A major Malaysian offshore oilfield, which is currently operating under waterflooding for a quite long time and declining in oil production, plan to convert as chemical enhanced oil recovery (CEOR) injection. The CEOR journey started since the first oil production in year 2000 and proximate waterflooding, with research and development in determining suitable method, encouraging field trial results and a series of field development plans to maximize potential recovery above waterflooding and prolong the production field life. A comprehensive EOR study including screening, laboratory tests, pilot evaluation, and full field reservoir simulation modelling are conducted to reduce the project risks prior to the full field investment and execution. Among several EOR techniques, Alkaline-Surfactant (AS) flooding is chosen to be implemented in this field. Several CEOR key parameters have been studied and optimized in the laboratory such as chemical concentration, chemical adsorption, interfacial tension (IFT), slug size, residual oil saturation (Sor) reduction, thermal stability, flow assurance, emulsion, dilution, and a chemical injection scheme. Uncertainty analysis on CEOR process was done due to the large well spacing in the offshore environment as compared to other CEOR projects, which are onshore with shorter well spacing. The key risks and parameters such as residual oil saturation (Sorw), adsorption and interfacial tension (IFT) cut-off in the dynamic chemical simulator have been investigated via a probabilistic approach on top of deterministic method. The laboratory results from fluid-fluid and rock-fluid analyses ascertained a potential of ultra-low interfacial tension of 0.001 dyne/cm with adsorption of 0.30 mg/gr-of-rock, which translated to a 50-75% reduction in Sor after waterflooding. The results of four single well chemical tracer tests (SWCTT) on two wells validated the effectiveness of the Alkaline Surfactant by a reduction of 50-80% in Sor. The most suitable chemical formulation was found 1.0 wt. % Alkali and 0.075 wt. % Surfactant. The field trial results were thenceforth upscaled to a dynamic chemical simulation; from single well to full field modeling, resulting an optimal chemical injection of three years or almost 0.2 effective injection pore volume, coupled with six months of low salinity treated water as pre-flush and post-flush injection. The latest field development study results yield a technical potential recoverable volume of 14, 16, and 26 MMstb (above waterflooding) for low, most likely, and high cases, respectively, which translated to an additional EOR recovery factor up to 5.6 % for most-likely case by end of technical field life. Prior to the final investment decision stage, Petronas’ position was to proceed with the project based on the techno-commerciality and associated risks as per milestone review 5, albeit it came to an agreement to have differing interpretations towards the technical basis of the project in the final steering committee. Subsequently, due to the eventual plunging global crude oil price, the project was then reprioritized and adjourned correspondingly within Petronas’ upstream portfolio management. Further phased development including a producing pilot has been debated with the main objective to address key technical and business uncertainties and risks associated with applying CEOR process.

Integrated reservoir modeling with representative data is crucial for an effective reservoir management and depletion plan. Both analytical and numerical approaches benefit from this integrated process. The objective of this study is to incorporate the outcomes of analytical techniques such as rate transient analysis (RTA) and pressure transient analysis (PTA) into numerical reservoir model to have a better understanding of drive mechanisms, reservoir connectivity with minimal time-consuming for history matching efforts but a more reliable production forecast. In order to demonstrate the methodology, a clastic reservoir from Malay basin was considered. Sedimentology and sequence stratigraphy studies were performed to have a better picture of heterogeneity and zonation of the reservoir. All production and injection data were investigated along with pressure data to filter data inconsistency. Shut-in time should be long enough to take representative reservoir pressure and accordingly material balance study conducted for accessible volume for a given area. However, flowing material balance is able to be applied with no restriction on the production data for evaluation of historical data and prediction cases. The boundary of the channel sand was constructed based on the well log data and seismic attributes. Amplitude impedance was used as a guide for lithofacies and porosity distribution in the geological model. In addition, stratigraphy definition with further details were incorporated. Lithofacies, petrophysical and SCAL data were incorporated in rock-type classification and accordingly saturation-height-function were modelled. Analytical approaches including PTA, material balance, and RTA were utilized to have a better understanding of fluid flow and drive mechanisms. The well and reservoir properties and also connected volume from analytical approaches were utilized as a tuning tool of static model. This approach considerably reduced the iteration between static and dynamic models for history matching exercise. Afterwards, the production forecast were conducted with two development opportunities identified. In this study, an integrated methodology was applied to mitigate the complexity of history matching task. Moreover, it is demonstrated that using such analytical methods help to improve the development plan of a given field significantly.

