Described embodiments generally relate to a computer-implemented method for modelling a subsurface formation. The method comprises receiving measurement data related to the subsurface formation, the measurement data comprising a plurality of data points; determining at least one rock physics model, each rock physics model relating to a rock type; assigning each data point of the measurement data to at least one initial rock class membership; fitting each determined rock physics model of the at least one rock physics model to the data points of the measurement data to produce at least one fitted rock physics model; reassigning each data point to at least one rock class based on the fitted rock physics models; determining whether a convergence criterion has been met; and responsive to the convergence criterion not being met, repeating the fitting and reassigning steps.

Methods, systems, and non-transitory computer readable media for analyzing type curve regions in a subsurface volume of interest are disclosed. Exemplary implementations may include: obtaining initial clusters of type curve regions in the subsurface volume of interest; obtaining production values as a function of position; generating an autocorrelation correction factor; attributing the autocorrelation correction factor to the production values as a function of position; generating type curve mean values; generating range distribution values; generating a type curve cluster probability value for each of the type curve regions; generating a first representation of the type curve regions as a function of position; clustering the type curve regions in updated clusters; generating a second representation of the type curve regions as a function of position; and displaying one or more of the first representation and the second representation.

A method includes obtaining total stresses and pore pressures of each porous medium of a formation, determining a first and second set of effective stresses for the formation, determining an individual collapse and fracturing mud weight for each porous medium of the formation using a first set of associated failure criteria, wherein the first set of associated failure criteria are based on the first set of effective stresses, determining an overall collapse and fracturing mud weight for the formation using a second set of associated failure criteria, wherein the second set of associated failure criteria is based on the second set of effective stresses, determining a mud weight window for the formation using the individual collapse mud weight, the individual fracturing mud weight, the overall collapse mud weight, and the overall fracturing mud weight, and transmitting a command to a drilling system based on the mud weight window.

A system can determine a location for future wells using machine-learning techniques. The system can receive seismic data about a subterranean formation and may determine a set of seismic attributes from the seismic data. The system can block the set of seismic attributes into a set of blocked seismic attributes by distributing the set of seismic attributes onto a geo-cellular grid representative of the subterranean formation. The system can apply a trained machine-learning model to the set of blocked seismic attributes to generate a composite seismic parameter. The system can distribute the composite seismic parameter in the subterranean formation to characterize formation locations based on a predicted presence of hydrocarbons.

A method includes receiving well log data for a plurality of wells. A flag is generated based at least partially on the well log data. The wells are sorted into groups based at least partially on the well log data, the flag, or both. A model is built for each of the wells based at least partially on the well log data, the flag, and the groups.

Methods for determining reservoir characteristics of a well can include receiving a first core from the well; performing an experiment to determine the wave velocity associated with a first direction of the first core, the experiment including: transmitting an ultrasonic wave through the first core in the first direction; receiving the transmitted ultrasonic wave; and determining a directional wave velocity of the first core based on the transmitted ultrasonic wave and the received transmitted ultrasonic wave, wherein the directional wave velocity represents a wave velocity along the first direction; rotating the first core about a longitudinal axis of the first core; and performing the experiment along a second direction of the first core.

A system and method of simulating fluid flow in a hydrocarbon reservoir is disclosed. The method includes obtaining a coarse grid model of the hydrocarbon reservoir and a trajectory of a wellbore that penetrates the hydrocarbon reservoir, and determining an initial grid geometry surrounding the trajectory. The method further includes constructing a reservoir simulation grid, conformal to the initial grid geometry in a first region in a vicinity of the wellbore and conformal with the coarse grid model in a second region more distant from the wellbore than the first region, and performing a hydrocarbon reservoir simulation, modeling a flow of fluid in the hydrocarbon reservoir based, at least in part, on the reservoir simulation grid.

A method, computer program product, and computing system are provided for receiving reservoir data associated with the reservoir. A simulation may be performed on the reservoir data to generate simulated reservoir data. A subset of realizations including a minimal number of realizations from a plurality of realizations may be determined based upon, at least in part, one or more statistical moments of the simulated reservoir data. An optimized reservoir model associated with an objective may be generated based upon, at least in part, the subset of realizations including the minimal number of realizations.

An ensemble of machine learning models is trained to evaluate seismic and risk-related data in order to evaluate, value, or otherwise rank various prospective hydrocarbon reservoir (“prospects”) of a field. A classification machine learning model is trained to classify a prospect or region of a prospect based on the exploration risk level. From the seismic data, a frequency-filtered volume (FFV) for each prospect is calculated, where the FFV is a measure of reservoir volume which takes into account seismic resolution limits. Based on the risk classification and FFV, prospects of the field are ranked based on their economic value which is a combination of the risk associated with drilling and their potential reservoir volume.

Implementations described and claimed herein provide systems and methods for reservoir modeling. In one implementation, an input dataset comprising seismic data is received for a particular subsurface reservoir. Based on the input dataset and utilizing a deep learning computing technique, a plurality of trained reservoir models may be generated based on training data and/or validation information to model the particular subsurface reservoir. From the plurality of trained reservoir models, an optimized reservoir model may be selected based on a comparison of each of the plurality of reservoir models to a dataset of measured subsurface characteristics.