Understanding the subsurface is critical to the needs of the E&P industry for minimising drilling risk and maximising subsurface recovery
S-Cube was spun out from Imperial College in 2015 to advance data-driven automated earth parameter learning from raw exploration seismic data. It arose at a point when waveform inversion of seismic data had become computationally feasible in 3D and proven with field recordings from ocean-bottom hydrophone sensors.
The first major anisotropic application of waveform inversion was conducted by Imperial College researchers and founding members of S-Cube in this award-winning study. This was at a site in the North Sea with a known shallow gas accumulation in the clastic section situated above a chalk reservoir. The waveform inversion successfully defines the structure using a data feed from a ocean-bottom seismic survey as shown below.
The company grew on the basis of this work and a breakthrough to a long-standing limitation in localised optimisation methodology known as "cycle-skipping". The adaptive formulation of waveform inversion known as AWITM was patented, proven on field data and became the recipient of the EAGE Guido Bonarelli Award for 2015. It is now in mainstream use within Equinor, Woodside Energy and most recently in 2019 within Chevron.
Ocean-bottom survey receiver cables are positioned on the seafloor with 1440 individual hydrophone locations marked as dots shown on the aerial view below. The left column is the input model grid with a manually inserted low blob to approximate the gas accumulation whilst the right column shows the iterated model grid converting the low velocity zone into a defined structure.
Also displayed is the predicted data shifting closer and closer to observed data for a shot firing into a cable of receivers over the anomaly as the details of the gas charged interbedded layers and faults get captured into the velocity volume.
What happens if we eliminate the low velocity zone from the starting model completely? The gas cloud is detected but the model iterates to a local minimum trap from which it cannot recover. This is due to a phenomenon known as cycle-skipping where the predicted data shifts in a direction which causes it to misalign with the observed data.
As a result of cycle skipped mismatch, the wider applicability of the generic technique is limited given reliance on a-priori knowledge of the target, i.e. an accurate starting model. S-Cube formed to specifically tackle this through restructuring of the data fitting feedback loop optimisation cost function, making it sufficiently convex to be able to start further from the true answer without misconvergence.
S-Cube AWITM driven by an adaptive cost function reformulates generic waveform inversion to overcome cycle skipping
S-Cube formulates Adaptive Waveform Inversion (AWITM) using a mapping between the observed and predicted data and minimising the misfit to send the non zero lag filter coefficients to zero. This solves a long-standing theoretical problem in waveform inversion parameter learning succumbing to false local minima due to oscillatory seismic data.
Repeating the previous example with the adaptive cost function and changing nothing else gives the result below. It enables successful recovery of the gas cloud from a cycle-skipped initial position at which which point conventional waveform inversion with a least-squares norm cost function can be applied to return to the previously obtained global minimum result.
Pre-drill predictive power of AWITM shown in Woodside Energy exploration application offshore Myanmar
The application of AWITM to narrow azimuth towed steamer data leads to pre-drill prediction of gas bearing formation offshore Myanmar. The low velocity interval prediction is confirmed by a subsequent well penetration of the zone.
The XWITM toolbox enhances the ability of waveform inversion to extend the model update deeper with reflection energy
Reflection Waveform Inversion (RWI) specifically targeting the deepest penetrating energy in the survey is added to the XWITM toolkit.
On the left is the final result using XWITM toolkit which had the closest agreement with true answer out of all industry contestants. On the right is the equivalent result using generic waveform inversion with spurious velocity variations occurring below the diving wave zone.
Cloud-native XWITM launches on Amazon Web Services with pilot user Tullow Oil
Users can now access the industry’s leading FWI algorithm on demand in a managed VPC environment hosted on the cloud with the benefit of multiple cases running concurrently to enhance convergence to the basin of the misfit global minimum.
S-Cube is on a journey ... it's not yet reached its destination but currently has an industry leading product and with AWS it is positioned for accelerated growth.
Run XWI on the Cloud
Discover an unprecedented increase in the resolution of your velocity model.