Aspects of the present disclosure relate to confidence volumes for earth modeling using machine learning. A method includes receiving detected data, wherein the detected data includes formation attributes relating to one or more depth points along or near a wellbore. The method further includes providing inputs to a plurality of machine learning models based on the detected data. The method further includes receiving output values from the plurality of machine learning models based on the inputs. The method further includes determining a measure of variance among the output values. The method further includes generating a confidence indicator related to the output values based on the measure of variance.

Systems and methods for developing hydrocarbon reservoirs based on constrained natural fracture parameters. A natural fracture modeling is generated for a reservoir, an initial set of fracture model parameters is determined, and a fracture model optimization is conducted to determine an optimized set of fracture model parameters. The optimized set of fracture model parameters are used as a basis for modeling the reservoir, and the modeling is used to generate a simulation of the reservoir.

Improved reservoir simulation methods and systems are provided that employ a new velocity model in conjunction with a sequential implicit (SI) formulation or Sequential Fully Implicit (SF) formulation for solving the discrete form of the system of nonlinear partial differential equations. In embodiments, the new velocity model employs a fluid transport equation part based on calculation of phase velocity for a number of fluid phases that involves capillary pressure and a modification coefficient. In embodiments, the modification coefficient can be based on a derivative of capillary pressure with respect to saturation. In another aspect, the new velocity model can employ an estimate of the phase velocity of the water phase v.sub.w_est that is based on one or more derivatives of capillary pressure of the water phase as a function of water saturation.

Systems and methods are provided for determining a formation body wave slowness from an acoustic wave. Waveform data is determined by logging tool measuring the acoustic wave. Wave features are determined from the waveform data and a model is applied to the wave features to determine data-driven scale factors The data-driven scale factors can be used to determine a body wave slowness within a surrounding borehole environment and the body wave slowness can be used to determine formation characteristics of the borehole environment.

A method may include generating a first well intervention plan for a well site automatically based on a predetermined scheduling criterion. The first well intervention plan is generated using a first set of data inputs regarding one or more well intervention providers and one or more well conditions. The method may further include obtaining first well data regarding the well site. The method may further include adjusting the predetermined scheduling criterion to produce an adjusted scheduling criterion using the first well data. The adjusted scheduling criterion corresponds to a second set of data inputs that are different from the first set of data inputs. The method may further include generating a second well intervention plan for the well site based on the adjusted scheduling criterion. The method may further include transmitting a command to the well site that adjusts well operations based on the second well intervention plan.

Aspects of the subject technology relate to systems and methods for optimizing multi-unit pumping operations at a well site. Systems and methods are provided for receiving sensor data from a hydraulic fracturing fleet equipment at an equipment system, designating an event as being flagged based on the sensor data from the hydraulic fracturing fleet equipment, determining a physical action based on the flagged event and a priority list of actions, and providing instructions to a first pump of the hydraulic fracturing fleet equipment to perform the physical action based on the flagged event and the priority list of actions.

A computer-implemented method may include obtaining well logs data pertaining to a well of interest. The method may further include training a physics-constrained machine learning (PCML) model using the obtained well logs data as inputs. The method may further include outputting one or more sonic logs and mechanical properties of interest determined by using the trained PCML model and the obtained well logs data for the well of interest. The method may further include updating the determined sonic logs and mechanical properties of interest based on a breakout model and field breakout data for the well of interest. The method may further include outputting the final sonic logs for the well of interest. The method may further include determining one or more mechanical properties for well planning based on the final sonic logs for the well of interest.

A method may include obtaining, from various sensors, acquired sensor data regarding various multiphase flows in a multiphase flow meter that are sampled at a predetermined sampling frequency. The acquired sensor data may describe various transient signals that correspond to various gas droplets. The method may further include generating, based on the acquired sensor data, a flow model for the multiphase flow meter. The method may further include updating the flow model to produce a first updated flow model using simulated flow data. The method may further include updating the first updated flow model to produce a second updated flow model using simulated sensor data. The second updated flow model may be used to determine one or more flow rates within a multiphase flow.

A method of identifying events includes obtaining an acoustic signal from a sensor, determining one or more frequency domain features from the acoustic signal, providing the one or more frequency domain features as inputs to a plurality of event detection models, and determining the presence of one or more events using the plurality of event detection models. The one or more frequency domain features are obtained across a frequency range of the acoustic signal, and at least two of the plurality of event detection models are different.

A drilling system includes a drill string, a plurality of sensors, and a computing system. The drill string includes a downhole motor. The sensors are coupled to the drill string. The computing system is coupled to the sensors. The computing system is configured to compute, based on measurements provided by the sensors, a motor stall index, and to determine, by comparing the motor stall index to a motor stall threshold, whether the downhole motor has stalled. The computing system is also configured to, responsive to a determination that the downhole motor has stalled, adjust operation of the drill string to restart the downhole motor.