scholarly journals Unsupervised machine learning for time-lapse seismic studies and reservoir monitoring

2021 ◽  
pp. 1-59
Author(s):  
Marwa Hussein ◽  
Robert R. Stewart ◽  
Deborah Sacrey ◽  
David H. Johnston ◽  
Jonny Wu
Keyword(s):  
Machine Learning ◽  
Reservoir Simulation ◽  
Seismic Data ◽  
Water Saturation ◽  
Seismic Analysis ◽  
Time Lapse ◽  
Seismic Attributes ◽  
Rock Properties ◽  
Unsupervised Machine Learning ◽  
4D Seismic

Time-lapse (4D) seismic analysis plays a vital role in reservoir management and reservoir simulation model updates. However, 4D seismic data are subject to interference and tuning effects. Being able to resolve and monitor thin reservoirs of different quality can aid in optimizing infill drilling or locating bypassed hydrocarbons. Using 4D seismic data from the Maui field in the offshore Taranaki basin of New Zealand, we generate typical seismic attributes sensitive to reservoir thickness and rock properties. We find that spectral instantaneous attributes extracted from time-lapse seismic data illuminate more detailed reservoir features compared to those same attributes computed on broadband seismic data. We develop an unsupervised machine learning workflow that enables us to combine eight spectral instantaneous seismic attributes into single classification volumes for the baseline and monitor surveys using self-organizing maps (SOM). Changes in the SOM natural clusters between the baseline and monitor surveys suggest production-related changes that are caused primarily by water replacing gas as the reservoir is being swept under a strong water drive. The classification volumes also facilitate monitoring water saturation changes within thin reservoirs (ranging from very good to poor quality) as well as illuminating thin baffles. Thus, these SOM classification volumes show internal reservoir heterogeneity that can be incorporated into reservoir simulation models. Using meaningful SOM clusters, geobodies are generated for the baseline and monitor SOM classifications. The recoverable gas reserves for those geobodies are then computed and compared to production data. The SOM classifications of the Maui 4D seismic data seems to be sensitive to water saturation change and subtle pressure depletions due to gas production under a strong water drive.

Unsupervised machine learning using 3D seismic data applied to reservoir evaluation and rock type identification

2021 ◽  
pp. 1-69
Author(s):  
Marwa Hussein ◽  
Robert R. Stewart ◽  
Deborah Sacrey ◽  
Jonny Wu ◽  
Rajas Athale
Keyword(s):  
Machine Learning ◽  
Seismic Data ◽  
Rock Type ◽  
Seismic Attributes ◽  
Reservoir Quality ◽  
3D Seismic ◽  
Reservoir Properties ◽  
Unsupervised Machine Learning ◽  
Amplitude Data ◽  
3D Seismic Data

Net reservoir discrimination and rock type identification play vital roles in determining reservoir quality, distribution, and identification of stratigraphic baffles for optimizing drilling plans and economic petroleum recovery. Although it is challenging to discriminate small changes in reservoir properties or identify thin stratigraphic barriers below seismic resolution from conventional seismic amplitude data, we have found that seismic attributes aid in defining the reservoir architecture, properties, and stratigraphic baffles. However, analyzing numerous individual attributes is a time-consuming process and may have limitations for revealing small petrophysical changes within a reservoir. Using the Maui 3D seismic data acquired in offshore Taranaki Basin, New Zealand, we generate typical instantaneous and spectral decomposition seismic attributes that are sensitive to lithologic variations and changes in reservoir properties. Using the most common petrophysical and rock typing classification methods, the rock quality and heterogeneity of the C1 Sand reservoir are studied for four wells located within the 3D seismic volume. We find that integrating the geologic content of a combination of eight spectral instantaneous attribute volumes using an unsupervised machine-learning algorithm (self-organizing maps [SOMs]) results in a classification volume that can highlight reservoir distribution and identify stratigraphic baffles by correlating the SOM clusters with discrete net reservoir and flow-unit logs. We find that SOM classification of natural clusters of multiattribute samples in the attribute space is sensitive to subtle changes within the reservoir’s petrophysical properties. We find that SOM clusters appear to be more sensitive to porosity variations compared with lithologic changes within the reservoir. Thus, this method helps us to understand reservoir quality and heterogeneity in addition to illuminating thin reservoirs and stratigraphic baffles.

