US20200240259A1

US20200240259A1 - Flow network model analysis

Info

Publication number: US20200240259A1
Application number: US16/750,710
Authority: US
Inventors: Mahadevan Balasubramaniam; Arun Karthi Subramaniyan; Shyam Sivaramakrishnan; Fabio Nonato de Paula
Original assignee: GE Inspection Technologies LP
Current assignee: Waygate Technologies USA LP
Priority date: 2019-01-25
Filing date: 2020-01-23
Publication date: 2020-07-30
Also published as: WO2020154468A1

Abstract

In one implementation, a method includes generating a manifold predictive model configured to calculate an initial virtual measurement associated with an oil field comprising a plurality of oil wells. The manifold predictive model can be based on one or more predictive models associated with one or more components of the oil field. The method also includes receiving data characterizing one or more pressure measurements and flow measurements obtained in the oil field. The method further includes determining a prospective sensor location of a first prospective sensor in the oil field. The first prospective sensor can be configured to detect an oil field parameter. The manifold predictive model can be configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement. The method also includes providing the prospective sensor location and the identity of the first prospective sensor.

Description

RELATED APPLICATION

This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No. 62/797,097 filed on Jan. 25, 2019, the entire contents of which are hereby expressly incorporated by reference herein.

BACKGROUND

Process values of process variables at a process plant (e.g., oil flow at an oil rig) can be tracked (e.g., at regular interval) to monitor the operation of the plant. Observing the process variables can allow an operator to ensure desirable operation of the plant. The process values can be measured, for example, by sensors (e.g., fluid flow meters, pressure gauges, thermocouples, accelerometers) located at the process plant. However, it may not be possible to detect values of all the desirable processes and/or values of a process at multiple locations in the process plant. This can be due to prohibitive cost of installing multiple sensors. Additionally, sensors that can detect certain processes (e.g., multi-phase fluid flow) can be expensive.
Numerical simulation based on regression models can be used to predict process values that cannot be directly measured. The numerical simulations can use process values measured by one or more sensors added to the process plant as outputs of the regression models. Such techniques may not be accurate as they do not model the actual processes at the plant and can be prone to over fitting. Additionally, these regression-based methods may require a large set of additional data for building the regression model.

SUMMARY

In general, apparatus, systems, methods and articles of manufacture for determination of virtual process parameters are provided.
In one implementation, a method includes generating a manifold predictive model configured to calculate an initial virtual measurement associated with an oil field comprising a plurality of oil wells. The manifold predictive model can be based on one or more predictive models associated with one or more components of the oil field. The method also includes receiving data characterizing one or more pressure measurements and flow measurements obtained in the oil field. The method further includes determining a prospective sensor location of a first prospective sensor in the oil field. The first prospective sensor can be configured to detect an oil field parameter. The manifold predictive model can be configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement. The method also includes providing the prospective sensor location and the identity of the first prospective sensor.
One or more of the following features can be included in any feasible combination.
In one implementation, a first sensitivity associated with the updated virtual measurement can be smaller than a second sensitivity associated with the initial virtual measurement. In another implementation, the one or more predictive models can include an oil well predictive model and a pipeline predictive model. In yet another implementation, at least one of the one or more measurement is at a manifold of the oil field.
In one implementation, the manifold predictive model is associated with a first manifold in the oil field, the first manifold coupled to a plurality of oil wells in the oil field. The one or more predictive models includes one or more of flow models associated with pipes connecting the first manifold to the plurality of oil wells, and physical equations and/or sensor measurements associated with one or more of the plurality of oil wells.
In one implementation, the manifold predictive model is configured to calculate a probabilistic estimation indicative of an operating range of a flow rate associated with one or more of oil, gas and water from the first manifold. In another implementation, the calculation of probabilistic estimation of flow rates is based on one or more sensor measurements at the first manifold. In yet another implementation, the manifold predictive model is configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement.
Non-transitory computer program products (i.e., physically embodied computer program products) are also described that store instructions, which when executed by one or more data processors of one or more computing systems, causes at least one data processor to perform operations herein. Similarly, computer systems are also described that may include one or more data processors and memory coupled to the one or more data processors. The memory may temporarily or permanently store instructions that cause at least one processor to perform one or more of the operations described herein. In addition, methods can be implemented by one or more data processors either within a single computing system or distributed among two or more computing systems. Such computing systems can be connected and can exchange data and/or commands or other instructions or the like via one or more connections, including a connection over a network (e.g. the Internet, a wireless wide area network, a local area network, a wide area network, a wired network, or the like), via a direct connection between one or more of the multiple computing systems, etc.
These and other capabilities of the disclosed subject matter will be more fully understood after a review of the following figures, detailed description, and claims.

