Machine Learning to Reduce Cycle Time for Time-lapse Seismic Data Assimilation into Reservoir Management

The proposed low cost 4D QI workflow using Machine Learning fills the gap between qualitative interpretation of 4D attribute maps and 4D probabilistic inversion of seismic wiggles, thus enabling the rapid quantification of reservoir property changes. The estimated water saturation changes can then be used to provide key geophysical input to analyze injection efficiency, update the reservoir model, and support decisions on the water flood optimization, especially for multiple repeated 4D seismic surveys.