Auto-windowed Super-virtual Interferometry via Machine Learning: A Strategy of First-arrival Traveltime Automatic Picking for Noisy Seismic Data

Abstract See how application of a fully trained Artificial Intelligence (AI) / Machine Learning (ML) technology applied to 3D seismic data volumes delivers an unbiased data driven assessment of entire volumes or corporate seismic data libraries quickly. Whether the analysis is undertaken using onsite hardware or a cloud based mega cluster, this automated approach provides unparalleled insights for the interpretation and prospectivity analysis of any dataset. The Artificial Intelligence (AI) / Machine Learning (ML) technology uses unsupervised genetics algorithms to create families of waveforms, called GeoPopulations, that are used to derive Amplitude, Structure (time or depth depending on the input 3D seismic volume) and the new seismic Fitness attribute. We will show how Fitness is used to interpret paleo geomorphology and facies maps for every peak, trough and zero crossing of the 3D seismic volume. Using the Structure, Amplitude and Fitness attribute maps created for every peak, trough and zero crossing the Exploration and Production (E&P) team can evaluate and mitigate Geological and Geophysical (G&G) risks and uncertainty associated with their petroleum systems quickly using the entire 3D seismic data volume.

Download Full-text

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

Interpretation ◽

10.1190/int-2020-0108.1 ◽

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.

Download Full-text

Deep Reinforcement and Machine Learning for Seismic Data Processing and Automated QC

10.3997/2214-4609.202112542 ◽

2021 ◽

Author(s):

J. Brittan ◽

R. O’Driscoll ◽

J. Walpole

Keyword(s):

Machine Learning ◽

Data Processing ◽

Seismic Data ◽

Seismic Data Processing

Download Full-text

Machine Learning Assisted Interpretation: Integrated Fault Prediction Extraction Case Study from the Groningen Gas Field, Netherlands

Interpretation ◽

10.1190/int-2021-0137.1 ◽

2021 ◽

pp. 1-67

Author(s):

Stewart Smith ◽

Olesya Zimina ◽

Surender Manral ◽

Michael Nickel

Keyword(s):

Machine Learning ◽

Fault Detection ◽

Seismic Data ◽

Fault Prediction ◽

Machine Learning Techniques ◽

Gas Field ◽

Seismic Fault ◽

Learning Techniques ◽

Groningen Gas Field

Seismic fault detection using machine learning techniques, in particular the convolution neural network (CNN), is becoming a widely accepted practice in the field of seismic interpretation. Machine learning algorithms are trained to mimic the capabilities of an experienced interpreter by recognizing patterns within seismic data and classifying them. Regardless of the method of seismic fault detection, interpretation or extraction of 3D fault representations from edge evidence or fault probability volumes is routine. Extracted fault representations are important to the understanding of the subsurface geology and are a critical input to upstream workflows including structural framework definition, static reservoir and petroleum system modeling, and well planning and de-risking activities. Efforts to automate the detection and extraction of geological features from seismic data have evolved in line with advances in computer algorithms, hardware, and machine learning techniques. We have developed an assisted fault interpretation workflow for seismic fault detection and extraction, demonstrated through a case study from the Groningen gas field of the Upper Permian, Dutch Rotliegend; a heavily faulted, subsalt gas field located onshore, NE Netherlands. Supervised using interpreter-led labeling, we apply a 2D multi-CNN to detect faults within a 3D pre-stack depth migrated seismic dataset. After prediction, we apply a geometric evaluation of predicted faults, using a principal component analysis (PCA) to produce geometric attribute representations (strike azimuth and planarity) of the fault prediction. Strike azimuth and planarity attributes are used to validate and automatically extract consistent 3D fault geometries, providing geological context to the interpreter and input to dependent workflows more efficiently.

Download Full-text

Fluid Prediction on Seismic Data Using Different Machine Learning Methods

10.3997/2214-4609.202150051 ◽

2021 ◽

Author(s):

V Kalashnikova ◽

E Karaseva ◽

R Overas ◽

T.R Sharafutdinov

Keyword(s):

Machine Learning ◽

Seismic Data ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Machine Learning analysis of seismic signals recorded at Stromboli Volcano

10.5194/egusphere-egu21-13166 ◽

2021 ◽

Author(s):

Darius Fenner ◽

Georg Rümpker ◽

Horst Stöcker ◽

Megha Chakraborty ◽

Wei Li ◽

...

