Machine learning model selection based on feature merging for a spatial location across multiple time windows

A system comprising:

• at least one sensor to be positioned in a wellbore created in a subsurface formation;
• a processor;
• and
• a machine-readable medium having program code executable by the processor to cause the processor to:
• receive, in a real time stream, a current dataset for a current time window from the at least one sensor, wherein the current dataset comprises values of a number of current features of the subsurface formation at a spatial location in the wellbore;
• select at least one previous time window from a number of previous time windows that includes a previously cached dataset that was detected by the at least one sensor or a different sensor in the wellbore and that spatially overlaps with the spatial location for the current dataset;
• merge the current dataset with the previously cached dataset across the current time window and the at least one selected previous time window to create a merged dataset;
• and
• predictively evaluate the subsurface formation in real time during hydrocarbon recovery operations using a machine learning model selected from a plurality of machine learning models for the spatial location in the wellbore based on a comparison between a probability distribution of a training dataset and a probability distribution for the merged dataset.

One or more non-transitory machine-readable media comprising program code executable by a processor to cause a processor to:

• receive, in a real time stream, a current dataset for a current time window from at least one sensor in a wellbore created in a subsurface formation, wherein the current dataset comprises values of a number of current features of the subsurface formation at a spatial location in the wellbore;
• select at least one previous time window from a number of previous time windows that includes a previously cached dataset that was detected by the at least one sensor or a different sensor in the wellbore and that spatially overlaps with the spatial location for the current dataset;
• merge the current dataset with the previously cached dataset across the current time window and the at least one selected previous time window to create a merged dataset that comprises at least a portion of the previously cached dataset and at least a portion of the current dataset;
• and
• predictively evaluate the subsurface formation in real time during hydrocarbon recovery operations using a machine learning model selected from a plurality of machine learning models for the spatial location in the wellbore based on a comparison between a probability distribution of a training dataset and a probability distribution for the merged dataset.