pumped to determine the relative production contribution from the stages and fracture fluid cleanup, respectively, from the stages. All the jobs could be successfully pumped without any screen outs. Challenges were faced in setting sand plugs and isolating stages. Large fracture widths and low leak-off into the formation led to difficulty in forming sand bridges at the perforations and concentrating sand in the wellbore for the plugs. Surface tiltmeters showed excessive fracture height growth. Tracer results showed that 20-30% of the stages contributed to 50-60% of the production. Stages with higher treating pressures contributed less towards production. This could be attributed to near wellbore tortuosity in these stages. Proppant flowback was encountered in one well, and after an effective clean up the production rose.The study illustrates how integration of various aspects such as completion design, fracture pressure analysis and diagnostics combined with geologic and reservoir information can help in identifying challenges and finding potential solutions of hydraulic fracturing. The findings highlight that the technology most suitable for vertical well stimulation might not be favorable for horizontal well stimulation.

To improve the reliability of reservoir performance predictions, subsurface uncertainties must be accounted for in production forecasts. Probabilistic methods are commonly used to understand and quantify the impact of uncertainties on reservoir behavior. This paper presents a structured and practical probabilistic history-matching and production forecasting workflow that was successfully applied to 6 reservoirs in a West-Africa field with several years of production history and a challenging data monitoring environment. The workflow was found to be very efficient as the 6 reservoir models were constructed, history-matched and run in predictions in less than three months. A recent look-back on the probabilistic predictions with a year of new production data proved the robustness of the workflow. The procedure used in this paper starts with a thorough review of subsurface uncertainties. All available static and dynamic data is analyzed to define uncertainty parameters and corresponding ranges. Next, a first set of simulations is performed, with each uncertainty parameter varied independently in order to analyze its effect on history-matched quality and future reservoir performance. The parameters with little impact are screened out during this step. The key parameters retained are then used to define a new set of simulations through experimental design. The models are run and the results are used to generate response surfaces for each history-match parameter and reservoir performance metric. Using a Monte-Carlo sampling procedure, thousands of uncertainty parameter combinations are tested using the response surfaces and screened using tolerances on various history-match parameters. This approach avoids the cumbersome and subjective definition of an objective function and allows the selection of a large number of parameter combinations that yield a history-match. Several models were selected to represent the 10th, 50th and 90th percentile of original oil in place and reservoir ultimate oil recovery. These probabilistic models are then run into prediction under different development scenarios, allowing for optimization of well locations and field operational constraints.

A new method is introduced to calculate the average reservoir pressure around wells from pressure buildup or falloff tests regardless of the shape of the drainage area or the types of boundaries surrounding it. Practicing engineers can apply the proposed method using any of the widely available Well Testing programs. Average reservoir pressure is used in just about all reservoir engineering studies conducted via simulation, material balance, or even decline curve analysis. Existing techniques, such as the Matthews-Brons-Hazebroek (MBH) method (1954), are valid for limited drainage shapes and boundary conditions. In particular, they cannot handle secondary recovery projects (waterflooding, steam injection, etc.), or irregular drainage shapes which are routinely used in the industry. As a result, typically the value of p* is used as a substitute for average reservoir pressure, which may result in significant error. The new method enables petroleum engineers to calculate the average reservoir pressure accurately for every reservoir type by integrating transient, production, and static data. Our technique calculates a robust approximation of the drainage area based on the production/injection history for each well in the reservoir. A given well may be offset by other injection and production wells and no-flow and/or constant-pressure boundaries, all of which are taken into account for estimation of the drainage area. Our technique also accounts for areal variations in reservoir thickness and depth. The size and shape of the drainage area directly impacts the average pressure characteristics, which can be calculated with the help of relatively simple runs using a reservoir simulator. The results are generalized by creating type curves similar to the MBH type curves for each well in the reservoir. The new method was applied to several wells in a large oil field offshore Nigeria with the methodology and results presented in this paper. The new technique enables us to calculate the average reservoir pressure accurately for every reservoir type, all drainage area shapes and boundary conditions, a variety of production patterns, and for multiphase flow conditions. It is difficult to overestimate the value of properly calculated average reservoir pressure. It is one of the main parameters used to history match field data. It affects multiple drilling and production optimization decisions like, infill drilling, mud pressure, and determining flow rates of producers and injectors. With accurate average reservoir pressure, optimization of field management is possible by determining the number and the locations of infill wells, the level of injection and production from different wells, and other reservoir performance related decisions. The new method significantly improves our ability to produce valid values of average reservoir pressure.

A new light weight blended cement has been de-developed in Canada for solving a series of oper-ational problems for cementing across low pres-sure formations were foam cements are usually applied. The slurry density can be varied from 1,100 to 1,400 Kg/m This system imparts good slurry properties such as uniform cement column density, zero free wa-ter, non-settling, reliable rheology, fluid loss and thickening time control behaviors. During the transition from liquid to solid state, the system develops stable gel strength, high shear bond strength, and high compressive strength even at low temperatures. References and illustrations at end of paper In field applications, the system reduces signif-icantly equipment and manpower requirements when compared with foam cementing job oper-ations, and can eliminate the need for a stage collar. No cement fall-back has been seen after implementing more than 25 jobs. Case histories including an on-site computer job record are dis-cussed. INTRODUCTION Light weight cements are solutions for cement-ing across rock formations which are fractured or having a low Eracture gradient, highly perme-abable, vuggy or cavemous. These weak forma-tions usually have very low pressure that they may not be able to support the hydrostatic pres-sure from a collumn of normal cement slurries.