Demulsifier analysis framework

Abstract

In one embodiment, a method may include receiving, by a demulsifier analysis framework, tensiometer data for an oil and water field sample including an added candidate demulsifier, where the tensiometer data includes rheology data regarding frequency, and wherein the demulsifier analysis framework includes a model derived based at least in part on bottle test data of the demulsifier in the oil and aqueous emulsion; using the tensiometer data as an input to a model of the demulsifier analysis framework to determine performance characteristics of candidate demulsifiers relative to the oil and water field samples without bottle testing of the oil and water field samples; and outputting, by the demulsifier analysis framework, performance characteristics of the candidate demulsifiers relative to the emulsion breaking the oil and water field samples.

Description

[0001] Related applications

[0002] This application claims priority to and the benefit of U.S. Provisional Application Serial No. 63 / 320,414, filed on March 16, 2022, which is incorporated herein by reference in its entirety. Background Art

[0003] A demulsifier is a chemical substance that can be used to break up an emulsion so that the emulsified phases can separate. An emulsion can be a dispersion of one immiscible liquid in another immiscible liquid, which can be formed by one or more processes. For example, the presence of one or more surface-active agents (surfactants), physical mixing, temperature, pressure, etc. are considered as one or more processes. Emulsions can be stable or unstable. Stable emulsions can include one or more chemicals, such as one or more surfactants, which reduce the interfacial tension (IFT) between the two liquids to achieve stability. Stable emulsions can pose challenges to the oil and gas industry, especially in processes where oil and water are to be separated.

[0004] Demulsifiers can be applied to various situations, such as separation processes that can include one or more separators. For example, consider a multiphase separator, which can be a container that separates well fluid into at least two types of liquids, i.e., oil and water. The type of demulsifier selected can depend on the type of emulsion (e.g., oil-in-water or water-in-oil). Demulsifiers are also found to be used in chemical analysis of oil and synthetic muds, and in processing hydrocarbons for production. Demulsifiers can play a role due to one or more different functional groups (e.g., amines, polyols, sulfonates, polymers, etc.). This functional group can provide different performance characteristics for different types of demulsifiers for different types of demulsification tasks.

[0005] As an example, demulsifiers can be applied to help ensure that the residual water and salt content in the crude oil and the oil content in the separated water meet certain specifications. Demulsifiers can disrupt the ordered structure of one or more natural surfactants that stabilize the emulsion, thereby enabling the dispersed droplets to coalesce and form larger droplets that can move to the oil / water interface (e.g., oil on water due to buoyancy).

[0006] Cost-effective demulsifiers may be formulated for one or more applications, facilities, etc. Various factors, such as injection location, oil and produced water characteristics, environmental requirements, and equipment and facility limitations may be evaluated in an effort to select or formulate a demulsifier and, for example, minimize operational disruptions and costs.

[0007] As mentioned above, demulsifiers can perform a variety of functions for various processes. In the case of water and oil emulsions, demulsifiers can facilitate the separation of water and oil, which in turn can improve downstream processing. Summary of the invention

[0008] A method may include receiving, by a demulsifier analysis framework, tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency, and wherein the demulsifier analysis framework includes a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions; using the tensiometer data as input to the model of the demulsifier analysis framework to determine performance characteristics of the candidate demulsifier relative to the oil and water field sample without performing a bottle test on the oil and water field sample; and outputting, by the demulsifier analysis framework, the performance characteristics of the candidate demulsifier relative to breaking an emulsion in the oil and water field sample. A system may include a processor; a memory operably coupled to the processor; and processor-executable instructions stored in the memory to direct the system to: receive tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency; use the tensiometer data as input to a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions to determine performance characteristics of the candidate demulsifier relative to the oil and water field sample without performing bottle testing of the oil and water field sample; and output the performance characteristics of the candidate demulsifier relative to the oil and water field sample. One or more computer-readable storage media may include processor-executable instructions to instruct a computing system to: receive tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency; use the tensiometer data as input to a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions to determine performance characteristics of the candidate demulsifier relative to the oil and water field sample without performing bottle testing of the oil and water field sample; and output performance characteristics of the candidate demulsifier relative to the oil and water field sample. Various other methods, systems, etc. are also described herein.

[0009] This Summary is provided to introduce a selection of some concepts that are further described below in the Detailed Description. This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to help limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The features and advantages of the described embodiments may be more readily understood by referring to the following description in conjunction with the accompanying drawings.

