Drilling rig operations may be monitored using a variety of sensors and/or other data sources. Erroneous, faulty, and/or missing data may be cleansed prior to using the data for modeling and/or monitoring drilling operations. Erroneous, faulty, and/or missing data may be identified by comparing received data to anticipated values based on historical operations, other physically related sensor readings, and known operating ranges. Cleansing may comprise replacing erroneous, faulty, and/or missing data with a modeled value or omitting a reading entirely.

Drilling rig operations may be monitored using a variety of sensors and/or other data sources. Erroneous, faulty, and/or missing data may be cleansed prior to using the data for modeling and/or monitoring drilling operations. Erroneous, faulty, and/or missing data may be identified by comparing received data to anticipated values based on historical operations, other physically related sensor readings, and known operating ranges. Cleansing may comprise replacing erroneous, faulty, and/or missing data with a modeled value or omitting a reading entirely.

An apparatus used to provide preprocessed variables to a predictive engine. The predictive engine generates predictive results, based on the variables, to automate well site operations. The apparatus comprises an analysis module, a pattern recognition module, and a library module. The analysis module identifies a well site operation by examining a well site operation variable, determines categories and standard operating procedures associated with the categories using the well site operation and a-priori information, and searches a library of historical information using the categories. The historical information comprising classified procedures and recommendations of historic well site operations. The pattern recognition module identifies a pattern using a statistics based algorithm. The algorithm uses the standard operating procedures, the categories, and the classified procedures and recommendations. The pattern indicating a deviation in the standard operating procedure. The library module classifies the well site operation variables and stores the classified variables.

Disclosed embodiments provide techniques for estimating imputation algorithm performance. Multiple imputer algorithms are selected, and an evaluation of how well each of the imputer algorithms can estimate the missing data is performed. Disclosed embodiments obtain an imputer candidate dataset (ICD). The imputer candidate dataset is compared to the incomplete data range, and a similarity metric is determined between the data range and the ICD. When the similarity metric exceeds a predetermined threshold, an imputer evaluation dataset (IED) is created from the ICD by removing one or more data points from the ICD. Each imputer algorithm is evaluated by applying the IED to it, and computing an imputer evaluation metric based on its performance. The multiple imputer algorithms are ranked based on the imputer evaluation metric. The best ranked imputer algorithm can then be selected for use on the incomplete data range within the measurement dataset.

Methods and systems that provide data privacy for implementing a neural network-based inference are described. A method includes injecting stochasticity into the data to produce perturbed data, wherein the injected stochasticity satisfies an -differential privacy criterion and transmitting the perturbed data to a neural network or to a partition of the neural network for inference.

Methods and systems that provide data privacy for implementing a neural network-based inference are described. A method includes injecting stochasticity into the data to produce perturbed data, wherein the injected stochasticity satisfies an -differential privacy criterion and transmitting the perturbed data to a neural network or to a partition of the neural network for inference.

A computer-implemented method for data analysis comprises obtaining a plurality of first observations, each one of the plurality of first observations including one or more values of one or more first parameters, the plurality of first observations grouped into a plurality of groups; constructing a first histogram using the values of at least one of the one or more first parameters, included in the plurality of first observations; constructing, for each one of the plurality of groups, a second histogram having bins corresponding to bins of the first histogram, wherein each one of the bins of the second histogram includes a count of the first observations, among the first observations that belong to the one of the plurality of groups, having one or more values corresponding to the one of the bins for the at least one of the one or more first parameters; and outputting the second histograms.

A computer-implemented method for data analysis comprises obtaining a plurality of first observations, each one of the plurality of first observations including one or more values of one or more first parameters, the plurality of first observations grouped into a plurality of groups; constructing a first histogram using the values of at least one of the one or more first parameters, included in the plurality of first observations; constructing, for each one of the plurality of groups, a second histogram having bins corresponding to bins of the first histogram, wherein each one of the bins of the second histogram includes a count of the first observations, among the first observations that belong to the one of the plurality of groups, having one or more values corresponding to the one of the bins for the at least one of the one or more first parameters; and outputting the second histograms.

A computer-implemented method for data analysis comprises obtaining a plurality of first observations, the plurality of first observations including one or more values of one or more first parameters, the plurality of first observations grouped into a plurality of groups; constructing a first histogram using the values of at least one of the one or more first parameters, included in the plurality of first observations; constructing, for each of the plurality of groups, a second histogram having bins corresponding to bins of the first histogram, wherein each of the bins of the second histogram includes a count of the first observations, among the first observations that belong to the one of the plurality of groups, having one or more values corresponding to the one of the bins; and outputting the second histograms constructed for the plurality of groups.

A computer-implemented method for data analysis comprises obtaining a plurality of first observations, the plurality of first observations including one or more values of one or more first parameters, the plurality of first observations grouped into a plurality of groups; constructing a first histogram using the values of at least one of the one or more first parameters, included in the plurality of first observations; constructing, for each of the plurality of groups, a second histogram having bins corresponding to bins of the first histogram, wherein each of the bins of the second histogram includes a count of the first observations, among the first observations that belong to the one of the plurality of groups, having one or more values corresponding to the one of the bins; and outputting the second histograms constructed for the plurality of groups.