Classifying downhole test data

A non-transitory, machine-readable medium having instructions stored thereon that are executable by a computing device to perform operations comprising:

• determining one or more variable types in a multivariate test vector within a test data set, the test data set generated by the downhole measurement tools in a downhole environment;
• for a plurality of machine-learning models, determining a closest match between variable types used by the plurality of machine-learning models and the determined one or more variable types for the multivariate test vector;
• in response to determining the closest match for one of the plurality of machine-learning models, selecting a corresponding machine-learning model and classifying the multivariate test vector using the corresponding machine-learning model to determine formation properties of the downhole environment;
• and
• in response to determining the closest match for multiple machine-learning models of the plurality of machine-learning models,
• determining a similarity between a probability distribution for the test data set and probability distributions for the multiple machine-learning models to generate similarity values for each of the multiple machine-learning models;
• and
• in response to one of the similarity values exceeding a threshold value, selecting a machine-learning model corresponding to the exceeding similarity value and classifying the multivariate test vector using the selected machine-learning model to determine formation properties of the downhole environment;
• and
• controlling well equipment by modifying operation settings of the well equipment based on the classification of the multivariate test vector.

An apparatus for classifying test data from downhole measurement tools, said apparatus comprising:

• a processor;
• and
• a computer-readable medium having instructions stored thereon that are executable by the processor to cause the apparatus to,
• determine one or more variable types in a multivariate test vector within a test data set, the test data set generated by the downhole measurement tools in a downhole environment;
• for a plurality of machine-learning models, determine a closest match between variable types used by the plurality of machine-learning models and the determined one or more variable types for the multivariate test vector;
• in response to determining the closest match for one of the plurality of machine-learning models, selecting a corresponding machine-learning model and classifying the multivariate test vector using the corresponding machine-learning model to determine formation properties of the downhole environment;
• and
• in response to determining the closest match for multiple machine-learning models of the plurality of machine-learning models,
• determine a similarity between a probability distribution for the test data set and probability distributions for the multiple machine-learning models to generate similarity values for each of the multiple machine-learning models;
• and
• in response to one of the similarity values exceeding a threshold value, select a machine-learning model corresponding to the exceeding similarity value and classifying the multivariate test vector using the selected machine-learning model to determine formation properties of the downhole environment;
• and
• control well equipment by modifying operation settings of the well equipment based on the classification of the multivariate test vector.