[0011] Figure 1 is a series of diagrams of examples of example environments and systems.

[0012] Figure 2 is a schematic diagram of an example of a system.

[0013] Figure 3 is a schematic diagram of an example of a system.

[0014] Figure 4 is a series of plots of example plots of tensiometer data.

[0015] Figure 5 is a schematic diagram of an example of a system.

[0016] Figure 6 is a diagram of the example workflow.

[0017] Figure 7 is a diagram of an example of a framework.

[0018] Figure 8 are diagrams of examples of methods and examples of systems.

[0019] Fig. 9 Example components of systems and networked systems are shown. DETAILED DESCRIPTION

[0020] The following description should not be taken in a limiting sense but is merely for the purpose of describing the general principles of the embodiments. The scope of the described embodiments should be determined with reference to the issued claims.

[0021] The selection and application of demulsifier chemicals in the oilfield often requires screening a large number of chemicals and bottle testing various mixtures of these chemicals to identify high-performing products. Bottle testing is a procedure in which different chemicals are added to bottle samples of emulsions to determine which chemical is most effective in breaking or separating the emulsion into oil and water. Once an effective chemical is identified, varying amounts of the chemical can be added to sample bottles of the emulsion to determine the minimum amount required to effectively break the emulsion.

[0022] Bottle test results can be expressed as water separated from the sample relative to time. For example, the sample can be placed in a tube and subjected to test conditions, including temperature, shaking, etc. As an example, consider a protocol that requires a certain amount of mechanical energy to be input into one or more standardized tubes containing a predetermined amount of sample (with or without demulsifier) ​​over a period of time. The tube may include a scale (e.g., a scale) that allows the separated aqueous phase to be read. Over time, unstable emulsions will separate so that the amount of separated water increases over time. In this method, a model can be used to characterize the demulsification of the emulsion (e.g., a linear model, a nonlinear model, etc.). An effective demulsifier can break the emulsion faster than the natural process, and note that a stable emulsion can remain stable for a considerable period of time (e.g., hours, days, etc.). By adding different levels of different demulsifiers, the best or effective demulsifier can be determined. The demulsifier can be a mixture of components, which may include one or more solvents and one or more so-called intermediates, where the intermediate is a demulsifier. Table 1 below shows an example of some bottle test data.

[0023] Table 1 Example of bottle test results.

[0024] serial number DM dose 2 minutes 4 minutes 6 minutes 8 minutes water Oil Int 1 none 0 - - - - - Br - 2 DM-A 50 0.15 1.4 6 18 C B B 3 DM-B 50 - 0.05 0.3 1.5 C B B 4 DM-C 50 - - 0.25 0.5 C Br B

[0025] In Table 1, the first column lists the bottle number, the second column lists the demulsifier tested, or "None" if the sample was tested without a demulsifier, the third column lists the dosage (e.g., in microliters, etc.), the fourth to seventh columns list the amount of free water at various times in appropriate volume units (e.g., milliliters), and the eighth to tenth columns list the subjective characteristics of the sample with respect to water, oil, and interface, including indicators such as "C" for clear, "B" for black or sagging, and "Br" for brown. For these listings, water may include clear and transparent, transparent, cloudy, water with oil, glass with oil, oil film on glass; oil may include black, brown, dark, light, and visible water droplets; and the interface may include sharp, smooth, sagging, broken, loose, and tapered. As an example, additional data may include, for example, longer retention time water droplets measured after 1 hour and 90 minutes and / or Karl-Fischer (KF) titration, which provides an estimate of the remaining water content in the oil phase. For example, for DM-A in Table 1, the KF result is 1%; however, without the demulsifier, the KF result is 40%. As for DM-A and DM-B, at 90 minutes, they both have a free water value of 30, while DM-B has a KF result of 0.8%. Therefore, DM-A and / or DM-B may be considered effective based on one or more factors that may be relevant to a field, where if time is a consideration, then DM-A may be selected over DM-B, and if the difference between 1% and 0.8% in the KF result is a consideration, then DM-B may be selected. The selection process may also consider cost, handling, impact on equipment, downstream processing, environment, etc.

