Production from multi-well and multi-bench unconventional development is time-consuming to model in physics-based simulators and requires multiple runs. Even with the use of high-performance computing or cloud computing, each single run can take several minutes to a few hours depending on the model complexity. It is very computationally demanding and almost impractical to perform full subsurface uncertainty and multiple scenario realizations. Reservoir simulation is highly mature and complex in terms of both implemented physics as well as the numerics used to solve the governing partial differential equations. Production from typical multi-well and multi-bench unconventional developments is a spatio-temporal problem and highly dynamic in nature.
We found that the proxy model predictions accurately match the trends as well as magnitudes when compared to computationally expensive, high-fidelity numerical simulations across several real examples of multi-well and multi-bench developments.
A lightning-fast reservoir proxy model significantly reduces the cycle-time for using physics-based models and workflows and captures subsurface uncertainty more holistically. The proxy model workflow benefits from standard features of machine learning systems, including interpretability and confidence scores that provide the user with richer information prior to deciding. Therefore, the physics-based proxy model is a powerful addition to an engineer’s toolkit who is involved in optimizing unconventional development.
Other Highlights of the Paper:
Reducing the cycle time of such modeling studies by several orders of magnitude by creating surrogate models that are trained on unconventional production models created offline for a wide range of inputs covering all the unconventional assets.
- Science-Based AI Benefits
The AI/ML algorithms developed in this work can fully understand the highly non-linear relationships between 100’s of model inputs and the full time series production outputs of every well.
The platform infrastructure supports inference at scale (e.g. ability to run over millions of inferences per month) as well as integrating the final proxy model into other workflows such as history matching, sensitivity analysis, uncertainty quantification, economic modeling etc. using an API.