BRIEF DESCRIPTION OF THE FIGURES

These and other features will be more readily understood from the following detailed description taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a schematic illustration of an oil field;

FIG. 2 illustrates an exemplary estimation of handoff flowrate;

FIG. 3 illustrates a representation of an exemplary well and lift model;

FIG. 4 illustrates exemplary gas flow virtual measurements of well models for naturally flowing reservoir and gas gifted wells;

FIG. 5 illustrates an exemplary pair plot of manifold pressures;

FIG. 6 illustrates virtual output of flow meter determined for simulation points in FIG. 5;

FIG. 7 illustrates an exemplary sensitivity plot for FIGS. 5 and 6;

FIG. 8 illustrates measurements and uncertainties associated with the measurements at the three manifolds identified in FIG. 2;

FIG. 9 illustrates a probabilistic estimation of a flowrate associated with the manifold C in FIG. 2;

FIG. 10 illustrates a probabilistic estimation of a flowrate associated with the manifold A1 in FIG. 2;

FIG. 11 illustrates flowrates at A1 and NE1 manifolds of FIG. 2;

FIG. 12 illustrates flowrates at A1, NE1 and NE2 manifolds when no measurements are enforced;

FIG. 13 illustrates flowrates at A1, NE1 and NE2 manifolds when measurement at A1 is enforced;

FIG. 14 illustrates flowrates at A1, NE1 and NE2 manifolds when measurements at A1 and NE2 are enforced;

FIG. 15 illustrates flowrates at A1, NE1 and NE2 manifolds when measurements at A1, NE1 and NE2 are enforced; and

FIG. 16 is a flow chart of an exemplary method for determining a prospective sensor location for improving a virtual measurement.