[0026] As an example, a bottle test may be conducted at least in part in a subjective manner, where subjective information is available, which may be handwritten information on a piece of paper. As an example, the results on a piece of paper may be digitized, where information may be extracted for storage in a database. Such information may include data and metadata. As an example, data of water dropout relative to time may be extracted, and may be presented in the form of a plot (e.g., water dropout relative to time), which may be represented as an image, a model (e.g., a rate model), etc. For example, consider the generation of pixel image data of water dropout relative to time, where such pixel image data may be used to train one or more machine learning models (ML models) that may perform image analysis. As an example, the generated image may include a hybrid image formed using bottle test data and other data such as tensiometer data, where, for example, the ML model may receive the tensiometer data and output corresponding bottle test data, which may be synthetic bottle test data based on the training bottle test data. In this approach, the framework may utilize an ML model of the image to characterize one or more candidate demulsifiers. This approach can be similar to training an ML model for plants, where the plants can have associated neighboring plants, such as in a plant ecosystem, where the training data can include images of the plants and associated neighboring plants, and where the trained ML model can be provided with individual plant images to output representations of the associated neighboring plants, which can be taken from a database and / or synthetically generated. As an example, various types of data can be represented in the form of images, which can be images of plots (e.g., a plot of bottle test data of water drops versus time, a plot of tensiometer data of properties versus frequency, etc.).

[0027] As explained, the bottle test can include information of one or more demulsifiers (e.g., individual and / or formulations of multiple demulsifiers), which can be referred to as intermediates in demulsifier solutions with solvents. The dosage or concentration can vary. As shown in Table 1, the water dripping out can be read from the scale (e.g., scale) on the bottle and recorded at two-minute intervals. Similarly, in Table 1, for the oil field sample from the Middle East oil field, DM-A in bottle No. 2 performed well, showing 18 ml of water dripping out after 8 minutes.

[0028] As an example, the framework may include features for digitizing bottle test data and / or for receiving digital files of bottle test data that may be input through a graphical user interface (GUI). Such a framework may provide for organizing bottle test data, for example, about oil fields, types of oil, brine properties, dosages (e.g., concentrations), time, and types of one or more demulsifiers (e.g., in a formulation). The framework may also provide for the collection of tensiometer data, and may provide, for example, the collection of field data about separator performance. One or more databases may be generated with data sufficient to establish one or more models, which may utilize tensiometer data, which may include rheological data of candidate demulsifiers (e.g., potential demulsifiers) as input to generate one or more demulsifier performance characteristics of the candidate demulsifiers as output. In such an example, the output may be generated without having to perform a bottle test on the candidate demulsifier, which, as explained, may take considerable time in setting up the bottle test and running the bottle test (e.g., considering that the setup and execution of the bottle test in Table 1 takes more than an hour). As an example, the framework can accelerate candidate demulsifier performance characterization, which can be performed relatively quickly using tensiometer data obtained using a tensiometer that generates digital data that can be automatically input into one or more models (e.g., a user does not have to write down the information on paper). The tensiometer can be equipped with an oilfield sample brine and an oilfield sample crude oil, into which one or more demulsifiers are introduced, and produce results at the push of a button or other controller.

[0029] Tensiometer data can be reproducible objective data; however, as explained, bottle test data can include subjective data that can vary from person to person and can vary depending on a person's ability to read the scale on the bottle regarding fluid layers (e.g., phases, interfaces, etc.). Given a sufficiently large amount of bottle test data, outliers can be identified and optionally excluded, and, for example, metrics (e.g., probabilistic, statistical, etc.) regarding quality and / or range can be generated. As an example, a framework can receive tensiometer data including rheological data and output one or more performance characteristics, optionally with one or more confidence intervals, etc. As for demulsifier efficiency as a performance characteristic, it can be defined as the volume of water separated divided by the initial volume of water in the emulsion of oil and water oilfield sample. As shown in Table 1, demulsifier efficiency can be evaluated by observing the volume of water that drips out of the emulsion; note that the original water content in the emulsion can vary. As an example, the performance characteristics of a demulsifier can be relative, for example, relative to the performance characteristics of one or more other demulsifiers. As an example, the framework may provide a ranking of candidate demulsifiers, such as relative to one or more other demulsifiers, which may be other candidates and / or demulsifiers having corresponding data in a database on which the framework is integrated or otherwise built.

[0030] As mentioned previously, bottle testing can be used to select effective demulsifiers (such as high-efficiency demulsifiers). After performing bottle test demulsifier screening to select a short list of test candidates, the final demulsifier selection can be made after large-scale field testing of the short list demulsifiers in actual oilfield process scenarios. Although time-consuming, bottle testing plays a role in selecting the best demulsifier formulation to reduce the time and cost of field testing.

