Identification of reservoir types in deep carbonates based on mixedkernel machine learning using geophysical logging data

Identification of reservoir types in deep carbonates has always been a great challenge due to complex logging responses caused by the heterogeneous scale and distribution of storage spaces.Traditional cross-plot analysis and empirical formula methods for identifying reservoir types using geophysical logging data have high uncertainty and low efficiency,which cannot accurately reflect the nonlinear relationship between reservoir types and logging data.Recently,the kernel Fisher discriminant analysis(KFD),a kernel-based machine learning technique,attracts attention in many fields because of its strong nonlinear processing ability.However,the overall performance of KFD model may be limited as a single kernel function cannot simultaneously extrapolate and interpolate well,especially for highly complex data cases.To address this issue,in this study,a mixed kernel Fisher discriminant analysis(MKFD)model was established and applied to identify reservoir types of the deep Sinian carbonates in central Sichuan Basin,China.The MKFD model was trained and tested with 453 datasets from 7 coring wells,utilizing GR,CAL,DEN,AC,CNL and RT logs as input variables.The particle swarm optimization(PSO)was adopted for hyper-parameter optimization of MKFD model.To evaluate the model performance,prediction results of MKFD were compared with those of basic-kernel based KFD,RF and SVM models.Subsequently,the built MKFD model was applied in a blind well test,and a variable importance analysis was conducted.The comparison and blind test results demonstrated that MKFD outperformed traditional KFD,RF and SVM in the identification of reservoir types,which provided higher accuracy and stronger generalization.The MKFD can therefore be a reliable method for identifying reservoir types of deep carbonates.

Application of 9-component S-wave 3D seismic data to study sedimentary facies and reservoirs in a biogasbearing area:A case study on the Pleistocene Qigequan Formation in Taidong area,Sanhu Depression,Qaidam Basin,NW China

To solve the problems in restoring sedimentary facies and predicting reservoirs in loose gas-bearing sediment,based on seismic sedimentologic analysis of the first 9-component S-wave 3D seismic dataset of China,a fourth-order isochronous stratigraphic framework was set up and then sedimentary facies and reservoirs in the Pleistocene Qigequan Formation in Taidong area of Qaidam Basin were studied by seismic geomorphology and seismic lithology.The study method and thought are as following.Firstly,techniques of phase rotation,frequency decomposition and fusion,and stratal slicing were applied to the 9-component S-wave seismic data to restore sedimentary facies of major marker beds based on sedimentary models reflected by satellite images.Then,techniques of seismic attribute extraction,principal component analysis,and random fitting were applied to calculate the reservoir thickness and physical parameters of a key sandbody,and the results are satisfactory and confirmed by blind testing wells.Study results reveal that the dominant sedimentary facies in the Qigequan Formation within the study area are delta front and shallow lake.The RGB fused slices indicate that there are two cycles with three sets of underwater distributary channel systems in one period.Among them,sandstones in the distributary channels of middle-low Qigequan Formation are thick and broad with superior physical properties,which are favorable reservoirs.The reservoir permeability is also affected by diagenesis.Distributary channel sandstone reservoirs extend further to the west of Sebei-1 gas field,which provides a basis to expand exploration to the western peripheral area.

Early Description of Reservoir by Using a Single Well Data

ＥａｒｌｙＤｅｓｃｒｉｐｔｉｏｎｏｆＲｅｓｅｒｖｏｉｒｂｙＵｓｉｎｇａＳｉｎｇｌｅＷｅｌｌＤａｔａ￥ＱｉｎＳｈｕｎｔｉｎｇ；ＭｏＭｉｎｇｄａｏ；ＣｈｅｎＪｕｎ；ａｎｄＮａｎＸｉａｏｚｈｅｎ（ＧｅｏｌｏｇｉｃａｌＲｅｓｅａｒｃｈＩｎｓｔｉｔｕｔｅ，Ｂｕ...