DETAILED DESCRIPTION

Simulations can be used to estimate variables of processes that cannot be directly or indirectly measured by sensors, or are measured by sensors that are inaccurate (e.g., due to wear and tear). Estimation of process variables (referred to as virtual measurement) can be desirable, for instance, when installing and/or replacing a sensor in an oil field requires shutting down oil production (e.g., which can result in revenue losses). However, such simulations may be slow, inaccurate, and/or may not capture the operating principles of the process. Accordingly, systems and corresponding methods for improved virtual measurement are provided. Improvements in virtual measurement process can be achieved, for example, based on the realization that oil wells in an oil field (or an oil production facility) can be interconnected by a system of pipelines, via oil and gas industrial machines (e.g., valves) and the like. Therefore, measurement from sensors distributed over the oil fields can be used to generate/calibrate predictive model for virtual measurements and/or determine uncertainty (or accuracy) associated with the virtual measurements.
Virtual measurements by predictive models can result in virtual process parameters indicative of various properties of an oil field (e.g., oil flow, manifold pressure, etc.). The oil field can include multiple clusters of oil wells. The output of the oil wells (e.g., oil, gas, water or a mixture thereof) can be connected via a system of pipelines. For example, output of oil wells in a cluster can be transferred to a cluster manifold where the various outputs can be combined and/or separated into oil, gas and water. Sensors (e.g., pressure sensors, flow sensors, etc.) can be deployed at various locations in the oil fields to detect pressure and flow of output from an oil field (e.g., oil output).
Virtual measurement and sensor measurement for a given oil field parameter may differ. This can result from measurement errors associated with predictive models, measurement sensors or both. For example, sensors deployed in the oil field may be old and may not provide accurate measurement. Additionally or alternately, predictive models may not be calibrated, which can result in inaccurate virtual measurements. Errors in sensor and virtual measurements (or discrepancy between them) can result in erroneous determination of oil production from an oil field and can lead to loss in revenue. Therefore, it is desirable to develop a predictive model that can improve the measurement accuracy of oil production (e.g., by calculation of virtual process parameters).
FIG. 1 is a schematic illustration of an oil field 100. The oil field 100 can include oil clusters 110 and 120 comprising multiple oil wells. The oil cluster 110 can include multiple oil wells 112, 114, 116, and the output of the oil wells (e.g., a multiphase fluid including oil, gas and water) can be transferred to a cluster manifold 118 via pipes 102, 104 and 106. The oil cluster 120 can include multiple oil wells 122, 124, 126, and the output of the oil wells (e.g., a multiphase fluid including oil, gas and water) can be transferred to a cluster manifold 128 via pipes 132, 134 and 136. At the cluster manifold 118, outputs of the oil wells 112, 114, 116 can be combined. The combined outputs from clusters 110 and 120 can be transferred to a cluster manifold 148 via pipes 142 and 144, respectively. Output from the cluster manifold 148 can be transferred to downstream facilities (e.g. gas processing facilities, oil facilities, etc.). The manifolds 118, 128 and 148 can include a separator that can separate various components of the multiphase fluid (e.g., oil, gas and water). For example, cluster manifold 148 can separate fluid output from wells upstream into an oil output 152, a gas output 154 and water output 156 (also referred to as hand off outputs). The oil output 152 can be transferred to an oil processing facility and the gas output 154 can be transferred to a gas processing facility. The oil and gas processing facilities can include export sensors (e.g., sensors that can be more accurate than the sensors in the oil field) that can detect the flow rates of one or more of the oil output 152, the gas output 154 and the water output 156.
During the initial phase of production, oil wells can be naturally flowing and fluid (e.g., oil) oozes out of the well due to pressure at the reservoir that can lift the oil naturally to the surface. The reservoir pressure can decrease, for example as the oil well ages, and an artificial lift mechanism (e.g., Electric submersible pumps, Gas Lift, Gas Injection, Rod Lift Pumps etc . . . ) may need to be used to extract oil. For example, the wells 112, 116 and 122, 126 can include pumps to extract oil. The wells can also include one or more flow sensors to measure the fluid output of the well, pressure sensors (e.g., to measure well head pressure) and sensors to detect the composition of the fluid output. These sensors can be located at one or more locations in the pipes (e.g., 102, 106, 132, 136, etc) and manifolds 118, 128 and 148.
It can be desirable to maintain a continuous production of oil (e.g., a predetermined flow of output from the cluster manifold 148) and prevent unplanned shutdowns. Replacing a sensor in the oil field that is producing inaccurate measurement can lead to downtime which is not be desirable. However, predictive models can be developed for the various sensors that can calculate virtual parameters associated with the sensors. In some implementations, virtual parameters can be calculated at a location where no sensor is present (e.g., virtual pressure detection at a location where no pressure sensor is present). The predictive models can be calibrated based on various sensor measurements in the oil field, physical model of sensors, physical model of oil wells, physical model of pipes, etc. Because the oil wells in the oil field are interconnected via a network of pipes, sensor measurement at various locations in the oil fields can be used to calibrate a predictive model (e.g., predictive model for a sensor measurement or a process) in the oil field (e.g., a predictive model of a sensor remote from the measurement location).
