The subject disclosure provides for a method of optical sensor calibration implemented with neural networks through machine learning to make real-time optical fluid answer product prediction adapt to optical signal variation of synthetic and actual sensor inputs integrated from multiple sources. Downhole real-time fluid analysis can be performed by monitoring the quality of the prediction with each type of input and determining which type of input generalizes better. The processor can bypass the less robust routine and deploy the more robust routine for remainder of the data prediction. Operational sensor data can be incorporated from a particular optical tool over multiple field jobs into an updated calibration when target fluid sample compositions and properties become available. Reconstructed fluid models adapted to prior field job data, in the same geological or geographical area, can maximize the likelihood of quality prediction on future jobs and optimize regional formation sampling and testing applications.

Methods and an alarming system for long-term carbon dioxide sequestration in a geologic reservoir are described. The geologic reservoir may be a water filled sandstone reservoir or a carbonate reservoir. A reservoir model is constructed to show the effects of varying injection pressures, the number of injection wells, the arrangement of injection wells, the boundary conditions and sizes of the reservoir on caprock uplift, fracture formation and fracture reactivation. The alarming system generates an alarm when caprock uplift that surpasses a threshold is detected. The injection pressures and the number of injection wells operating may be varied in response to the alarm.

A system for determining completion parameters for a wellbore includes a sensor and a computing device. The sensor can be positioned at a surface of a wellbore to detect data prior to finishing a completion stage for the wellbore. The computing device can receive the data, perform a history match for simulation and production using the sensor data and historical data, generate inferred data for completion parameters using the historical data identified during the history match, predict stimulated area and production by inputting the inferred data into a neural network model, determine completion parameters for the wellbore using Bayesian optimization on the stimulated area and production from the neural network model, profit maximization, and output the completion parameters for determining completion decisions for the wellbore.

Provided are embodiments that include segmenting a trajectory of a wellbore of a hydrocarbon well. For each segment, determining whether the segment includes a connection or a flow control element, and in response to determining that the segment includes a connection, generating a conduit link for the segment, in response to determining that the segment includes one or more flow control elements, generating an element link for the segment, or in response to determining that the segment does not include a connection or a flow control element, merging a modeling of the segment with modeling of a conduit link for an adjacent segment of the wellbore. Generating, based on the conduit links and element links generated, a reduced connectivity graph for the hydrocarbon well and a simulation of the hydrocarbon well.

A method and system for performing a pressure test. The method may comprise inserting a formation testing tool into a wellbore to a first location within the wellbore, identifying one or more tool parameters of the formation testing tool, performing a first pre-test with the pressure transducer when the pressure has stabilized to identify formation parameters, inputting the formation parameters and the one or more tool parameters into a forward model, changing the one or more tool parameters to a second set of tool parameters; performing a second pre-test with the second set of tool parameters; and comparing the first pre-test to the second pre-test. A system may comprise at least one probe, a pump disposed within the formation testing tool, at least one stabilizer, a pressure transducer disposed at least partially in the at least one fluid passageway, and an information handling system

A method, apparatus, and program product to determine distribution of a plurality of components amongst a plurality of phases for a multi-component, multi-phase system including a multi-component, multi-phase fluid. A plurality of phase boundaries of the multi-component, multi-phase fluid and a vapor-liquid equilibrium (VLE) are determined based on a plurality of geophysical parameters associated with an oilfield and using one or more computer processors, including by determining hydrocarbon partitioning in a water phase, based in part on applying empirical equilibrium multi-phase mole fraction ratios (K-values) of the multi-component, multi-phase system that are functions of temperature and pressure only. In addition, an amount of at least one fluid component distributed in a plurality of phases of the multi-component, multi-phase system is predicted by solving a set of flash equations with the one or more computer processors based on the plurality of phase boundaries.

Disclosed embodiments include well ranging apparatus, systems, and methods which operate to receive normal components of electromagnetic field strength azimuthal measurements within a first well at different tool azimuth angles in the first well. Further activities include determining an approximate range from the sensors to a second well that serves as a source of an electromagnetic field, via direct transmission or backscatter transmission, using the normal components of the electromagnetic field strength azimuthal measurements. In some embodiments, the approximate range can be determined without introducing sensor azimuthal separation into range calculations. Additional apparatus, systems, and methods are disclosed.

A tool for screening unconventional reservoirs to determine the location of economically important accumulations of hydrocarbons early in a reservoir development process is described. Once the accumulations are identified, subsequent process such as drilling wells and producing hydrocarbons can begin.

A system includes a receiver component that receives a model of a well system, the model comprising a representation of an electrically energized well casing and/or a fracture as a transmission line that leaks electric current as the current traverses the well casing and/or fracture. The receiver component also receives a value that indicates an amount of the current that is applied to the well casing and a location of a source of the current on the well casing. The system further includes an electromagnetic field calculator component that calculates an estimated electromagnetic field at at least one location on the surface of the earth based at least in part upon the representation of the electrically energized well casing and/or fracture, the value that indicates the amount of current that is applied to the well casing, and the location of the source of the current on the well casing.

The disclosure provides a method of generating a basin fill model using a set of known paleogeographic characteristic parameters, for a specified basin location and time interval. The basin fill model can be used to assist in predicting the location of submarine fan deposits containing commercially valuable hydrocarbons or minerals. The generated models and predicted locations can be used in a well system operation plan. A computer program product is also disclosed that can retrieve sets of known paleogeographic data and generate multiple interim models and parameters that can be used for further predictions on where, and at what depth, valuable deposits may be found. Addi-tionally, a basin fill modeling system is disclosed that can retrieve and store known characteristic parameters for various geographic locations and time periods and utilize those characteristic parameters in algorithms to generate basin fill models and to predict where valuable submarine fan deposits are located.