Full field reservoir modeling of shale assets using advanced data-driven analytics

Hydrocarbon production from shale has attracted much attention in the recent years. When applied to this prolific and hydrocarbon rich resource plays, our understanding of the complexities of the flow mechanism（sorption process and flow behavior in complex fracture systems- induced or natural） leaves much to be desired. In this paper, we present and discuss a novel approach to modeling, history matching of hydrocarbon production from a Marcellus shale asset in southwestern Pennsylvania using advanced data mining, pattern recognition and machine learning technologies. In this new approach instead of imposing our understanding of the flow mechanism, the impact of multi-stage hydraulic fractures, and the production process on the reservoir model, we allow the production history, well log, completion and hydraulic fracturing data to guide our model and determine its behavior. The uniqueness of this technology is that it incorporates the so-called ＂hard data＂ directly into the reservoir model, so that the model can be used to optimize the hydraulic fracture process. The ＂hard data＂ refers to field measurements during the hydraulic fracturing process such as fluid and proppant type and amount, injection pressure and rate as well as proppant concentration. This novel approach contrasts with the current industry focus on the use of ＂soft data＂（non-measured, interpretive data such as frac length, width,height and conductivity） in the reservoir models. The study focuses on a Marcellus shale asset that includes 135 wells with multiple pads, different landing targets, well length and reservoir properties. The full field history matching process was successfully completed using this data driven approach thus capturing the production behavior with acceptable accuracy for individual wells and for the entire asset.

Comprehensive security risk factor identification for small reservoirs with heterogeneous data based on grey relational analysis model

Identification of security risk factors for small reservoirs is the basis for implementation of early warning systems.The manner of identification of the factors for small reservoirs is of practical significance when data are incomplete.The existing grey relational models have some disadvantages in measuring the correlation between categorical data sequences.To this end,this paper introduces a new grey relational model to analyze heterogeneous data.In this study,a set of security risk factors for small reservoirs was first constructed based on theoretical analysis,and heterogeneous data of these factors were recorded as sequences.The sequences were regarded as random variables,and the information entropy and conditional entropy between sequences were measured to analyze the relational degree between risk factors.Then,a new grey relational analysis model for heterogeneous data was constructed,and a comprehensive security risk factor identification method was developed.A case study of small reservoirs in Guangxi Zhuang Autonomous Region in China shows that the model constructed in this study is applicable to security risk factor identification for small reservoirs with heterogeneous and sparse data.

An improved data space inversion method to predict reservoir state fields via observed production data

A data-space inversion(DSI)method has been recently proposed and successfully applied to the history matching and production prediction of reservoirs.Based on Bayesian theory,DSI can directly and effectively obtain good posterior flow predictions without inversion of geological parameters of reservoir model.This paper presents an improved DSI method to fast predict reservoir state fields(e.g.saturation and pressure profiles)via observed production data.Firstly,a large number of production curves and state data are generated by reservoir model simulation to expand the data space of original DSI.Then,efficient history matching only on the observed production data is carried out via the original DSI to obtain related parameters which reflects the weight of the real reservoir model relative to prior reservoir models.Finally,those parameters are used to predict the oil saturation and pressure profiles of the real reservoir model by combining large amounts of state data of prior reservoir models.Two examples including conventional heterogeneous and unconventional fractured reservoir are implemented to test the performances of predicting saturation and pressure profiles of this improved DSI method.Besides,this method is also tested in a real field and the obtained results show the high computational efficiency and high accuracy of the practical application of this method.

Application of Integrated Seismic Data Processing and Interpretation to Subtle Reservoir Survey

Nowadays, it becomes very urgent to find remain oil under the oil shortage worldwide.However, most of simple reservoirs have been discovered and those undiscovered are mostly complex structural, stratigraphic and lithologic ones. Summarized in this paper is the integrated seismic processing/interpretation technique established on the basis of pre-stack AVO processing and interpretation.Information feedbacks occurred between the pre-stack and post-stack processes so as to improve the accuracy in utilization of data and avoid pitfalls in seismic attributes. Through the integration of seismic data with geologic data, parameters that were most essential to describing hydrocarbon characteristics were determined and comprehensively appraised, and regularities of reservoir generation and distribution were described so as to accurately appraise reservoirs, delineate favorite traps and pinpoint wells.