In some implementations, a manifold predictive model can be generated (e.g., for manifolds 118, 128, 148, etc.). The manifold predictive model can be generated based on predictive models of oil wells and pipes that are upstream (or downstream) from the manifold. For example, manifold predictive model for manifold 118 can be based on models associated with wells 112 116 and pipes 102 106. The manifold predictive model can also be based on one or more sensor measurements taken upstream from the manifold (e.g., change in pressure of a fluid and/or change in phase of the fluid flowing along a segment of a pipeline upstream from the manifold). In some implementations, the manifold predictive model can be based on (or calibrated) sensor measurements downstream from the manifold. In some implementations, the manifold predictive model can include a thermodynamic model based on inenthalpic mixing of the fluid outputs from the various wells upstream from the manifold. In some implementations, a manifold can include a separator that can separate fluid arriving at the manifold from the wells upstream from the manifold. For example, the separator can separate oil, gas and water from the multiphase fluid arriving at the manifold. In some implementations, the manifold predictive model can calculate the flow rates of oil, gas, and water that are obtained from the above-mentioned separation.
The oil production and transfer system (e.g., including oil wells, oil and gas industrial machines [e.g., valves], oil carrying pipelines, and the like) can route output from different wells to be combined and/or re-routed through different production facilities as desired. In some implementations, two types of measurements can be available for determining handoff flowrates (e.g., outputs 152-156) from oil fields. A first measurement can be obtained from a physical sensor that can have an uncertainty (e.g., 8%), and a second measurement can be obtained from hand allocation calculations. The discrepancy between these measurements can be problematic for customers. A system of systems modeling approach (e.g., generation of predictive model for a manifold) can enable an objective mathematically sound estimate for measurements in the oil field. Such a system of systems model can be based on physical equations associated with oil wells, surface flow networks (e.g., including pipelines in the oil field) to infer unknowns at one location (virtual measurements) using sensor measurements at other locations in the network.
In some implementations, sensor measurements can be performed at different locations in the oil field, but the measurements may not be accurate (e.g., due to errors in compositional information of oil well outputs, unexpected deterioration of oil well, an overdue calibration and neglect of instrumentation, etc.). Calibration of sensors can require dedicated access to the pipeline that can obstruct oil production.
Uncertainty in determining the true amount of hydrocarbons (e.g., oil, gas, etc.) that are exported through the pipes can result in lost revenue (e.g., in the tune of billions of dollars per year for a large oil field). For example, an error in measurement accuracy (e.g., estimated difference between the mean value of flow of hydrocarbon and the actual value of hydrocarbon flow) can result in lost revenue. Measurement uncertainty, on the other hand, can be indicative of a range of variation around an estimate. It can be desirable that the measurement uncertainty is small. Measurement uncertainty can be indicative of how the system is responding to valve position adjustments. Knowledge of measurement uncertainty can lead to insight into the state of different components of the network, degradation of instrumentation, etc.
Multiple sensors in the oil field that have poor accuracy. However, the fact that the different parts of the oil field are connected in a network can be used with the physical equations of various components of the oil field to make accurate estimates of production of oil and/or gas. The disparate measurements can be considered as data which when connected through a physical model of the network can provide “accurate” insights.
Sensor measurements can be made at certain cumulative locations of the production network (e.g., at a customer's facility that receives oil output 152). But these measurements may not be used to increase the accuracy of handoff flowrate estimates. FIG. 2 illustrates an exemplary estimation of handoff flowrate. The estimation can be based on computed uncertainty of three oil flowrate measurements at different manifolds (A1, NE1 and NE2) shown in FIG. 2.
FIG. 3 illustrates a representation of an exemplary well and lift model. The blue dots 302-310 in FIG. 3 refer to various well and the associated lift mechanism. The predictive models for the reservoir, pump, fluid flow through horizontal and vertical pipe sections, manifold pressure at surface can be built and used to characterize how the production would be affected by valve settings at the top of the well. The various measurement points 312-320 (e.g., sensor locations) are shown in FIG. 3. For example, the measurement points can be located at well-heads, manifolds (e.g., for detecting multiphase flow), etc. In some implementations, these measurements can be around 2-3% accurate and can feed into the facility where the compressors and flow components are split into a single phase flow where measurement accuracy is about 1%.
Reduced order models of the wells can be built and calibrated based on sensor measurements in the oil field. The reduced order models can be a drop-in replacement to perform several (e.g., thousands) of calculations for different production estimation scenarios. FIG. 4 illustrates exemplary gas flow virtual measurements of well models for naturally flowing reservoir and gas lifted wells.