Keyword(s):

Machine Learning ◽

Seismic Data ◽

Noise Level ◽

Volcanic Eruptions ◽

Time Varying ◽

Seismic Signals ◽

Machine Learning Model ◽

Automatic Adaptation ◽

Consistent Manner ◽

Recent Developments

At Stromboli, minor volcanic eruptions occur at time intervals of approximately five minutes on average, making it one of the most active volcanoes worldwide. In addition to these mostly harmless events, there are also stronger eruptions and paroxysms which pose a serious threat to residents and tourists. In light of recent developments in Machine Learning, this study attempts to apply these new tools for the analysis of the time-varying volcanic eruptions at Stromboli. As input for the Machine-Learning approach, we use continuous recordings of seismic signals from two seismometers on the island. The data is available from IRIS &#160;and includes records starting in 2012 up to the present.&#160;One primary challenge is to label and classify the data, i.e., to discriminate events of interest from noise. The variety of signal-appearance in the recorded data is wide, in some periods the events are clearly distinguishable from noise whereas, in other cases relevant events are obscured by the high noise level. To enable the event-detection in all cases, we developed the following algorithm: in the first step, the seismic data is pre-processed with an STA/LTA-Filter, which allows detection of events based on a prominence threshold. However, due to the diversity of signal patterns, a fixed set of hyperparameters (STA- and LTA-window length, prominence threshold, correlation coefficient) fails to reliably extract the relevant events in a consistent manner. Therefore, the (time-varying) noise level of the recordings is used as an additional key indicator. After this, the hyperparameters are optimized. The automatic adaptation is then used for labeling the continuous seismic data.After extracting the events based on this approach, a machine learning model is trained to analyze the recordings for possible patterns in the interval times and the event amplitudes. This study is expected to provide constraints on the possibility to detect complex time-dependent patterns of the eruption history at Stromboli.

Download Full-text

Unravel Carbonate Reservoir Heterogeneities Through 3D Seismic Neural Network Application: A Machine Learning Based Approach and Workflow

10.2118/205664-ms ◽

2021 ◽

Author(s):

Dimmas Ramadhan ◽

Krishna Pratama Laya ◽

Ricko Rizkiaputra ◽

Esterlinda Sinlae ◽

Ari Subekti ◽

...

Keyword(s):

Neural Network ◽

Machine Learning ◽

Seismic Data ◽

Carbonate Reservoir ◽

3D Seismic ◽

Carbonate Facies ◽

Full Field ◽

Facies Classification ◽

Unsupervised Neural Network ◽

Tight Gas Reservoir

Abstract The availability of 3D seismic data undoubtedly plays an important role in reservoir characterization. Currently seismic technology continues to advance at a rapid pace not only in the acquisition but also in processing and interpretation domain. The advance on this is well supported by the digitalization era which urges everything to run reliably fast, effective and efficient. Thanks to continuous development of IT peripherals we now have luxury to process and handle big data through the application of machine learning. Some debates on the effectiveness and threats that this process may automating certain task and later will decrease human workforce are still going on in many forums but still like it or not this machine learning is already embraced in almost every aspect of our life including in oil & gas industry. Carbonate reservoir on the other hand has been long known for its uniqueness compared to siliciclastic reservoir. The term heterogeneous properties are quite common for carbonate due to its complex multi-story depositional and diagenetic facies. In this paper, we bring up our case where we try to unravel carbonate heterogeneity from a massive tight gas reservoir through our machine learning application using the workflow of supervised and unsupervised neural network. In this study, we incorporate 3D PSTM seismic data and its stratigraphic interpretation coupled with the core study result, BHI (borehole image) log interpretation, and our regional understanding of the area to develop a meaningful carbonate facies model through seismic neural network exercises. As the result, we successfully derive geological consistent carbonate facies classification and distribution honoring all the supporting data above though the limitation of well penetration in the area. This result then proved to be beneficial to build integrated 3D geomodel which later can explain the issue on different gas compositions happens in the area. The result on unsupervised neural network also able to serves as a quick look for further sweetspot analysis to support full-field development.

Download Full-text

An automatic S-phase picker

Bulletin of the Seismological Society of America ◽

10.1785/bssa0830010180 ◽

1993 ◽

Vol 83 (1) ◽

pp. 180-189

Author(s):

Artur Cichowicz

Keyword(s):

Total Energy ◽

Seismic Data ◽

Transverse Energy ◽

Deflection Angle ◽

S Phase ◽

Underground Mine ◽

Degree Of Polarization ◽

Reference Level ◽

Automatic Picking ◽

Analyze Data

Abstract An algorithm has been developed for the automatic picking of the S phase from three-component seismic data. Three parameters of the signal are calculated in the program: deflection angle, degree of polarization, and the ratio between transverse energy and total energy. The S phase is declared when the product of the three parameters increases above the reference level. Most parameters are computed automatically and modified if necessary. The S-phase picker is used to analyze data from a local underground mine seismic network.

Download Full-text