Spatiotemporal deformation characteristics of Outang landslide and identification of triggering factors using data mining

Since the impoundment of Three Gorges Reservoir(TGR)in 2003,numerous slopes have experienced noticeable movement or destabilization owing to reservoir level changes and seasonal rainfall.One case is the Outang landslide,a large-scale and active landslide,on the south bank of the Yangtze River.The latest monitoring data and site investigations available are analyzed to establish spatial and temporal landslide deformation characteristics.Data mining technology,including the two-step clustering and Apriori algorithm,is then used to identify the dominant triggers of landslide movement.In the data mining process,the two-step clustering method clusters the candidate triggers and displacement rate into several groups,and the Apriori algorithm generates correlation criteria for the cause-and-effect.The analysis considers multiple locations of the landslide and incorporates two types of time scales:longterm deformation on a monthly basis and short-term deformation on a daily basis.This analysis shows that the deformations of the Outang landslide are driven by both rainfall and reservoir water while its deformation varies spatiotemporally mainly due to the difference in local responses to hydrological factors.The data mining results reveal different dominant triggering factors depending on the monitoring frequency:the monthly and bi-monthly cumulative rainfall control the monthly deformation,and the 10-d cumulative rainfall and the 5-d cumulative drop of water level in the reservoir dominate the daily deformation of the landslide.It is concluded that the spatiotemporal deformation pattern and data mining rules associated with precipitation and reservoir water level have the potential to be broadly implemented for improving landslide prevention and control in the dam reservoirs and other landslideprone areas.

Reservoir Multiscale Data Assimilation Using the Ensemble Kalman Filter

In this paper we propose a way to integrate data at different spatial scales using the ensemble Kalman filter (EnKF), such that the finest scale data is sequentially estimated, subject to the available data at the coarse scale (s), as an additional constraint. Relationship between various scales has been modeled via upscaling techniques. The proposed coarse-scale EnKF algorithm is recursive and easily implementable. Our numerical results with the coarse-scale data provide improved fine-scale field estimates when compared to the results with regular EnKF (which did not incorporate the coarse-scale data). We also tested our algorithm with various precisions of the coarse-scale data to account for the inexact relationship between the fine and coarse scale data. As expected, the results show that higher precision in the coarse-scale data, yielded improved estimates.

An Integrated Method of Data Mining and Flow Unit Identification for Typical Low Permeability Reservoir Prediction

With the development of oilfield exploration and mining, the research on continental oil and gas reservoirs has been gradually refined, and the exploration target of offshore reservoir has also entered the hot studystage of small sand bodies, small fault blocks, complex structures, low permeability and various heterogeneous geological bodies. Thus, the marine oil and gas development will inevitably enter thecomplicated reservoir stage;meanwhile the corresponding assessment technologies, engineering measures andexploration method should be designed delicately. Studying on hydraulic flow unit of low permeability reservoir of offshore oilfield has practical significance for connectivity degree and remaining oil distribution. An integrated method which contains the data mining and flow unit identification part was used on the flow unit prediction of low permeability reservoir;the predicted results?were compared with mature commercial system results for verifying its application. This strategy is successfully applied to increase the accuracy by choosing the outstanding prediction result. Excellent computing system could provide more accurate geological information for reservoir characterization.