[0031] As an example, demulsifiers can be selected based in part on one or more residence times in an oilfield process. For example, residence times can be determined for various flows of various fluid handling equipment. As an example, the residence time of a conduit leading to a separator can be determined, wherein the conduit can have an injection port for injecting demulsifiers. As an example, the injection port position can be based on the residence time of a conduit and / or one or more other equipment, and, for example, the performance characteristics of one or more demulsifiers, which can contribute to the destruction of the emulsion produced. As an example, a demulsifier analysis framework can include one or more features for evaluating the performance in a specific field. For example, a framework that can simulate fluid flow and / or separation is considered, so that a demulsifier that can be appropriately implemented in a specific field can be selected. In this method, one or more types of simulators can be included in the framework and / or can be operably connected to the framework. For example, consider the PIPESIM framework (SLB, Houston, Texas), which can perform fluid flow simulation of multiphase fluids and can be integrated with the DELFI environment (SLB, Houston, Texas).

[0032] The PIPESIM simulator includes a solver that can provide simulation results, such as multiphase flow results (e.g., from the reservoir to the wellhead and beyond the wellhead, etc.), flow line and surface facility performance, etc. The PIPESIM simulator can be integrated with the AVOCET Production Operations Framework (SLB, Houston, Texas). As an example, the PIPESIM simulator can be an optimizer that can optimize one or more operating scenarios at least in part by simulating physical phenomena (e.g., considering separation and / or mixing of phases). The PIPESIM simulator can simulate various types of equipment, such as chokes, safety valves, separators, and chemical injectors. The PIPESIM simulator has several separator modeling options, including the following: separations defined with rigorous PIPESIM simulator engine flash calculations for various types of fluid definitions based on field pressure and temperature conditions; online separators (downhole or surface) that discard liquids, gases, or water (e.g., using specific separation efficiencies); network separators that track product and separation flows by adjusting the pressure continuity across the separators to allow boundary conditions for each outlet flow to match; and separator configurations arranged in series and in parallel to simulate conditions in multi-stage separation queues. As an example, demulsifier performance characteristics can be received by the PIPESIM simulator to simulate the performance of one or more separation processes, thereby generating simulation results, where one or more control actions can be taken in the field based on such results. In such an example, a demulsifier analysis framework can be operably connected to a PIPESIM simulator (e.g., a computing framework for performing oilfield process simulations).

[0033] The DELFI Environment, or DELFI Cognitive Exploration and Production (E&P) Environment, is a secure, cognitive, cloud-based collaborative environment that integrates data and workflows with digital technologies such as artificial intelligence and machine learning.

[0034] As an example, a model-based scientific approach can be followed, where demulsifier chemicals are surfactants that aggregate at the interface between the oil phase and the water phase and modify the interface. Interfacial properties such as interfacial tension (IFT) and interfacial viscosity have been identified as playing a crucial role in emulsion stability. See, for example, Berger et al., "Designing and Selecting Demulsifiers for Optimum Field Performance on the Basis of Production Fluid Characteristics", SPE Production Engineering 1988, 522-526, hereinafter referred to as Berger. It has been noted that there is a common misunderstanding about the role of IFT in emulsion stability, and in some cases reducing IFT is beneficial to the stability of the emulsion. See, e.g., Salager et al., "Surfactant-Oil-Water Systems Near the Affinity Inversion, Part I: Relationship Between Equilibrium Phase Behavior and Emulsion Type and Stability," J. Disp. Sci. Technol. 3, 1982, 279-92; Rosano et al., "Considerations on Formation and Stability of Oil / Water Dispersed Systems," JAOCS 59, 1982, 360-63; Graciaa et al., "Emulsion Stability and Phase Behavior for Ethoxylated Nonyl Phenol Sulfates", J. Colloid InterfaceSci.89, 1982, 217-25; and Bourrel et al., "The Relationship of Emulsion Stability to Phase Behavior and Interfacial Tension of Surfactant Systems", J. Colloid Interface Sci.72,1979,161-63.

