Abstract: Geomechanical parameters can be used to optimize a well configuration that includes one or more projected wells having locations and geometries. Formation data and regional stress information of a formation can be used to determine a local stress variation of the formation. A quality index can be generated by combining petrophysical properties with the local stress variation. Hydrocarbon recovery flow simulations can be generated by generating well configuration models based on the quality index, generating reservoir geomechanical model that includes hydraulic fracture propagation characteristics, determining new hydraulic fractures by simulating propagation through the reservoir geomechanical model and using geomechanical rules, and determining a projected hydrocarbon recovery rate by simulating flow with the new hydraulic fractures. A well placement plan can be selected using the projected hydrocarbon recovery rates. The well placement plan can be output to be used to plan one or more wellbores.

Abstract: A least one seismic attribute is determined for each voxel of the seismic volume. A first horizon is selected for mapping and a sparse global grid is generated which includes the horizon, at least one constraint point identifying the horizon, and a number of points having a depth in the seismic volume. A value of at least one seismic attribute is determined for each point and their depths are adjusted based on the value of the seismic attribute. A map of the horizon can be generated based on the adjusted depths. Multiple local grids can be generated based on the sparse global grid, and the depths of the local grid points adjusted to generate a map of the horizon at voxel level resolution. The seismic volume can be mapped into multiple horizons, where previously mapped horizons can function as constraints on the sparse global grid.

Abstract: A system is described for determining a likelihood of a type of fluid in a subterranean reservoir. The system may include a processor and a non-transitory computer-readable medium that includes instructions executable by the processor to cause the processor to perform various operations. The processor may receive pre-stack seismic data having seismically-acquired data elements for geometric locations in a subterranean reservoir. The processor may determine, using the pre-stack seismic data, input features for each geometric location and may execute a trained model on the input features for determining a likelihood of a type of fluid in the subterranean reservoir and for determining a list of features affecting the likelihood. The processor may subsequently output the likelihood and the list of features.

Abstract: A system can provide for determining characteristics loss in a wellbore. The system can include a processor and a non-transitory memory with instructions that are executable by the processor for causing the processor to execute operations. The operations can include receiving, from sensors in a wellbore, data corresponding to loss indicators. The operations can include determining a loss probability for each loss indicator. The operations can include determining a total loss probability of fluid loss in the wellbore based on the loss probabilities. The operations can include outputting the total loss probability to be used in a drilling operation in the wellbore.

Abstract: Aspects of the disclosed technology provide techniques for determining frictional forces bearing on a downhole drill string. In some implementations, a method of the disclosed technology can include steps for segmenting a plurality of continuous nodes of the drilling string into a first segment and a second segment, computing a first set of values corresponding with one or more nodes in the first segment using a first model, computing a second set of values corresponding with one or more nodes in the second segment using a second model, and determining a torque of the drill string based on the first set of values and the second set of values. In some aspects, the method can further include steps for determining a drag force on the drill string based on the first set of values and the second set of values. Systems and machine-readable media are also provided.

Abstract: Optimizing seismic to depth conversion to enhance subsurface operations including measuring seismic data in a subsurface formation, dividing the subsurface formation into a training area and a study area, dividing the seismic data into training seismic data and study seismic data, wherein the training seismic data corresponds to the training area, and wherein the study seismic data corresponds to the study area, calculating target depth data corresponding to the training area, training a machine learning model using training inputs and training targets, wherein the training inputs comprise the training seismic data, and wherein the training targets comprise the target depth data, computing, by the machine learning model, output depth data corresponding to the study area based at least in part on the study seismic data; and modifying one or more subsurface operations corresponding to the study area based at least in part on the output depth data.

Abstract: Certain aspects and features relate to a system that includes a drilling tool, a processor, and a non-transitory memory device that includes instructions that are executable by the processor to cause the processor to perform operations. The operations include receiving input data that corresponds to characteristics of at least one of drilling fluid, a drillstring, or a wellbore. The operations also include calculating at least one dynamic sideforce and at least one dynamic, hydraulic force based at least in part on the input data. The operations also include determining an equilibrium solution for an output value using the at least one dynamic sideforce and at least one dynamic, hydraulic force. The operations also include applying the output value to the drilling tool for controlling operation of the drilling tool.

Abstract: System and methods for predicting time-dependent rock properties are provided. Seismic data for a subsurface formation is acquired over a plurality of time intervals. A value of at least one rock property of the subsurface formation is calculated for each of the plurality of time intervals, based on the corresponding seismic data acquired for that time interval. At least one of a trend or a spatio-temporal relationship in the seismic data is determined based on the value of the at least one rock property calculated for each time interval. A value of the at least one rock property is estimated for a future time interval, based on the determination. The estimated value of the at least one rock property is used to select a location for a wellbore to be drilled within the subsurface formation. The wellbore is then drilled at the selected location.

Abstract: A method of managing oilfield activity with a control system is provided having a plurality of virtual sensors and integrating the virtual sensors into a virtual sensor network. The method includes determining interdependencies among the virtual sensors, obtaining operational information from the virtual sensors, and providing virtual sensor output to the control system based on the determined interdependencies and the operational information.

Abstract: A system can assign a value to one or more sustainability factors for a wellbore operation based on historical data. The system can determine, for each of the one or more sustainability factors, a weight. The system can determine a sustainability index corresponding to a predicted carbon footprint for the wellbore operation based on the weight and the value for each of the one or more sustainability factors. The system can output a command for adjusting the wellbore operation based on the sustainability index.