Quality of commingled production data and reliability of back-allocation from stacked reservoirs with numerous platforms, wells, and strings play an important role in reservoir simulation modeling for history matching and prediction. A long historical data, limited surveillance data including routine well tests, pressure, and PLT, relying on conventional back-allocation for high water-cut strings with tubing integrity issues are the common pain points. The pre-HM tool is developed to provide a reliable and clean dataset for modeling. Pre-HM tool includes advanced functionalities such as data quality index (DQI), areal-vertical multiphase allocation, integrated allocation, leak identification and quantification. DQI is designed for a systematic interpretation to identify missing data, potential errors, and outliers in historical data. Multi-phase allocation helps to improve areal and vertical allocation based on the water cut behavior type curve compared to the conventional KH-method. Integrated allocation is designed to handle the uncertainties associated with metering and back-allocation using the multi-solutions approach. Clustering and ranking processes for similar solutions are validated against material balance to propose the best solutions for modeling. This technology has been deployed in several fields to examine the capability of different modules on the dataset. In most of the studies, the DQI module could demonstrate a rapid analysis of potential errors and outliers’ identifications in historical production data which was quantified by a 26% improvement. Moreover, the quality checking and cleaning process shows an improvement of 20 to 50% in timesaving compared to the conventional approach. Utilizing the advanced allocation results post DQI which were validated through material balance at reservoir sand package levels as input for the history matching process of simulation modeling works obtained a very good and reasonable history matching quality index (HMQI) improvement up to 25 to 35% in well levels. This improvement resulted in reasonable history matching without using unrealistic multipliers and reducing the uncertainties during prediction works. Leak assessment results proposed possible leaks in eleven wells which five of them were proven to have leaks based on available leak diagnostic job done by the operation team. An advanced multiple-solutions search engine combined with multi-phase deliverability models and material balance analysis to assess the uncertainty in the layer-phase allocation of different surveillance datasets compared to static KH-allocation. This tool assists reservoir engineering with data quality checks and analyzing the historical production in an effective manner and provides an alternative solution to facilitate the history matching process in dynamic simulation modeling and reduce the uncertainties range. The multi-phase splitting factors is a key advanced feature compared to the single-phase method which is commonly used in the industry. This technology can be potentially deployed during any FDP project and simulation work.

Thorough reservoir modeling studies have been performed for field ABC, however there are still challenges to be addressed in modelling of some specific sand reservoir depositional systems i.e. meandering fluvial reservoirs (point bars and crevasse splays). The current modelling approaches especially for fluvial reservoirs are mainly controlled by wells and have contributed to uncertainties in lateral variation based on geostatistic (variograms etc) between and away from well control. Moreover, the existing modelling approach is using sixth to fifth order (lower order) hierarchical architecture elements and this project further refines the model up to third order (higher order) which enables capturing lateral accretion of point bars. Advanced fluvial workflow (AFW) have been developed to improve the understanding of the reservoir architecture of fluvial reservoirs. It comprises of three main steps which are, first, details study on fluvial reservoir sedimentology characteristics derived from core analysis and literature. Second, qualitative geophysical study and interpretation derived from seismic dataset. Third, integration between the first and second steps into a three dimensional (3-D) reservoir model. As a result of AFW implementation in field ABC, this has led to better representation of the reservoir heterogeneities, more accurate STOIIP assessment, improved history matching quality index (HMQI) and enhanced subsurface risks and uncertainties understanding. This enable optimization of future field development plan such as infill well reactivation, water flood and chemical enhanced oil recovery (EOR). The AFW is a robust modelling method that can be used in any reservoir modelling platform (PETREL, CMG, RMS, TNAV) with multiple realizations capability using automated workflows.

Excessive water production associated with a decrease in hydrocarbon production is becoming a big challenge in matured offshore fields. Producing a barrel of water requires more energy that creates major economic impact on the profitability of an oil-field project. This paper presents a case study for water shut off treatment with a novel relative permeability modifier (RPM) (nano-clay). The nano-clay demonstrated high resistance to water flow (RRFw >10) and less effect to oil flow (RRFo <5) and capable to change the rock surface's wettability to more water wet. The main pilot objective was to assess the chemicals performance as part of production enhancement effort to reduce the water production from 90% to 50% water cut and to accelerate the oil production. We discussed the overall workflow, pilot execution, challenges and best practices including the laboratory results with the reference during research and development stage. The well treatment consists of bull-heading a pill of pre-flush of treated sea water for injectivity test, followed by nano-clay injection, post-flush with treated sea water, soaking for 48 hours and flow back the well. Pilot execution was completed successfully and safely within the target execution plan and are currently in monitoring stage. The post-treatment results and the overall economic success will then decide the future replication plan of this new water shut off technology.