[0035] As an example, reducing interfacial viscosity may be a possible mechanism by which demulsifiers break emulsions; note that relative solubility values, commonly used as a criterion for demulsifier evaluation, actually show no correlation with demulsifier performance, as tested in the 2400 field bottle test. See Berger. In various cases, a reduction in IFT itself is not a reliable indicator of demulsification performance, as it may simply mean rapid migration to the oil / water interface, where the mode of action depends on the structure and other components present at the interface. Therefore, a scientific approach is needed, including a comprehensive multivariate analysis of various measurable interfacial properties and their correlation with actual demulsifier performance, as inferred by bottle tests, which is not available in the current open literature. See, for example, Poindexter et al., "Applied Statistics: Crude Oil Emulsions and Demulsifiers", Journal of Dispersion Science and Technology 25, 2004, 311-320.

[0036] As an example, a demulsifier analysis framework may include a processor and a memory including processor executable instructions that may be executed to receive demulsifier data and analyze the data to determine the efficiency of the demulsifier. In such an example, the framework may access a database that may be created to provide correlations with the efficiency of one or more known demulsifier chemicals, such as may be inferred from bottle tests of interfacial properties of oil / water systems. In such an example, the data may include tensiometer data, such as data from an oscillating pendant drop tensiometer. Such data may include various types of rheological data, for example, considering one or more types of moduli (e.g., viscosity, elasticity, viscoelasticity, etc.).

[0037] As an example, the framework may provide for building and / or accessing a database. For example, consider a method that can correlate actual field efficiencies of demulsifier intermediates and commercial products inferred from standard bottle tests with interfacial property measurements. In such examples, correlation of various interfacial properties at the oil / water interface, such as interfacial tension (IFT), viscoelastic modulus (E), elastic modulus (E'), and viscous modulus (E"), with standard bottle test results can provide for the generation of a database that in turn facilitates rapid screening of new intermediates for their demulsifier efficiency by way of rapid and easy interfacial property measurements. Such a method can include obtaining interfacial properties via a drop shape analyzer tensiometer that allows for the measurement of oscillating droplets to infer the viscoelastic properties of oil droplets in an aqueous bulk phase. An estimate of demulsifier efficiency can be obtained, for example, from a regression model that relates the efficiency measured in the bottle test to the interfacial properties measured at the oil / water interface. In one embodiment, the interfacial properties can be the interfacial tension, viscoelastic modulus, elastic modulus, and viscous modulus measured at the oil / water interface before and after the addition of the corresponding demulsifier chemical to the analysis. In one or more other embodiments, one or more alternative interfacial or bulk properties can be used to estimate demulsifier efficiency.

[0038] As an example, the database can be used to train one or more machine learning models (ML models). For example, consider a trained ML model that can receive tensiometer data (e.g., including IFT and one or more moduli) of a demulsifier, and can output the efficiency of the demulsifier. As an example, the framework can operate with or without the ML model. For example, the framework can operate using the database or using one or more ML models that can be trained using data from the database.

[0039] As an example, the emulsion and / or a sample from the emulsion may be obtained in the field, for example, from one or more production wells. In such an example, the sample may be transported to a facility that includes a tensiometer, which may be an instrument that is sensitive to external vibrations. As such, the tensiometer may be located in a facility that is substantially isolated from external vibrations and / or mounted using one or more anti-vibration (e.g., vibration damping) mechanisms.

[0040] Various types of tensiometers are commercially available. Consider, for example, a hanging drop tensiometer, which may be particularly useful for relatively high interfacial tensions, although measurements down to about 1 mN / m can be made. In a hanging drop tensiometer, a drop of a denser liquid is suspended at the end of a syringe needle, where the drop is large enough that its shape is deformed by gravity, but not large enough to break away from the syringe. In this approach, its shape can be determined by the balance of interfacial tension and gravity. The interfacial tension can be derived from the drop shape and the densities of the two liquids. This approach can also be applied to drops of low-density liquids by inverting the syringe. For example, consider a drop of a low-density phase formed at the end of a syringe, facing the surrounding high-density phase in the direction of gravity. In such an example, buoyancy can cause the drop of the low-density phase to orient upward from the end of the syringe.

[0041] As an example, a tensiometer can include one or more oscillating mechanisms that can provide energy that can cause a force to impact a droplet. In such an example, the droplet can be subjected to a force that can react, which can allow one or more rheological properties to be obtained. For example, consider a tensiometer that can generate vibrations at one or more frequencies to obtain rheological property data. In such an example, a plot of modulus versus frequency of emulsion components can be generated, the emulsion can include one or more demulsifiers, which can be at one or more levels, ratios, etc.

