Abstract: Disclosed are systems and methods for obtaining input data comprising properties associated with at least one parent well and a child well associated with the at least one parent well, dividing the input data into a training data subset, a validation data subset, and a test data subset, selecting at least one machine learning model using the training data subset, the validation data subset, and the test data subset based on k-fold cross validation, tuning hyper parameters for each of the at least one machine learning model, and generating a learning output using the at least one machine learning model and the hyper parameters for each of the at least one machine learning model, the learning output indicating a test root-mean-square error (RMSE) and a training RMSE.

Abstract: Disclosed are systems and methods for receiving historical production data associated with at least one hydraulic fracturing well, receiving time-series data associated with the at least one hydraulic fracturing well, the time-series data representing at least one type of data, receiving non-temporal data associated with the at least one hydraulic fracturing well, generating a machine learning model based on the historical production data, the time-series data associated with the at least one hydraulic fracturing well and based on an original job design during a first stage of the job at a particular hydraulic fracturing well, and the non-temporal data, determining an optimized job design for the particular hydraulic fracturing well having an objective function using a prediction based on the machine learning model, and implementing the optimized job design for the particular hydraulic fracturing well.

Abstract: A wellbore trajectory survey, having an ordered plurality of survey points, is acquired. Each of the plurality of survey points has a measured depth, an inclination, and a geographic location of the point in the wellbore corresponding to the survey point. The wellbore trajectory survey is ordered on measured depth. The processor identifies a minimum lateral measured depth (min_LMD), a maximum lateral measured depth (max_LMD). The processor identifies a mid-lateral point (mid_LMP) in the plurality of survey points whose measured depth (mid_LMD) is greater than min_LMD and less than max_LMD.

Abstract: A wellbore trajectory survey includes an ordered plurality of survey points, each of which has a measured depth, an inclination, and a geographic location. The wellbore trajectory survey is ordered on measured depth. A minimum lateral measured depth (min_LMD) as a measured depth of a first point in the ordered plurality of survey points for which the inclination exceeds an inclination threshold angle is identified. A maximum lateral measured depth (max_LMD) as a measured depth of a last point in the ordered plurality of survey points for which the inclination exceeds the inclination threshold angle is identified. The processor determines from the wellbore trajectory survey that the wellbore at measured depths between the min_LMD and the max_LMD exhibits porpoising into and out of a formation.

Abstract: The disclosure is directed to methods to design and revise hydraulic fracturing (HF) job plans. The methods can utilize one or more data sources from public, proprietary, confidential, and historical sources. The methods can build mathematical, statistical, machine learning, neural network, and deep learning models to predict production outcomes based on the data source inputs. In some aspects, the data sources are processed, quality checked, and combined into composite data sources. In some aspects, ensemble modeling techniques can be applied to combine multiple data sources and multiple models. In some aspects, response features can be utilized as data inputs into the modeling process. In some aspects, time-series extracted features can be utilized as data inputs into the modeling process. In some aspects, the methods can be used to build a HF job plan prior to the start of work at a well site.

Abstract: The disclosure is directed to methods to design and revise hydraulic fracturing (HF) job plans. The methods can utilize one or more data sources from public, proprietary, confidential, and historical sources. The methods can build mathematical, statistical, machine learning, neural network, and deep learning models to predict production outcomes based on the data source inputs. In some aspects, the data sources are processed, quality checked, and combined into composite data sources. In some aspects, ensemble modeling techniques can be applied to combine multiple data sources and multiple models. In some aspects, response features can be utilized as data inputs into the modeling process. In some aspects, time-series extracted features can be utilized as data inputs into the modeling process. In some aspects, the methods can be used to build a HF job plan prior to the start of work at a well site.

Abstract: The disclosure is directed to methods to design and revise hydraulic fracturing (HF) job plans. The methods can utilize one or more data sources from public, proprietary, confidential, and historical sources. The methods can build mathematical, statistical, machine learning, neural network, and deep learning models to predict production outcomes based on the data source inputs. In some aspects, the data sources are processed, quality checked, and combined into composite data sources. In some aspects, ensemble modeling techniques can be applied to combine multiple data sources and multiple models. In some aspects, response features can be utilized as data inputs into the modeling process. In some aspects, time-series extracted features can be utilized as data inputs into the modeling process. In some aspects, the methods can be used to build a HF job plan prior to the start of work at a well site.

Abstract: A wellbore trajectory survey includes an ordered plurality of survey points, each of which has a measured depth, an inclination, and a geographic location. The wellbore trajectory survey is ordered on measured depth. A minimum lateral measured depth (min_LMD) as a measured depth of a first point in the ordered plurality of survey points for which the inclination exceeds an inclination threshold angle is identified. A maximum lateral measured depth (max_LMD) as a measured depth of a last point in the ordered plurality of survey points for which the inclination exceeds the inclination threshold angle is identified. The processor determines from the wellbore trajectory survey that the wellbore at measured depths between the min_LMD and the max_LMD exhibits porpoising into and out of a formation.