Abstract: Embodiments of the subject technology for deep learning based reservoir modelling provides for receiving input data comprising information associated with one or more well logs in a region of interest. The subject technology determines, based at least in part on the input data, an input feature associated with a first deep neural network (DNN) for predicting a value of a property at a location within the region of interest. Further, the subject technology trains, using the input data and based at least in part on the input feature, the first DNN. The subject technology predicts, using the first DNN, the value of the property at the location in the region of interest. The subject technology utilizes a second DNN that classifies facies based on the predicted property in the region of interest.

Abstract: A system and method for controlling a drilling tool inside a wellbore makes use of simulated annealing and Bayesian optimization to determine optimum controllable drilling parameters. In some aspects, a computing device generates sampled exploration points using simulated annealing and runs a Bayesian optimization using a loss function and the exploration points to optimize at least one controllable drilling parameter to achieve a predicted value for a selected drilling parameter. In some examples, the selected drilling parameter is rate-of-penetration (ROP) and in some examples, the controllable drilling parameters include such parameters as rotational speed (RPM) and weight-on-bit (WOB). In some examples, the computing device applies the controllable drilling parameter(s) to the drilling tool to achieve the predicted value for the selected drilling parameter and provide real-time, closed-loop control and automation in drilling.

Abstract: A method comprises determining a measure of drilling efficiency, such as a friction factor or mechanical specific energy, of a drill bit used in a drilling operation of a wellbore and performing video analytics of at least one video that includes a substantially complete view of the wear surfaces of a drill bit to determine drill bit wear of the drill bit that is a result of the drilling operation of the wellbore. The method includes determining a cause of the drill bit wear based on the measure of drilling efficiency and the drill bit wear determined by performing video analytics. Based on correlation or modeling of drill bit wear and the measure of drilling efficiency, drill bit wear can be predicted and some types of drilling dysfunction mitigated in subsequent drilling runs.

Abstract: A system includes a processor and a memory. The memory includes instructions that are executable by the processor to cause the processor to receive basin data of a target basin including an area of the target basin, a number of exploration wells in the target basin, and a number of discovery wells in the target basin. Additionally, the instructions are executable to cause the processor to provide the basin data as input to a trained machine-learning model to determine a predicted trajectory of exploration efficiency of the target basin. Further, the instructions are executable to cause the processor to, in response to providing the basin data as input to the trained machine-learning model, receive an output from the trained machine-learning model indicating the predicted trajectory of exploration efficiency in the target basin.

Abstract: A self-adapting digital twin of a wellbore environment can be created. The self-adapting digital twin incorporates a standalone Ensemble Kalman Filter (EnKF) module with a constant parameter digital twin developed for a fixed environment. The standalone EnKF module receives streaming measurement data from multiple sensors and prediction data from the digital twin and executes the standalone EnKF module using the streaming measurement data and the prediction data from the digital twin. The results of executing the standalone EnKF module are input parameter corrections for the digital twin that are communicated to the digital twin. Output predictions of the digital twin are used to modify operational parameters of an oil or gas recovery process.

Abstract: The invention concerns a computer implemented method for underbalanced drilling. In one embodiment, the method includes determining a plurality of gas injection rate versus bottom hole pressure curves for a plurality of liquid injection rates for a specified minimum and maximum gas injection rate and a minimum and maximum liquid injection rate. Next, the method determines a set of four interception curves including a minimum motor equivalent liquid rate interception curve, a minimum vertical liquid velocity intercept curve, a minimum horizontal liquid velocity intercept curve, and a maximum motor equivalent liquid rate intercept curve for a specified minimum and maximum mud motor rate range and a minimum horizontal and vertical annulus velocity.

Abstract: A system can include one or more sensors at a wellsite. The system can detect first emission data about emissions with respect to a wellbore over a first period of time. The system can detect second emission data about emissions with respect to the wellbore over a second period of time. The system can determine an adjustment to a plug and abandonment operation with respect to the wellbore based on the first emission data. The system can determine an amount of reduction in emissions from the wellbore using the first emission data and the second emission data. The system can output the amount of reduction in the emissions.

Abstract: A system for analysis and display of hydrocarbon play information according to some aspects determines a probability of source rock occurrence according to source rock age based on a proven play concept. The system can also determine a relative probability of migration for hydrocarbons from a source rock of a proposed petroleum play concept to a reservoir. A relative probability of wellbore success for the proposed play concept can be determined at least in part based on these probabilities. The system can display the relative probability of wellbore success for the proposed play concept, either alone as part of a displayed inventory of proposed hydrocarbon play concepts. The system can produce accurate results that facilitate rapid play concept investigations for hydrocarbon exploration.

Abstract: A system can receive fault likelihood data about a subterranean environment and apply a binary mask filter using a tuning parameter to convert the fault likelihood data to binary distribution data having a plurality of pixels arranged in a plurality of profiles in at least two directions. The system can perform, for each profile of the plurality of profiles, fault skeletonization on the binary distribution data to form fault skeletonization data with pixels connected that represent part of a fracture. The system can convert the fault skeletonization data to seismic volume data and combine and filter the seismic volume data in the at least two directions to form combined seismic volume data. The system can output the combined seismic volume data as an image for use in detecting objects to plan a wellbore operation.

Abstract: A system and method for controlling a drilling tool inside a wellbore makes use of Bayesian optimization with range constraints. A computing device samples observed values for controllable drilling parameters such as weight-on-bit (WOB) and drill bit rotational speed in RPM and evaluates a selected drilling parameter such a rate-of-penetration (ROP) for the observed values using an objective function. Range constraints can be continuously learned by the computing device as the range constraints change. A Bayesian optimization, subject to the range constraints and the observed values, can produce an optimized value for the controllable drilling parameter to achieve a predicted value for the selected drilling parameter. The system can then control the drilling tool using the optimized value to achieve the predicted value for the selected drilling parameter.