[0042] As for viscoelasticity, these can relate to interfacial rheology, such as the viscoelastic behavior of the interface and stress relaxation. As an example, rheological data can include viscosity data, elastic data, or viscosity and elastic data. In a simple system of water and air, the IFT is a constant and is independent of the area being changed. However, if a surfactant is present, the physics can be quite complex. Surfactants can adsorb at the interface and form a layer. In equilibrium, the molecules or particles occupy a certain amount of space at the interface with an average distance from each other. The resulting equilibrium interfacial concentration is characteristic of the system and depends on parameters such as the participating phases, the interfacial active components and their volume concentrations. The interfacial active components react to deviations from equilibrium when the available interfacial area is changed by changing the volume of the droplets or by moving a barrier at the interface.

[0043] When the interfacial area decreases, the interfacial concentration increases and the system attempts to reach its equilibrium by desorbing the active species from the interface. When the interfacial area increases, the interfacial concentration decreases and the equilibrium state is restored by adsorbing the active species.

[0044] As the interfacial concentration of active components changes, the interfacial tension between the phases changes accordingly. For example, when the surfactant concentration at the interface increases by reducing the available area, the interfacial tension also decreases. If the area increases, the interfacial tension also increases. The reaction of active components at the interface can occur at different rates. This may depend on various factors, such as the mobility of the components inside the bulk phase, the bulk concentration, and the interfacial energy between the phases.

[0045] As an example, the interfacial behavior can be described by the complex viscoelastic modulus E* (or E), which includes an elastic part E' and a viscous part E": E* = E' + iE". The complex viscoelastic modulus can be measured by continuously reducing and increasing the interfacial area in an oscillating manner. When the interfacial area varies sinusoidally, the interfacial tension also varies sinusoidally. These two sinusoidal waves are phase-shifted. The phase shift depends on the reaction speed of the interfacial active components. The faster the interfacial active components react by adsorption and desorption to the interface, the larger the phase shift becomes. The elastic modulus E' and the viscous modulus E" can be calculated from the sinusoidal area and tension changes:

[0046] E'=Δσ(A o / ΔA)cosφ

[0047] E"=Δσ(A o / ΔA)sinφ

[0048] Above, Δσ and ΔA are the peak-to-peak amplitudes of interfacial tension and interfacial area, Ao is the average interfacial area, and φ is the phase shift.

[0049] As explained above, the tensiometer may include features capable of measuring the viscoelastic modulus. For example, using an optical contact angle goniometer and a drop shape analysis system, an oscillating pendant droplet may be generated. As an example, using a spinning drop video tensiometer, the rotational velocity of a spinning droplet may be oscillated, or using a dynamic contact angle measurement device and a force tensiometer, the position of a barrier of a Langmuir trough may be oscillated. Once the viscoelastic behavior is optimized, the effect on emulsion stability or instability may be analyzed, for example, with a dispersion stability analysis system.

[0050] As explained previously, the selection and application of demulsifier chemicals to ensure improved control of oil / water emulsions in oilfield operations is often a complex and tedious task. Many factors of chemical and physical properties directly affect the stability of the emulsion and thus the separation process. Demulsifier formulations usually need to be specifically selected for each respective process system. Due to the large number of interacting factors, representative testing and screening of suitable demulsifiers often follows a holistic approach, taking into account many variables. However, to date, there is no universal screening test other than the bottle test, which in turn is not standardized due to its individual customization for specific oilfield situations. Therefore, bottle tests are performed at different locations following different procedures. Bottle tests require various individual steps, such as manually shaking the emulsion of various products in the bottle, and visual observation of multiple interfacial properties, such as phase clarity and smoothness of the interface, and are therefore also highly subjective and require experienced experts. Replacing this tedious practice with automated and objective volumetric or interfacial measurements allows for the rapid screening of multiple intermediates and formulations based on their potential demulsification efficiency, which is needed in industry.

[0051] Considering that emulsion stability is often strongly defined by interfacial properties, which are physical quantities measurable using tensiometers, the demulsifier analysis framework can be used to rapidly screen demulsifier chemistry efficiency based on tensiometer interfacial measurements.