Meteorological Data Generation for the Numerical Simulation of Stratified Flow in the Joumine Reservoir, Tunisia

Dynamic numerical simulation of water conditions is useful for reservoir management. In remote semi-arid areas, however, meteorological and hydrological time-series data needed for computation are not frequently measured and must be obtained using other information. This paper presents a case study of data generation for the computation of thermal conditions in the Joumine Reservoir, Tunisia. Data from the Wind Finder web site and daily sunshine duration at the nearest weather stations were utilized to generate cloud cover and solar radiation data based on meteorological correlations obtained in Japan, which is located at the same latitude as Tunisia. A time series of inflow water temperature was estimated from air temperature using a numerical filter expressed as a linear second-order differential equation. A numerical simulation using a vertical 2-D （two-dimensional） turbulent flow model for a stratified water body with generated data successfully reproduced seasonal thermal conditions in the reservoir, which were monitored using a thermistor chain.

Turbidite Dynamics and Hydrocarbon Reservoir Formation in the Tano Basin: A Coastal West African Perspective

This study examines the turbidite dynamics and hydrocarbon reservoir formation in Ghana’s Tano Basin, which is located in coastal West Africa. Through an exploration of geological processes spanning millions of years, we uncover key factors shaping hydrocarbon accumulation, including source rock richness, temperature, pressure, and geological structures. The research offers valuable insights applicable to exploration, management, and sustainable resource exploitation in coastal West Africa. It facilitates the identification of exploration targets with higher hydrocarbon potential, enables the anticipation of reservoir potential within the Tano Basin, and assists in tailoring exploration and management strategies to specific geological conditions of the Tano Basin. Analysis of fluvial channels sheds light on their impact on landscape formation and hydrocarbon exploration. The investigation into turbidite systems unveils intricate interactions involving tectonics, sea-level fluctuations, and sedimentation patterns, influencing the development of reservoirs. An understanding of sediment transport and depositional settings is essential for efficient reservoir management. Geomorphological features, such as channels, submarine canyons, and distinct channel types, are essential in this situation. A detailed examination of turbidite channel structures, encompassing canyons, channel complexes, convex channels, and U-shaped channels, provides valuable insights and aids in identifying exploration targets like basal lag, channel levees, and lobes. These findings underscore the enduring significance of turbidite systems as conduits for sediment transport, contributing to enhanced reservoir management and efficient hydrocarbon production. The study also highlights how important it is to examine the configuration of sedimentary layers, stacking patterns, and angular laminated facies to identify turbidites, understand reservoir distribution, and improve well design. The dynamic nature of turbidite systems, influenced by basin characteristics such as shape and slope, is highlighted. The research provides valuable insights essential for successful hydrocarbon exploration, reservoir management, and sustainable resource exploitation in coastal West Africa.

Controlling the uncertainty in reservoir stochastic simulation

Unexpected noise in reservoir stochastic simulation realization may be too high to make the realization useful, especially when there is a lack of hard data. Through discussing the uncertainties, we present two ways to control the uncertainty ratio that is brought by the algorithm of stochastic simulation. By reasonably reducing the random value of the stochastic simulation result, the unexpected values introduced by the residual that associates with random series can be controlled. Another way when the data disperse unevenly is to control the stochastic simulation order by grouping the points that need to be simulated to make those points which can be simulated by more neighborhood hard data calculated first. Both methods do not go against the core stochastic simulation algorithm.

Prediction of Subtle Thin Gas Reservoir in the Loess Desert Area in the North of Ordos Basin

For thin gas reservoir of low-porosity and low-permeability in the loess desert area, a suite of lateral reservoir prediction techniques has been developed by Changqing Oil Company and the excellent effects achieved in exploration and exploitation in the areas such as Yulin, Wushenqi,Suligemiao, Shenmu etc., so that the Upper Paleozoic gas reserve has been stably increasing for eight years in Changqing Oilfield. The paper analyzed the effects and experience of the application of these techniques in detail.