The flow from different wells can be combined in a manifold and transported through surface pipelines. Pressure loss models and phase change models can predict the state of the fluid being transferred through the pipelines. Combine fluid flow models using manifolds and split flow can be generated. These models can utilize thermodynamics to model the inenthalpic mixing process to accurately assess the enthalpy of the inlet and exit streams along with the pressure, flow and composition of the fluid. In some implementations, once the different components of the oil field are connected, parameters of control (e.g., manifold pressures) and multiple measurement at different locations (e.g., well head and manifold flow measurements) can be used in a Monte Carlo sweep of the entire design space to understand/predict characteristics of the oil field. Monte Carlo sweep can include repeated random sampling to obtain numerical results. For example, randomness can be used to solve problems that may be deterministic in nature.
FIG. 5 illustrates an exemplary pair plot of manifold pressures (first five columns) versus the resulting production of liquid and gaseous components (shown in last two columns). For each of the simulation point in FIG. 5, the predictive model of the oil field can be evaluated to determine the virtual output at the flow meter locations (e.g., at various manifolds). FIG. 6 illustrates virtual output of flow meter determined for the simulation points in FIG. 5.
In some implementations, more than 10000 hybrid simulations of predictive models can be executed in the order of seconds to determine the data for FIGS. 5 and 6. These figures can provide information on sensitivities of the manifold pressures to overall production (e.g., oil and/or gas production) and/or the optimal instrumentation locations (e.g., sensor location) in order to achieve benefit (e.g., maximal benefit) from an accurate sensor measurement in the oil field.
While system of systems models (e.g., generated using predictive models of various components of the oil field) are great at estimating missing inputs, they may require several inputs (e.g., above a threshold value) to predict outputs. If no inputs are available, the model can be used to understand the most critical location where a sensor is needed. Missing sensor locations can be inferred from the sensitivity plots.
FIG. 7 illustrates an exemplary sensitivity plot for FIGS. 5 and 6. The sensitivity plot can be used in identifying one or more locations (e.g., most important location) for adding sensors in the oil field network. FIG. 7 can indicate that for estimating liquid production, it can be desirable to have a sensor measurement in A1 manifold (see FIG. 2). Such sensitivity charts can be useful for understanding relative importance of each measurement to the accuracy of handoff flowrate estimation.
In some implementations, system of systems models can be created using physics based models of networks in the oil field, oil wells, artificial lift equipment etc. Such physics based models can be converted to reduced order machine learning models for faster execution. Such physics or reduced order models can provide a frequentist view of the outputs (e.g., they provide a single set of outputs given a single set of inputs). In some implementations, Bayesian methods can be used to understand probabilistic relationships between different variables in the network and estimate uncertainties in estimates (e.g., in virtual measurement). Such probabilistic computations can be aided by the power of cloud computing which can bring unprecedented speed to such techniques.
FIG. 8 illustrates measurements and uncertainties associated with the measurements at the three manifolds identified in FIG. 2. The green lines 802, 804 and 806 at the center can be indicative of measurements performed on the day of interest. Green bars 812-816 and 822-826 at either side of the mean value are uncertainties based on historical values (e.g., historical sensor values, historical virtual measurements, etc.). With these three measurements and a reduced-order system of systems model of the entire network, one can estimate probabilistically the expected handoff flowrate associated with the manifold C in FIG. 2 (e.g., flow rate of oil output 152, gas output 154 water output 156, etc.).
FIG. 9 illustrates a probabilistic estimation 900 of a flowrate associated with the manifold C in FIG. 2. The center red line 902 is a measured value from handoff flowmeter (e.g., associated with the location C) and can serve as a reference. FIG. 9 illustrates that the estimated flow rate has a median value of 3498.5 barrel oil per day (BPD) with an uncertainty of 58 BPD or about 2% which is much smaller than uncertainty of current measurement of 8%. The uncertainty can be smaller if measurement uncertainties of individual flowmeters were lower. The approach of probabilistic estimation of median value can be better (e.g., more accurate) than simply summing up production rates from individual flow lines. While the above result indicate that existing sensors can be used with a probabilistic system of systems model to achieve better accuracy, such system of systems models also lead a number of other desirable outcomes.
In some implementations, if the sensor in the A1 manifold is not available, the system of systems model can be used to estimate measurements in this manifold. While this estimate can be less accurate and more uncertain than the previous scenario where an actual measurement is available in the manifold, it can allow for determination of a reasonable value (e.g., accuracy above a threshold value) using the rest of the network. The estimate for total handoff flowrate for oil however can have an uncertainty because of this missing measurement.
FIG. 10 illustrates a probabilistic estimation 1000 of a flowrate associated with the manifold A1 in FIG. 2. The uncertainty in this instance, is still smaller than 8% uncertainty in spite of no actual measurement at the manifold A1. FIG. 11 illustrates flowrates at A1 and NE1 manifolds of FIG. 2. FIG. 11 indicates that it may be possible to estimate handoff flowrate with one flowmeter at the NE2 manifold. The flowrates at A1 and NE1 can be estimated by the system of systems model.
