Hydraulic Flow Unit Classification and Prediction Using Machine Learning Techniques: A Case Study from the Nam Con Son Basin, Offshore Vietnam

The test study area is the Miocene reservoir of Nam Con Son Basin, offshore Vietnam. In the study we used unsupervised learning to automatically cluster hydraulic flow units (HU) based on flow zone indicators (FZI) in a core plug dataset. Then we applied supervised learning to predict HU by combining core and well log data. We tested several machine learning algorithms. In the first phase, we derived hydraulic flow unit clustering of porosity and permeability of core data using unsupervised machine learning methods such as Ward’s, K mean, Self-Organize Map (SOM) and Fuzzy C mean (FCM). Then we applied supervised machine learning methods including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Boosted Tree (BT) and Random Forest (RF). We combined both core and log data to predict HU logs for the full well section of the wells without core data. We used four wells with six logs (GR, DT, NPHI, LLD, LSS and RHOB) and 578 cores from the Miocene reservoir to train, validate and test the data. Our goal was to show that the correct combination of cores and well logs data would provide reservoir engineers with a tool for HU classification and estimation of permeability in a continuous geological profile. Our research showed that machine learning effectively boosts the prediction of permeability, reduces uncertainty in reservoir modeling, and improves project economics.

Artificial Intelligence Method of Flow Unit Division Based on Waveform Clustering: A Case Study on Zhetybay Oil Field, South Mangyshalk Basin, Kazakhstan

The Zhetybay Field is located in the South Mangyshlak Sub-basin, a delta front sedimentary reservoir onshore western Kazakhstan. It was discovered in 1961 and first produced by waterflooding in 1967. After more than 50 years of waterflooding development, the reservoirs are generally in the mid-to-high waterflooded stage and oil-water distribution becomes complicated and chaotic. It is very difficult to handle and identify so much logging data by hand since the oilfield has the characteristics of high-density well pattern and contains rich logging information with more than 2000 wells. The wave clustering method is used to divide the sedimentary rhythm of the logging curve. Sedimentary microfacies manifested as a regression sequence, with four types of composite sand bodies including the composite estuary bar and distributary channel combination, the estuary bar connected to the dam edge and the distributing channel combination, the isolated estuary bar and distributing channel combination, and the isolated beach sand. In order to distinguish the flow units, the artificial intelligence algorithm-support vector machine (SVM) method is established by learning the non-linear relationship between flow unit categories and parameters based on developing flow index and reservoir quality factor, summarizing permeability logarithm and porosity degree parameters in the sedimentary facies, and analyzing the production dynamic. The flow units in Zhetybay oilfield were classified into 4 types: A, B1, B2 and B3, and the latter three are the main types. Type A is distributed in the river, type B1 is distributed in the main body of the dam, type B2 is mainly distributed in the main body of the dam, and some of B2 is distributed in the dam edge, and B3 is located in the dam edge, sheet sand and beach sand. The results show that the accuracy of flow unit division by support vector machines reaches 91.1%, which clarifies the distribution law of flow units for oilfield development. This study is one of the significant keys for locating new wells and optimizing the workovers to increase recoverable reserves. It provides an effective guidance for efficient waterflooding in this oilfield.

Integrating XRD and Well Logging Data to Establish Electro-Facies and Permeability Models for an Unconventional Heterogeneous Tight Gas Reservoir, Obaiyed Giant Gas Field

Formation evaluation in heterogeneous reservoirs can be very challenging especially in fields that extend over several kilometers in area where the permeability varies from 0.1 mD up to 1000 D within the same porosity. The porosity, hydrocarbon saturation and net sand thickness in most of Obaiyed field wells are consistent; hence, the productivity of these wells is enormously dependent on the reservoir permeability. Since the permeability is highly heterogeneous, initial production rate of the wells varies between few MMSCFD to almost one hundred MMSCFD. The huge permeability variation led to a tremendous uncertainty in the dynamic modeling, which resulted in an inaccurate production forecast affecting the field economics estimation. Understanding permeability distribution and heterogeneity in Obaiyed field is the key factor for establishing a realistic permeability model, which will lead to a successful field development strategy. Extensive work was performed to understand key factors that govern the permeability in Obaiyed using the data of 1-kilometer length of cores acquired in more than 50 wells covering different reservoir properties in the field. Core data were used to separate the reservoir into different Hydraulic Flow Units (HFU) according to Amaefule's work performed on the Kozeny-Carmen model. Afterwards, a correlation between the HFU and well logs was established using IPSOM Electro-Facies module in order to define the flow units in un-cored wells. The result of this correlation was used to calibrate a Porosity-Permeability relationship for each flow unit. The next step was examining the clay-type distribution and diagenesis in each flow unit using the petrographic analysis (XRD) results from the core xdata. All factors controlling the permeability can now be represented in hydraulic flow units which are considered as a method of measurement of the reservoir quality. Consequently, property maps were constructed showing the location and continuity of each of the flow units, leading to a more deterministic approach in the well placement process. Based on this new work methodology, a production cut-off criteria relating the reservoir productivity to both clay minerals presence and percentages was established for multiple wells scenarios. As a result, the development strategy of the field changed from only vertical wells to include horizontal wells as well which proved to be the only economic approach to produce the Illite dominated zones. This paper presents a workflow to provide a representative estimation of permeability in extremely heterogeneous reservoirs especially the ones dominated by complex clay distribution.