Reservoir simulation in time‐lapse (4D) seismic analysis

1997 ◽  
Author(s):  
Paul J. Hicks ◽  
Zhiyuan Cai
Keyword(s):  
Reservoir Simulation ◽  
Seismic Analysis ◽  
Time Lapse ◽  
4D Seismic

The combined effects of pressure and cementation on 4D seismic data

Geophysics ◽  
2015 ◽  
Vol 80 (2) ◽  
pp. WA135-WA148 ◽  
Author(s):  
Matthew Saul ◽  
David Lumley
Keyword(s):  
Pore Pressure ◽  
Seismic Data ◽  
Time Pressure ◽  
Time Lapse ◽  
Reservoir Rock ◽  
Elastic Theory ◽  
Rock Properties ◽  
Baseline Survey ◽  
Core Data ◽  
4D Seismic

Time-lapse seismology has proven to be a useful method for monitoring reservoir fluid flow, identifying unproduced hydrocarbons and injected fluids, and improving overall reservoir management decisions. The large magnitudes of observed time-lapse seismic anomalies associated with strong pore pressure increases are sometimes not explainable by velocity-pressure relationships determined by fitting elastic theory to core data. This can lead to difficulties in interpreting time-lapse seismic data in terms of physically realizable changes in reservoir properties during injection. It is commonly assumed that certain geologic properties remain constant during fluid production/injection, including rock porosity and grain cementation. We have developed a new nonelastic method based on rock physics diagnostics to describe the pressure sensitivity of rock properties that includes changes in the grain contact cement, and we applied the method to a 4D seismic data example from offshore Australia. We found that water injection at high pore pressure may mechanically weaken the poorly consolidated reservoir sands in a nonelastic manner, allowing us to explain observed 4D seismic signals that are larger than can be predicted by elastic theory fits to the core data. A comparison of our new model with the observed 4D seismic response around a large water injector suggested a significant mechanical weakening of the reservoir rock, consistent with a decrease in the effective grain contact cement from 2.5% at the time/pressure of the preinjection baseline survey, to 0.75% at the time/pressure of the monitor survey. This approach may enable more accurate interpretations and future predictions of the 4D signal for subsequent monitor surveys and improve 4D feasibility and interpretation studies in other reservoirs with geomechanically similar rocks.

Machine learning to reduce cycle time for time-lapse seismic data assimilation into reservoir management

2019 ◽  
Vol 7 (3) ◽  
pp. SE123-SE130
Author(s):  
Yang Xue ◽  
Mariela Araujo ◽  
Jorge Lopez ◽  
Kanglin Wang ◽  
Gautam Kumar
Keyword(s):  
Machine Learning ◽  
Data Assimilation ◽  
Seismic Data ◽  
Reservoir Management ◽  
Time Lapse ◽  
Water Flooding ◽  
Reservoir Models ◽  
Reservoir Property ◽  
Property Changes ◽  
4D Seismic

Time-lapse (4D) seismic is widely deployed in offshore operations to monitor improved oil recovery methods including water flooding, yet its value for enhanced well and reservoir management is not fully realized due to the long cycle times required for quantitative 4D seismic data assimilation into dynamic reservoir models. To shorten the cycle, we have designed a simple inversion workflow to estimate reservoir property changes directly from 4D attribute maps using machine-learning (ML) methods. We generated tens of thousands of training samples by Monte Carlo sampling from the rock-physics model within reasonable ranges of the relevant parameters. Then, we applied ML methods to build the relationship between the reservoir property changes and the 4D attributes, and we used the learnings to estimate the reservoir property changes given the 4D attribute maps. The estimated reservoir property changes (e.g., water saturation changes) can be used to analyze injection efficiency, update dynamic reservoir models, and support reservoir management decisions. We can reduce the turnaround time from months to days, allowing early engagements with reservoir engineers to enhance integration. This accelerated data assimilation removes a deterrent for the acquisition of frequent 4D surveys.

scholarly journals Time-Lapse Seismic Monitoring of Onshore Reservoirs in Niger Delta, Field ‘K’ as a Case Study