Application of Spectral Decomposition to Detection of Fracture-Cavity Carbonate Reservoir Beds in the Tahe OUfield,Tarim Basin,NW China

Ordovician fracture-cavity carbonate reservoir beds are the major type of producing formations in the Tahe oilfield, Tarim Basin. The seismic responses of these beds clearly changes depending on the different distance of the fracture-cavity reservoir bed from the top of the section. The seismic reflection becomes weak or is absent when the fracture-cavity reservoir beds are less than 20 ms below the top Ordovician. The effect on top Ordovician reflection became weaker with deeper burial of fracture-cavity reservoir beds but the developed deep fracture-cavity reservoir beds caused stronger reflection in the interior of the Ordovician. This interior reflection can be divided into strong long-axis, irregular and bead string reflections, and was present 80 ms below the top Ordovician. Aimed at understanding reflection characteristics, the spectral decomposition technique, which uses frequency to ＂tune-in＂ bed thickness, was used to predict Ordovician fracture-cavity carbonate formations in the Tahe oilfield. Through finely adjusting the processing parameters of spectral decomposition, it was found that the slice at 30 Hz of the tuned data cube can best represent reservoir bed development. Two large N-S-trending strong reflection belts in the mid-western part of the study area along wells TK440- TK427-TK417B and in the eastern part along wells TK404-TK409 were observed distinctly on the 30 Hz slice and 4-D time-frequency data cube carving. A small N-S trending reflection belt in the southern part along wells T403-TK446B was also clearly identified. The predicted reservoir bed development area coincides with the fracture-cavities connection area confirmed by drilling pressure testing results. Deep karst cavities occur basically in three reservoir bed-development belts identified by the Ordovician interior strong reflection. Spectral decomposition proved to be a useful technique in identifying fracture-cavity reservoir beds.

On the Use of Landsat-5 TM Satellite for Assimilating Water Temperature Observations in 3D Hydrodynamic Model of Small Inland Reservoir in Midwestern US

Accuracy of hydrodynamic and water quality numerical models developed for a specific site is dependent on multiple model parameters and variables whose values are attained via calibration processes and/or expert knowledge. Real time variations in the actual aquatic system at a site necessitate continuous monitoring of the system so that model parameters and variables are regularly updated to reflect accurate conditions. Multiple sources of observations can help adjust the model better by providing benefits of individual monitoring technology within the model updating process. For example, remote sensing data provide a spatially dense dataset of model variables at the surface of a water body, while in-situ monitoring technologies can provide data at multiple depths and at more frequent time intervals than remote sensing technologies. This research aims to present an overview of an integrated modeling and data assimilation framework that combines three-dimensional numerical model with multiple sources of observations to simulate water column temperature in a eutrophic reservoir in central Indiana. A variational data assimilation approach is investigated for incorporating spatially continuous remote sensing temperature observations and spatially discrete in-situ observations to change initial conditions of the numerical model. The results demonstrate the challenges in improving the model performance by incorporating water temperature from multi-spectral remote sensing analysis versus in-situ measurements. For example, at a eutrophic reservoir in Central Indiana where four images of multi-spectral remote sensing data were assimilated in the numerical model, the overall error for the four images reduced from 20.9% (before assimilation) to 15.9% (best alternative after the assimilation). Additionally, best improvements in errors were observed on days closer to the starting time of model’s assimilation time window. However, when the original and updated model results for the water column temperature were compared to the in-situ measurements during the data assimilation period, the error was found to have actually increased from 1.8℃ (before assimilation) to 2.7℃ (after assimilation). Sampling depth differences between remote sensing observations and in-situ measurements, and spatial and temporal sampling of remote sensing observations are considered as possible reasons for this contrary behavior in model performance. The authors recommend that additional research is needed to further examine this behavior.