FIG. 12 illustrates flowrates at A1, NE1 and NE2 manifolds when no measurements are enforced. FIG. 13 illustrates flowrates at A1, NE1 and NE2 manifolds when measurement at A1 is enforced. FIG. 14 illustrates flowrates at A1, NE1 and NE2 manifolds when measurements at A1 and NE2 are enforced. FIG. 15 illustrates flowrates at A1, NE1 and NE2 manifolds when measurements at A1, NE1 and NE2 are enforced.
Systems and methods described in this application can provide several advantages and novelty. In some implementations, oil field network architecture can allow for lego block style assembling of predictive models of components of the oil field. In some implementations, large-scale networks that include predictive models (e.g., Hybrid Physics Models) of components that works seamlessly with other connected components can be created. The concept of a network can be used to tie different disparate information (e.g., sensor measurements) across multiple phases, flow, pressure and temperature.
In some instances, the network model can solve the production estimation problem. Uncertainties across all or some nodes of the network can be rolled up (e.g., combined). The network predictive model can be calibrated to match physical reality and can then be executed in prediction mode. If the virtual measurement from the network predictive model and sensor measurement diverge, the network predictive model can be recalibrated. The network predictive model can suggest the best places of observability for instrumentation (e.g., works with partial instrumentation).
FIG. 16 is a flow chart of an exemplary method for determining a prospective sensor location for improving a virtual measurement. At 1602 a manifold predictive model can be generated. The manifold predictive model can be generated for (or associated with) a given manifold in the oil field (e.g., manifold A1, A2, A3, etc.). In some implementations, the manifold predictive model can be generated based on predictive models (e.g., predetermined) associated with one or more portions of the oil field. The manifold predictive model can be based on predictive models of manifolds and/or oil well that are coupled to the given manifold (e.g., manifolds/oil well from which oil/gas/water flows into the given manifold, manifolds/oil wells to which oil/gas/water flows out of the given manifold, etc.), physical equations of the coupled manifolds/oil fields and pipes coupling the manifolds/oil wells to the given manifold. For example, the manifold predictive model of manifold A1 in FIG. 2 can be based on predictive models, physical equations and sensor measurements at manifolds A2 and A3 (or manifolds/oil wells coupled to the manifolds A2 and A3).
In some implementations, the manifold predictive model can be used to calculate a probabilistic estimation indicative of an operating range of a flow rate associated with one or more of oil, gas and water from the first manifold. For example, as illustrated in FIG. 10, the manifold predictive model can calculate a probabilistic estimation of flowrate associated with manifold A1. The probabilistic estimation can include a range of flowrates at A1 and the probability associated with the various flow rates. Additionally or alternately, a median flowrate value and an uncertainty around the medial flowrate value can also be calculated. In some implementations, the calculation of probabilistic estimation of flow rates can be based on one or more sensor measurements at the given manifold (e.g., manifold A1). The sensor can allow for an actual measurement of flowrate which can be used to by the manifold model to generate the probabilistic estimation. Presence of a sensor measurement can reduce the uncertainty associated with the probabilistic estimation. Additionally or alternately, the probabilistic estimation can be used to determine if the sensor at the given manifold is operating properly. For example, if the sensor measurement and the probabilistic estimation do not match (e.g., the difference between the sensor measurement and the median of the probabilistic estimation is greater than a predetermined value [e.g., one or more standard deviations]), it can be determined that the sensor needs to be replaced or recalibrated. The manifold predictive model can be configured to calculate an initial virtual measurement associated with an oil field comprising a plurality of oil wells. The manifold predictive model can be based on one or more predictive models associated with one or more components of the oil field.
At 1604, data characterizing one or more pressure measurements and flow measurements obtained in the oil field can be received. At 1606, a prospective sensor location of a first prospective sensor in the oil field can be determined. The first prospective sensor can be configured to detect an oil field parameter. The manifold predictive model can be configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement. At 1608, the prospective sensor location and the identity of the first prospective sensor can be provided (e.g., in a graphical user interface display space).
Exemplary technical effects of the methods, systems, and devices described herein include, by way of non-limiting example, expediting the calculation of virtual measurement values, for example, due to parallelization of the simulation. Further, applying an iterative algorithm to the simulation of process flow algorithm can result in accurate and robust determination of virtual measurement values.
Certain exemplary embodiments are described herein to provide an overall understanding of the principles of the structure, function, manufacture, and use of the systems, devices, and methods disclosed herein. One or more examples of these embodiments are illustrated in the accompanying drawings. Those skilled in the art will understand that the systems, devices, and methods specifically described herein and illustrated in the accompanying drawings are non-limiting exemplary embodiments and that the scope of the present invention is defined solely by the claims. The features illustrated or described in connection with one exemplary embodiment may be combined with the features of other embodiments. Such modifications and variations are intended to be included within the scope of the present invention. Further, in the present disclosure, like-named components of the embodiments generally have similar features, and thus within a particular embodiment each feature of each like-named component is not necessarily fully elaborated upon.