2017 ◽  
Vol 1 (1) ◽  
pp. 23-27 ◽  
Author(s):  
A. Ogbamikhumi ◽  
T. Tralagba ◽  
E. E. Osagiede
Keyword(s):  
Seismic Data ◽  
Acoustic Impedance ◽  
Niger Delta ◽  
Rock Physics ◽  
Time Lapse ◽  
Seismic Survey ◽  
Impedance Change ◽  
Data Set ◽  
Reservoir Fluid ◽  
4D Seismic

Field ‘K’ is a mature field in the coastal swamp onshore Niger delta, which has been producing since 1960. As a huge producing field with some potential for further sustainable production, field monitoring is therefore important in the identification of areas of unproduced hydrocarbon. This can be achieved by comparing production data with the corresponding changes in acoustic impedance observed in the maps generated from base survey (initial 3D seismic) and monitor seismic survey (4D seismic) across the field. This will enable the 4D seismic data set to be used for mapping reservoir details such as advancing water front and un-swept zones. The availability of good quality onshore time-lapse seismic data for Field ‘K’ acquired in 1987 and 2002 provided the opportunity to evaluate the effect of changes in reservoir fluid saturations on time-lapse amplitudes. Rock physics modelling and fluid substitution studies on well logs were carried out, and acoustic impedance change in the reservoir was estimated to be in the range of 0.25% to about 8%. Changes in reservoir fluid saturations were confirmed with time-lapse amplitudes within the crest area of the reservoir structure where reservoir porosity is 0.25%. In this paper, we demonstrated the use of repeat Seismic to delineate swept zones and areas hit with water override in a producing onshore reservoir.

Faster Convergence in Seismic History Matching by Dividing and Conquering the Unknowns

SPE Journal ◽  
2010 ◽  
Vol 15 (04) ◽  
pp. 1077-1088 ◽  
Author(s):  
F.. Sedighi ◽  
K.D.. D. Stephen
Keyword(s):  
Regression Analysis ◽  
Seismic Data ◽  
History Matching ◽  
Time Lapse ◽  
Business Decisions ◽  
Order Of Magnitude ◽  
Reservoir Simulation Model ◽  
The Uk ◽  
Pressure Changes ◽  
4D Seismic

Summary Seismic history matching is the process of modifying a reservoir simulation model to reproduce the observed production data in addition to information gained through time-lapse (4D) seismic data. The search for good predictions requires that many models be generated, particularly if there is an interaction between the properties that we change and their effect on the misfit to observed data. In this paper, we introduce a method of improving search efficiency by estimating such interactions and partitioning the set of unknowns into noninteracting subspaces. We use regression analysis to identify the subspaces, which are then searched separately but simultaneously with an adapted version of the quasiglobal stochastic neighborhood algorithm. We have applied this approach to the Schiehallion field, located on the UK continental shelf. The field model, supplied by the operator, contains a large number of barriers that affect flow at different times during production, and their transmissibilities are highly uncertain. We find that we can successfully represent the misfit function as a second-order polynomial dependent on changes in barrier transmissibility. First, this enables us to identify the most important barriers, and, second, we can modify their transmissibilities efficiently by searching subgroups of the parameter space. Once the regression analysis has been performed, we reduce the number of models required to find a good match by an order of magnitude. By using 4D seismic data to condition saturation and pressure changes in history matching effectively, we have gained a greater insight into reservoir behavior and have been able to predict flow more accurately with an efficient inversion tool. We can now determine unswept areas and make better business decisions.

Fluid dynamics at the base of hydrate-bearing sediments at the Vestnesa Ridge inferred from 5 years of high-resolution 4D seismic surveying

2020 ◽  
Author(s):  
Malin Waage ◽  
Stefan Bünz ◽  
Kate Waghorn ◽  
Sunny Singhorha ◽  
Pavel Serov
Keyword(s):  
High Resolution ◽  
Gas Hydrate ◽  
Seismic Data ◽  
Time Lapse ◽  
Stability Zone ◽  
Seismic Surveys ◽  
Free Gas ◽  
4D Seismic ◽  
Hydrate Bearing Sediments ◽  
Hydrate Stability