Artificial Intelligence-Based, Automated Rapid Reservoir Assurance and Reservoir Health Diagnostics in a Complex Offshore Mature Field

The QQ Field in the Gulf of Suez is a mature, geologically complex with multiple stacked, faulted reservoirs, with commingled production between different reservoirs. This paper illustrates the power of an automated tool to perform systematic, rapid, and detailed assessment of the reservoir performance, identify the key recovery obstacles and prepare remedial plans to enable the reservoir to produce to its full potential. The well and reservoir data were processed to compute a series of metrics and key performance indicators at various levels (well, layer, reservoir, well groups, etc.). The tool has several automated modules to facilitate rapid, metric-driven reservoir assurance and management. These modules include: (i) well production/injection allocation, (ii) wells decline curve analysis including event-detection, (iii) pressure and voidage analysis, and (iv) Contact analysis. Using performance analytics, the study quickly identified ways to improve the health of the reservoir and maximize its value. The QQ Field predominantly produces from two formations: Nubia and Nezzazat. Furthermore, there are multiple sub-layers in each formation. Reliable flow unit allocation is critical to gauge contribution of each layer, identify the undrained areas of the reservoir, and locate future development opportunities. The flow unit allocation module incorporates all available data such as PLT/ILT data, completion history, permeability of each flow unit at well level, relative pressures, and water influx model. Based on the allocated production, the current recovery factors in Nubia and Nezzazat are approximately 60% and 20% respectively. Analysis of RFT data reveals good vertical communication across Nubia. However, in some areas there is clear pressure discontinuity across layers. The reservoir pressure has dropped below the bubble point in both formations. As a result, water injection was initiated. The pressure in all parts of Nubia was restored above bubble point. Aquifer influx is sufficient to support the current withdrawal rates and further water injection is unnecessary. However, Nezzazat has a significantly higher reservoir complexity and therefore, shows a large variation in pressure behavior. It needs water injection to maintain the reservoir pressure above the bubble point in all parts of the reservoir. Based on the flow-unit allocation, the voidage replacement ratio (VRR) was calculated for each area and each layer. Even though the overall VRR in the waterflooded areas is above one, the distribution of the injected water is uneven. Redistributing injected water and ensuring that all the areas and all the layers are adequately supported will help to maximize recovery. The prolonged dip in oil price demands extreme efficiency. Sound reservoir management must not require unreasonable time or manpower. The rapid, automated analysis enables quick identification of the key areas for improvement in reservoir management practices and maximize the value of the asset.

Rock typing based on hydraulic and electric flow units for reservoir characterization of Nubia Sandstone, southwest Sinai, Egypt

The present work describes and evaluates the reservoir quality of the sandstone of the Nubia Formation at the Gebel Abu Hasswa outcrop in southwest Sinai, Egypt. Hydraulic flow unit (HFU) and electrical flow unit (EFU) concepts are implied to achieve this purpose. The Paleozoic section made up of four formations has been studied. The oldest is Araba Formation followed by Naqus formations (Nubia C and D) overlay by Abu Durba, Ahemir and Qiseib formations (Nubia B), where the Lower Cretaceous (Nubia A) is represented by the Malha Formation. The studied samples have been collected from Araba, Abu Durba, Ahemir and the Malha formations. The hydraulic flow unit (HFU) discrimination was carried out based on permeability and porosity relationship, whereas the electrical flow unit (EFU) differentiation was carried out based on the relationship between formation resistivity factor and porosity. Petrographic investigation of the studied thin sections illustrates that the studied samples are mainly quartz arenite. Important roles to enhance or reduce the pore size and/or pore throats controlling the reservoir petrophysical behavior are due to the diagenetic processes. The present study used the reservoir quality index (RQI) and Winland R35 as additional parameters applied to discriminate the HFUs. The study samples have five hydraulic flow units of different rock types, where the detected electrical flow units are only three. The differences between them are may be due to the cementation process with iron oxides that might act as pore filling, lining and pore bridging, sometimes bridges helping to decrease permeability without serious reduction in porosity. The reduction between the number of EFUs and HFUs comes from the effect of diagenesis processes which is responsible for a precipitation of different cement types such as different clay minerals and iron oxides.

Hydraulic flow unit classification from core data: case study of the Z gas reservoir, Poland

Permeability and porosity are essential parameters for estimating hydrocarbon production from reservoir rocks. They are combined in an additional factor, the Flow Zone Index (FZI), which is the basis for defining the hydraulic flow unit (HFU). Each HFU is a homogeneous section of a reservoir rock with stable parameters that allow for media flow. Hydraulic flow units are determined from the porosity and permeability of core or well logs. The simple statistical methods are applied for HFU classification and then improve permeability prediction. This paper also shows how to quickly apply the global hydraulic elements (GHE) method for HFU classification. The methodology is tested on the Miocene formation of a deltaic facies from the Carpathian Foredeep in South-Eastern Poland.