An improved oil recovery prediction method for volatile oil reservoirs

To describe the complex phase transformation in the process of depletion exploitation of volatile oil reservoir,four fluid phases are defined,and production and remaining volume of these phases are calculated based on the principle of surface volume balance,then the recovery prediction method of volatile oil reservoir considering the influence of condensate content in released solution gas and the correction method of multiple degassing experiments data are established.Taking three typical kinds of crude oil(black oil,medium-weak volatile oil,strong volatile oil)as examples,the new improved method is used to simulate constant volume depletion experiments based on the corrected data of multiple degassing experiment to verify the reliability of the modified method.By using"experimental data and traditional method","corrected data and traditional method"and"corrected data and modified method",recovery factors of these three typical kinds of oil are calculated respectively.The source of parameters and the calculation methods have little effect on the recovery of typical black oil.However,with the increase of crude oil volatility,the oil recovery will be seriously underestimated by using experimental data or traditional method.The combination of"corrected data and modified method"considers the influence of condensate in gas phase in both experimental parameters and calculation method,and has good applicability to typical black oil and volatile oil.The strong shrinkage of volatile oil makes more"liquid oil"convert to"gaseous oil",so volatile oil reservoir can reach very high oil recovery by depletion drive.

Detection of gas reservoirs by the joint use of P- and PS-waves: A case study on the Ordos basin, China

We present an example of using converted-waves for characterizing onshore gas reservoirs in the Ordos basin in Northwest China. The Ordos basin is the largest gas province in China. The main gas reservoirs （about 3 300 m in depth） are in upper Paleozoic sandstone that has low or reversed P-wave impedance and is immediately above a coal seam. This makes it very difficult to image the gas reservoirs using conventional P-wave data. Analysis of core, log and VSP data shows a weak PP reflection but a relatively strong PS-converted wave reflection, or both strong PP- and PS-reflections but with opposite polarity from the gas bearing sands, which indicates the potential of using PS-waves to image the gas reservoirs in the Ordos basin. Subsequently, thirteen seismic lines were acquired, processed and interpreted to verify the PP- and PS-responses, and two corresponding attributes （PP- and PS- amplitude ratio and polarity ratio） are used to map the reservoirs through joint PP and PS analysis.

Logging Data High-Resolution Sequence Stratigraphy

The recognition and contrast of bed sets in parasequence is difficult in terrestrial basin high-resolution sequence stratigraphy. This study puts forward new methods for the boundary identification and contrast of bed sets on the basis of manifold logging data. The formation of calcareous interbeds, shale resistivity differences and the relation of reservoir resistivity to altitude are considered on the basis of log curve morphological characteristics, core observation, cast thin section, X-ray diffraction and scanning electron microscopy. The results show that the thickness of calcareous interbeds is between 0.5 m and 2 m, increasing on weathering crusts and faults. Calcareous interbeds occur at the bottom of a distributary channel and the top of a distributary mouth bar. Lower resistivity shale （4-5 Ω · m） and higher resistivity shale （〉 10Ω·m） reflect differences in sediment fountain or sediment microfacies. Reservoir resistivity increases with altitude. Calcareous interbeds may be a symbol of recognition for the boundary of bed sets and isochronous contrast bed sets, and shale resistivity differences may confirm the stack relation and connectivity of bed sets. Based on this, a high-resolution chronostratigraphic frame- work of Xi-1 segment in Shinan area, Junggar basin is presented, and the connectivity of bed sets and oil-water contact is confirmed. In this chronostratigraphic framework, the growth order, stack mode and space shape of bed sets are qualitatively and quantitatively described.

Seismic Data Processing and Interpretation on the Loess Plateau, Part 2: Seismic Data Interpretation

The loess plateau covering the North Shaanxi slope and Tianhuan depression consists of a regional monocline, high in the east and low in the west, with dips of less than 1^0, Structural movement in this region was weak so that faults and local structures were not well developed. As a result, numerous wide and gentle noses and small traps with magnitudes less than 50 m were developed on the large westward-dipping monocline. Reservoirs, including Mesozoic oil reservoirs and Paleozoic gas reservoirs in the Ordos Basin, are dominantly lithologic with a small number of structural reservoirs. Single reservoirs are characterized as thin with large lateral variations, strong anisotropy, low porosity, low permeability, and low richness. A series of approaches for predicting reservoir thickness, physical properties, and hydrocarbon potential of subtle lithologic reservoirs was established based on the interpretation of erosion surfaces.