Paradise: Machine Learning Technology
The Next Generation in Seismic Interpretation
Paradise is a dramatic new geoscience analysis software featuring machine learning technology for oil and gas exploration and development. The platform enables geoscientists to rapidly analyze multiple attributes and other information simultaneously, generating these sources of value:
Identify thin beds
See detailed stratigraphy
Reveal Direct Hydrocarbon Indicators (DHIs)
Analyze reservoir properties
Expose faults and fracture trends
Isolate pressure-related facies
Geobodies is a dramatic new machine learning tool within Paradise that enables geoscientists to isolate and investigate areas of interest in the SOM results
Display 2D and 3D views of the data while using the 2D Colormap to gain an understanding of the classification results
Organize data and results into user-defined areas while facilitating collaboration among asset teams globally
The new Paradise HPC capability enables the generation of geometric and spectral decomposition attributes on clusters of machines, dramatically reducing computation time
Explore classification results in the Universal Viewer by highlighting the corresponding neurons in the 2D Colormap
Select specific attributes for elimination of dead traces in all attributes used for processing
Seismic attributes provide a framework for interpreting geologic features that define depositional environments. The Paradise Attribute Generator guides interpreters through the process in straightforward ThoughtFlows™, making a large library of attributes readily accessible. Using the Advanced Attribute Library, interpreters can systematically generate seismic attribute volumes for multi-attribute analysis. Employ an array of instantaneous, geometric, and spectral decomposition attributes to extract more from the seismic response.
The Advanced Attribute library is based on software developed by the world-class Attribute Assisted Seismic Processing & Interpretation (AASPI) Consortium at The University of Oklahoma. The Paradise Attribute Generator presents this powerful software in straightforward ThoughtFlows, making them available for all interpreters who use Paradise for attribute generation and analysis.
Families of Attributes
• Over 100 Attributes
• Structural Decomposition
Principal Component Analysis (PCA)
Attributes differ in their relative contribution to information in a given volume. PCA is a method of identifying those attributes that have the greatest contribution to the data and quantifying the relative contribution of each attribute. The PCA workflow in Paradise is applied on the attributes to assess their relative contribution to the SOM analysis.
Self Organizing Maps (SOMs)
The SOM is a neural network based machine learning process that is applied to multiple attribute volumes simultaneously. Applied at full seismic resolution in Paradise, the SOM produces a non-linear classification of the data in a region designated by the interpreter. Regions can be delimited by time, between horizons, or above and below a given horizon in Paradise. The interpreter uses the 2D Colormap to analyze SOM results. Each hexagonal object (neuron) in the interactive 2D Colormap represents a class of data in the region and a corresponding geologic condition. The SOM distills greater information from the seismic response by extracting more information from multiple attributes.
A SOM analysis with a corresponding 8x8 neural hexagonal 2D color bar that are interactive with the Universal Viewer
Interactive 2D Colormap
Paradise is the only software product in the industry with a 2D Colormap representing neuron classifications of attributes. The 2D Colormap is used interactively with the Paradise Universal Viewer to select and isolate specific neurons which have classified a set of seismic attributes according to where data is concentrated. The Paradise 2D Colormap focuses on specific anomalies, discovering geologic features, stratigraphic features, high pore pressure, or direct hydrocarbons indicators (DHI’s).
Paradise enables every interpreter to use powerful machine learning processes through straight forward, left-to-right guided ThoughtFlowsTM. Click below to learn more about how to set up and generate a PCA chart of attributes and SOM classification results, then use the unique 2D Colormap with the 3D Viewer to interpret geobodies.
The Universal Viewer presents your SOM process results through classification and probability volumes. Using the unique interactive 2D Colormap, geoscientists can selectively isolate, visualize, and analyze each individual neuron.
Specific features and anomalies can be investigated according to classification and probability results. The interactive capability of the Universal Viewer and interactive 2D Colormap enables a faster and higher fidelity through a 360 degree view of the volume, thereby reducing exploration risk. The Universal Viewer also offers a variety of data views,
- Chair cuts
- Horizon ghosting
- Volume-oriented data views
- Well paths
The selected data objects in the Universal Viewer windows do not need to be reselected as the interpreter navigates between Universal Viewer, 2D Viewer, and the World Map. Data is selected for viewers from a single workspace navigation tree.
Geobody generation and editing is an exciting part of the Paradise 3.2 release and unique to the Paradise machine learning platform. Geobodies are created from neuron groups selected from a Self Organizing Map (SOM), and are used to generate a three-dimensional visualization of a geological feature, such as channels, sand bars, deltaic formations, or any other area of interest. The number of Geobodies generated depend upon the SOM topology and the number of neurons originally selected.
Once Geobodies are generated, they can be used to help calculate the volumetrics of a geological feature; and therefore, the amount of recoverable oil or gas by using industry-standard calculations. Using Paradise to identify geobodies empowers interpreters with the ability to:
- Investigate geobodies at the sample level of each neuron
- Capture details on areas of interest, including volumetrics and statistics
- Edit/clean-up selected geobodies by filling in areas or pruning extraneous samples
- Export areas of interest to an interpretation system
Real-time interactive attribute analysis allows interpreters to idenitify the attribute contributions per neuron
Enhanced Data Management
Dashboards & Sessions
The dynamically updated Dashboard provides a convenient snapshot of work performed on specific portfolios, along with the corresponding SOM and PCA results, in an easy-to-view tabular form. Select an item to see the underlying details or return to a specific session to continue an analysis.
Paradise is a global analysis platform that enables interpreters to declare specific Workspaces within a regional setting. Workspaces are defined by seismic data, attribute volumes, and analysis results within a polygon on a map of any shape and size. The powerful combination of the World Map and user-identified Workspaces supports collaboration among asset teams and JV partners alike. Results within defined Workspaces can be exported to other interpretation tools to share the classification results from Paradise.
With the increase in the size of data volumes, traditional serversand workstations can limit the speed of interpretation. The newParadise HPC capability enables the generation of geometricand spectral decomposition attributes on clusters of machines, dramatically reducing computation time.
The Petrel connector allows Paradise users to import and export data from Paradise into and out of Petrel.
Import From Petrel
2D and 3D Seismic
Wells, including logs, velocity and markers
Export From Paradise
*Petrel is a registered trademark of Schlumberger
Below is a list of the minimum technical hardware and software requirements for running a single instance of the Paradise client and server on a dedicated workstation.
- Windows 7 or Windows 10 Pro
- MS SQL Server 2012
- 8+ processing cores
- 32GB RAM
- 4GB Nvidia graphics card
- File storage capacity depends on several different variables: See note below*
- 10Gb ethernet connection for networked data storage or USB 3.0 connection for external storage devices