Other embodiments are within the scope and spirit of the disclosed subject matter. One or more examples of these embodiments are illustrated in the accompanying drawings. Those skilled in the art will understand that the systems, devices, and methods specifically described herein and illustrated in the accompanying drawings are non-limiting exemplary embodiments and that the scope of the present invention is defined solely by the claims. The features illustrated or described in connection with one exemplary embodiment may be combined with the features of other embodiments. Such modifications and variations are intended to be included within the scope of the present invention. Further, in the present disclosure, like-named components of the embodiments generally have similar features, and thus within a particular embodiment each feature of each like-named component is not necessarily fully elaborated upon.
The subject matter described herein can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structural means disclosed in this specification and structural equivalents thereof, or in combinations of them. The subject matter described herein can be implemented as one or more computer program products, such as one or more computer programs tangibly embodied in an information carrier (e.g., in a machine-readable storage device), or embodied in a propagated signal, for execution by, or to control the operation of, data processing apparatus (e.g., a programmable processor, a computer, or multiple computers). A computer program (also known as a program, software, software application, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file. A program can be stored in a portion of a file that holds other programs or data, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers at one site or distributed across multiple sites and interconnected by a communication network.
The processes and logic flows described in this specification, including the method steps of the subject matter described herein, can be performed by one or more programmable processors executing one or more computer programs to perform functions of the subject matter described herein by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus of the subject matter described herein can be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).
Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processor of any kind of digital computer. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are a processor for executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile memory, including by way of example semiconductor memory devices, (e.g., EPROM, EEPROM, and flash memory devices); magnetic disks, (e.g., internal hard disks or removable disks); magneto-optical disks; and optical disks (e.g., CD and DVD disks). The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.
To provide for interaction with a user, the subject matter described herein can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, (e.g., a mouse or a trackball), by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well. For example, feedback provided to the user can be any form of sensory feedback, (e.g., visual feedback, auditory feedback, or tactile feedback), and input from the user can be received in any form, including acoustic, speech, or tactile input.
The techniques described herein can be implemented using one or more modules. As used herein, the term “module” refers to computing software, firmware, hardware, and/or various combinations thereof. At a minimum, however, modules are not to be interpreted as software that is not implemented on hardware, firmware, or recorded on a non-transitory processor readable recordable storage medium (i.e., modules are not software per se). Indeed “module” is to be interpreted to always include at least some physical, non-transitory hardware such as a part of a processor or computer. Two different modules can share the same physical hardware (e.g., two different modules can use the same processor and network interface). The modules described herein can be combined, integrated, separated, and/or duplicated to support various applications. Also, a function described herein as being performed at a particular module can be performed at one or more other modules and/or by one or more other devices instead of or in addition to the function performed at the particular module. Further, the modules can be implemented across multiple devices and/or other components local or remote to one another. Additionally, the modules can be moved from one device and added to another device, and/or can be included in both devices.
The subject matter described herein can be implemented in a computing system that includes a back-end component (e.g., a data server), a middleware component (e.g., an application server), or a front-end component (e.g., a client computer having a graphical user interface or a web browser through which a user can interact with an implementation of the subject matter described herein), or any combination of such back-end, middleware, and front-end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (“LAN”) and a wide area network (“WAN”), e.g., the Internet.
Approximating language, as used herein throughout the specification and claims, may be applied to modify any quantitative representation that could permissibly vary without resulting in a change in the basic function to which it is related. Accordingly, a value modified by a term or terms, such as “about” and “substantially,” are not to be limited to the precise value specified. In at least some instances, the approximating language may correspond to the precision of an instrument for measuring the value. Here and throughout the specification and claims, range limitations may be combined and/or interchanged, such ranges are identified and include all the sub-ranges contained therein unless context or language indicates otherwise.

