Conventional inversion possesses a major limitation related to bandwidth in seismic reservoir characterization studies: Seismic frequency content governs the resolution power of 3D characterization models

Amplitude Spectrum Comparison: Conventional Inversion Versus TROID Inversion

Multi-Physics Technologies offers TROID machine learning seismic inversion as an alternative to circumvent this limitation, utilizing the power of artificial intelligence.

Reduce subsurface uncertainty in seismic reservoir characterization studies by predicting subsurface properties in 3D using machine learning. Our TROID algorithm uses ensemble learning to predict acoustic, elastic, petrophysical, and reservoir properties from seismic and well data with resolution beyond the input seismic bandwidth

- Predict elastic, petrophysical or reservoir properties
- Input data types include prestack 3D, multi-component, 4D, and azimuthal seismic data
- One step process (no seismic inversion in between)
- Bandwidth beyond seismic
- Depth or time domain

Geosolutions

Project Experience

OTC

Subsurface Characterization Using Ensemble Machine Learning – Presented at the OTC Annual Conference and Exhibition at NRG Park in Houston, Texas, USA

NExT Course - Karachi, Pakistan

Multi-Physics Technologies Geophysicist Teaches A Modern Approach to Seismic Interpretation with Petrel in Karachi, July 31 – August 4, 2023

GSH Journal

Geophysical Society of Houston Journal: Spectral Extrapolation and Acoustic Inversion for the Characterization of an Ultra-Thin Reservoir

NExT Course - Quito, Ecuador

Multi-Physics Technologies Geophysicist Teaches AVO and Seismic Inversion With Petrel to PetroEcuador in Quito, May 29 – June 2 2023

AAGP

Resolving Sedimentary Features in a Lower Miocene Clastic Reservoir, Jeanerette Field, Louisiana – Presented at the joint AAPG and SEG International Meeting for Applied Geoscience and Energy, Denver, USA