Abstract: Disclosed are systems and methods for receiving historical production data associated with at least one hydraulic fracturing well, receiving time-series data associated with the at least one hydraulic fracturing well, the time-series data representing at least one type of data, receiving non-temporal data associated with the at least one hydraulic fracturing well, generating a machine learning model based on the historical production data, the time-series data associated with the at least one hydraulic fracturing well and based on an original job design during a first stage of the job at a particular hydraulic fracturing well, and the non-temporal data, determining an optimized job design for the particular hydraulic fracturing well having an objective function using a prediction based on the machine learning model, and implementing the optimized job design for the particular hydraulic fracturing well.

Abstract: Disclosed are systems and methods for obtaining input data comprising properties associated with at least one parent well and a child well associated with the at least one parent well, dividing the input data into a training data subset, a validation data subset, and a test data subset, selecting at least one machine learning model using the training data subset, the validation data subset, and the test data subset based on k-fold cross validation, tuning hyper parameters for each of the at least one machine learning model, and generating a learning output using the at least one machine learning model and the hyper parameters for each of the at least one machine learning model, the learning output indicating a test root-mean-square error (RMSE) and a training RMSE.

Abstract: A wellbore trajectory survey, having an ordered plurality of survey points, is acquired. Each of the plurality of survey points has a measured depth, an inclination, and a geographic location of the point in the wellbore corresponding to the survey point. The wellbore trajectory survey is ordered on measured depth. The processor identifies a minimum lateral measured depth (min_LMD), a maximum lateral measured depth (max_LMD). The processor identifies a mid-lateral point (mid_LMP) in the plurality of survey points whose measured depth (mid_LMD) is greater than min_LMD and less than max_LMD.

Abstract: One example of well treatment design based on three-dimensional wellbore shape can be implemented as a computer-implemented method. Wellbore data including acoustic logging data that defines an internal shape of a wall of the wellbore at multiple locations around the perimeter of the wellbore can be received. A volume of an open hole portion of the wellbore that includes the multiple locations can be determined using the wellbore data. Using the volume of the open hole portion of the wellbore, a volume of a fluid loss treatment to treat the portion of the wellbore for well fluid loss can be determined.

Abstract: One example of well treatment design based on three-dimensional wellbore shape can be implemented as a computer-implemented method. Wellbore data including acoustic logging data that defines an internal shape of a wall of the wellbore at multiple locations around the perimeter of the wellbore can be received. A volume of an open hole portion of the wellbore that includes the multiple locations can be determined using the wellbore data. Using the volume of the open hole portion of the wellbore, a volume of a fluid loss treatment to treat the portion of the wellbore for well fluid loss can be determined.

Abstract: A method, system, and apparatus, including a program encoded on computer-readable medium, for detecting duplicate data files to be stored in a well job data archive includes identifying a well job data file for storage in a well job data archive and performing an initial duplicate check and a secondary duplicate check to determine if the well job data file is a duplicate of a data file stored in the well job data archive, as well as a quality assurance test of the file data. At least one of the well job data file or one or more of the data files stored in the well job data archive is identified as a duplicate data file based on the initial and secondary duplicate checks, and the well job data file is stored in the well job data archive in accordance with the initial and secondary duplicate checks and the identification of a duplicate data file.

Abstract: A data mining and analysis system which analyzes a database of wellbore-related data in order to determine those predictor variables which influence or predict well performance.

Abstract: A method, system, and apparatus, including a program encoded on computer-readable medium, for detecting duplicate data files to be stored in a well job data archive includes identifying a well job data file for storage in a well job data archive and performing an initial duplicate check and a secondary duplicate check to determine if the well job data file is a duplicate of a data file stored in the well job data archive, as well as a quality assurance test of the file data. At least one of the well job data file or one or more of the data files stored in the well job data archive is identified as a duplicate data file based on the initial and secondary duplicate checks, and the well job data file is stored in the well job data archive in accordance with the initial and secondary duplicate checks and the identification of a duplicate data file.

Abstract: A method for generating an artificial neural network ensemble for determining stimulation design parameters. A population of artificial neural networks is trained to produce one or more output values in response to a plurality of input values. The population of artificial neural networks is optimized to create an optimized population of artificial neural networks. A plurality of ensembles of artificial neural networks is selected from the optimized population of artificial neural networks and optimized using a genetic algorithm having a multi-objective fitness function. The ensemble with the desired prediction accuracy based on the multi-objective fitness function is then selected.