Claims

What is claimed is:

1. A method comprising:

generating a manifold predictive model configured to calculate an initial virtual measurement associated with an oil field comprising a plurality of oil wells, the manifold predictive model based on one or more predictive models associated with one or more components of the oil field;

receiving data characterizing one or more pressure measurements and flow measurements obtained in the oil field,

determining a prospective sensor location of a first prospective sensor in the oil field, the first prospective sensor configured to detect an oil field parameter; and

providing the prospective sensor location and/or the identity of the first prospective sensor.

2. The method of claim 1, wherein the manifold predictive model is associated with a first manifold in the oil field, the first manifold coupled to a plurality of oil wells in the oil field, and wherein the one or more predictive models includes one or more of flow models associated with pipes connecting the first manifold to the plurality of oil wells, and physical equations and/or sensor measurements associated with one or more of the plurality of oil wells.

3. The method of claim 2, wherein the manifold predictive model is configured to calculate a probabilistic estimation indicative of an operating range of a flow rate associated with one or more of oil, gas and water from the first manifold.

4. The method of claim 3, wherein the calculation of probabilistic estimation of flow rates is based on one or more sensor measurements at the first manifold.

5. The method of claim 1, wherein the manifold predictive model is configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement.

6. The method of claim 5, wherein a first sensitivity associated with the updated virtual measurement is smaller than a second sensitivity associated with the initial virtual measurement.

7. The method of claim 1, wherein at least one of the one or more pressure measurements and flow measurements are obtained at a manifold of the oil field.

8. A system comprising:

at least one data processor;

memory coupled to the at least one data processor, the memory storing instructions to cause the at least one data processor to perform operations comprising:

9. The system of claim 8, wherein the manifold predictive model is associated with a first manifold in the oil field, the first manifold coupled to a plurality of oil wells in the oil field, and wherein the one or more predictive models includes one or more of flow models associated with pipes connecting the first manifold to the plurality of oil wells, and physical equations and/or sensor measurements associated with one or more of the plurality of oil wells.

10. The system of claim 9, wherein the manifold predictive model is configured to calculate a probabilistic estimation indicative of an operating range of a flow rate associated with one or more of oil, gas and water from the first manifold.

11. The system of claim 10, wherein the calculation of probabilistic estimation of flow rates is based on one or more sensor measurements at the first manifold.

12. The system of claim 8, wherein the manifold predictive model is configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement.

13. The system of claim 12, wherein a first sensitivity associated with the updated virtual measurement is smaller than a second sensitivity associated with the initial virtual measurement.

14. The system of claim 12, wherein at least one of the one or more pressure measurements and flow measurements are obtained at a manifold of the oil field.

15. A computer program product comprising a machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:

determining a prospective sensor location of a first prospective sensor in the oil field, the first prospective sensor configured to detect an oil field parameter;

16. The computer program product of claim 15, wherein the manifold predictive model is associated with a first manifold in the oil field, the first manifold coupled to a plurality of oil wells in the oil field, and wherein the one or more predictive models includes one or more of flow models associated with pipes connecting the first manifold to the plurality of oil wells, and physical equations and/or sensor measurements associated with one or more of the plurality of oil wells.

17. The computer program product of claim 16, wherein the manifold predictive model is configured to calculate a probabilistic estimation indicative of an operating range of a flow rate associated with one or more of oil, gas and water from the first manifold.

18. The computer program product of claim 17, wherein the calculation of probabilistic estimation of flow rates is based on one or more sensor measurements at the first manifold.

19. The computer program product of claim 15, wherein the manifold predictive model is configured to receive data characterizing the detected oil field parameter and generate an updated virtual measurement.

20. The computer program product of claim 19, wherein a first sensitivity associated with the updated virtual measurement is smaller than a second sensitivity associated with the initial virtual measurement.