<p>The transition from gas hydrate to gas-bearing sediments at the base of the hydrate stability zone (BHSZ) is commonly identified on seismic data as a bottom-simulating reflection (BSR). At this boundary, phase transitions driven by thermal effects, pressure alternations, and gas and water flux exist. Sedimentation, erosion, subsidence, uplift, variations in bottom water temperature or heat flow cause changes in marine gas hydrate stability leading to expansion or reduction of gas hydrate accumulations and associated free gas accumulations. Pressure build-up in gas accumulations trapped beneath the hydrate layer may eventually lead to fracturing of hydrate-bearing sediments that enables advection of fluids into the hydrate layer and potentially seabed seepage. Depletion of gas along zones of weakness creates hydraulic gradients in the free gas zone where gas is forced to migrate along the lower hydrate boundary towards these weakness zones. However, due to lack of “real time” data, the magnitude and timescales of processes at the gas hydrate – gas contact zone remains largely unknown. Here we show results of high resolution 4D seismic surveys at a prominent Arctic gas hydrate accumulation – Vestnesa ridge - capturing dynamics of the gas hydrate and free gas accumulations over 5 years. The 4D time-lapse seismic method has the potential to identify and monitor fluid movement in the subsurface over certain time intervals. Although conventional 4D seismic has a long history of application to monitor fluid changes in petroleum reservoirs, high-resolution seismic data (20-300 Hz) as a tool for 4D fluid monitoring of natural geological processes has been recently identified.<br><br>Our 4D data set consists of four high-resolution P-Cable 3D seismic surveys acquired between 2012 and 2017 in the eastern segment of Vestnesa Ridge. Vestnesa Ridge has an active fluid and gas hydrate system in a contourite drift setting near the Knipovich Ridge offshore W-Svalbard. Large gas flares, ~800 m tall rise from seafloor pockmarks (~700 m diameter) at the ridge axis. Beneath the pockmarks, gas chimneys pierce the hydrate stability zone, and a strong, widespread BSR occurs at depth of 160-180 m bsf. 4D seismic datasets reveal changes in subsurface fluid distribution near the BHSZ on Vestnesa Ridge. In particular, the amplitude along the BSR reflection appears to change across surveys. Disappearance of bright reflections suggest that gas-rich fluids have escaped the free gas zone and possibly migrated into the hydrate stability zone and contributed to a gas hydrate accumulation, or alternatively, migrated laterally along the BSR. Appearance of bright reflection might also indicate lateral migration, ongoing microbial or thermogenic gas supply or be related to other phase transitions. We document that faults, chimneys and lithology constrain these anomalies imposing yet another control on vertical and lateral gas migration and accumulation. These time-lapse differences suggest that (1) we can resolve fluid changes on a year-year timescale in this natural seepage system using high-resolution P-Cable data and (2) that fluids accumulate at, migrate to and migrate from the BHSZ over the same time scale.</p>

Time-lapse seismic analysis of overburden water injection at the Ekofisk field, southern North Sea

Geophysics ◽  
2020 ◽  
Vol 85 (1) ◽  
pp. B9-B21
Author(s):  
Filipe Borges ◽  
Martin Landrø ◽  
Kenneth Duffaut
Keyword(s):  
North Sea ◽  
Seismic Data ◽  
Seismic Event ◽  
Seismic Analysis ◽  
Water Injection ◽  
Time Lapse ◽  
Reservoir Compaction ◽  
Southern North Sea ◽  
Stress Changes ◽  
Before And After

On 7 May 2001, a seismic event occurred in the southern North Sea in the vicinity of the Ekofisk platform area. Analysis of seismological recordings of this event indicated that the epicenter is likely within the northern part of the field and its hypocenter lies in the shallow sedimentary layer. Further investigation in this same area revealed a small seabed uplift and identified an unintentional water injection in the overburden. The injection presumably caused the seabed uplift in addition to stress changes in the overburden. To better understand the consequences of this water injection, we analyze marine seismic data acquired before and after the seismological event. The 4D analysis reveals a clear traveltime shift close to the injection well, as well as a weak amplitude difference. We find that these measured time shifts correspond reasonably well with modeled time shifts based on a simple geomechanical model. The modeling also correlates well with the observed bathymetry changes at the seabed and with global positioning system measurements at the platforms. Although no explicit amplitude sign of the seismic event could be detected in the seismic data, the modeled stress changes, combined with the effect of decades of production-induced reservoir compaction, suggest a source mechanism for the far-field seismological recordings of the May 7th event.

Sign in / Sign up

Export Citation Format

Share Document