nuroTEEC

neuroTEEC provides the mapping of seismic facies by means of an unsupervised Neural Network. This classification technique is based on a patent granted to TEEC. The workflow comes with the assumption that changes in lithology, rock properties and fluid content should affect the seismic traces with respect to amplitude, shap and lateral continuity. This leads to the approach to take advantage of three dimensional seismic pattern where the individual characteristics of these pattern should be detected by the Neural Network Contrary to the generation of dozens of seismic attributes which may be related to the reservoir heterogeneity later on. The Neural Network comes with the spatial distribution of classes representing similar signal characteristics or simply seismic facies classes. These seismic facies classes often represent features of the reservoir hetereogeneity that are otherwise hard to detect or time consuming to interpret.

Features:

• Neural Network Seismic Facies Classifications
• Processing of Volume Attributes
• Quantitative Data Verification
• Institive Program Handling
• Simple Data In & Out

Workign with nuroTEEC is a four step approach:

• Extraction of 3D training patterns from the seismic data by a random process
• Setting up a network architecture, training and analysing of training results
• Classification of the whole seismic data
• Clibration of facies classes by means of key wells and geological information

nuroTEEC is not a black box software. The menu driven interface is user friendly and intuitive to learn. Working with neuroTEEC allows an easy handling of training patterns, generation of control plots at each stage of the workflow and analysing the classification results.