[0052] Figure 1 Examples of environments 101 are shown, including marine environment 102 and land environment 104, wherein marine environment 102 includes various equipment and land environment 104 includes various equipment. As shown, each environment 101 can include one or more wellheads 106 (e.g., wellhead equipment). The wellhead can be a ground terminal of a wellbore, which can include a system of reels, valves, and matching adapters, which can, for example, provide pressure control of a production well. The wellhead can be on the surface of land, on the surface of the seafloor (e.g., the seafloor, etc.), etc. As an example, conduits from multiple wellheads can be connected at one or more manifolds so that fluids from multiple wells can flow in a common conduit.

[0053] At different times, a well can be tested using a process known as well testing. During well testing, reservoir fluids can be produced at different rates at the separator, for example, according to a predetermined schedule. Such testing may take less than two days to evaluate a single well, or may take several months to evaluate a reservoir range. Well testing may include one or more of a variety of well testing operations, wherein the test may include, for example, one or more of pressure recovery, pressure drop, pressure drop, injection, and interference. In various cases, fluids can flow from one or more wells to the ground, where the fluids are subjected to one or more well testing operations and produce waste (e.g., waste fluids), which must be properly handled, for example, according to circumstances, regulations, etc. For example, it is considered that waste fluids are loaded into tank trucks for transportation to facilities capable of handling waste fluids, and waste fluids are handled by burning (e.g., flaring), etc.

[0054] about Figure 1 101, consider well testing as an operation that can be performed, for example, using equipment shown in marine environment 102 and / or using equipment shown in terrestrial environment 104. As an example, the environment may be in the exploration, development, evaluation, etc. phase, where such environment includes at least one well that can produce well fluids (e.g., via natural pressure, via fracturing, via artificial lift, via pumping, via water injection, etc.). In such an environment, various types of equipment can be on site, which can be operably connected to the well testing equipment.

[0055] Figure 1 An example of a system 110 is shown that can be operably coupled to one or more conduits that can, for example, transport well fluids from one or more wellheads. As shown, the system 110 can include a computing system 111 (CS) that can include one or more processors 112, a memory 114 accessible to at least one of the one or more processors 112, instructions 116 that can be stored in the memory 114 and executable by at least one of the one or more processors 112, and one or more interfaces 118 (e.g., wired, wireless, etc.). Figure 1 In the example of , system 110 is shown to include various communication symbols, which can be used for transmission and / or reception of information (e.g., data, commands, etc.), such as to and / or from computing system 111. As an example, computing system 111 can be a controller that can issue control instructions to one or more devices in an environment (e.g., marine environment 102 and / or terrestrial environment 104). As an example, computing system 111 can be local, can be remote, or can be distributed such that it is partially local and partially remote.

[0056] Referring again to wellhead 106, it may include various types of wellhead equipment, such as casing and tubing heads, Christmas trees, blowout preventers, etc. Fluid produced from the well may be routed through wellhead 106 and into system 110, which may be configured with various features for well testing operations.

[0057] exist Figure 1 In the example of FIG. 1 , system 110 is shown to include various segments, which may be categorized operationally. For example, consider well control segment 120 , separation segment 122 , fluid management segment 124 , and combustion segment 126 .

[0058] like Figure 1 As shown in the example, the well control section 120 is an assembly of various components, such as a manifold 130, a throttling manifold 132, a manifold 134, a heat exchanger 136, and an instrument 138; the separation section 122 includes a separator 142; the fluid management section 124 is an assembly of various components, such as a manifold and a pump 144, a manifold 146-1, a manifold 146-2, a tank 148-1, and a tank 148-2; and the combustion section 126 includes a burner 152 and one or more cameras 154 (e.g., for observing combustion, etc.).

[0059] As mentioned above, in Figure 1 In the example of , system 110 includes various features for one or more aspects of well testing operations; note that system 110 may include fewer features, more features, alternative features, etc. For example, consider one or more of a gas density meter, a water cut meter, a gas-oil ratio sensor, a carbon dioxide sensor, a hydrogen sulfide sensor, or a contraction measurement device. The various features may be upstream and / or downstream of a separator section or separator.

[0060] With respect to fluid flow from a well or multiple wells, such fluid may be received by the well control section 120 and then routed to the separation section 122 via one or more conduits. Figure 1 In an example of well control section 120 , heat exchanger 136 may be provided as a steam-heat exchanger, and meter 138 for measuring fluid flow through well control section 120 .

[0061] As described above, the well control section 120 can deliver fluids received from one or more wells to the separator 142. As examples, the separator 142 can be a horizontal separator or a vertical separator, and can be a two-phase separator (e.g., for separating gas and liquid) or a three-phase separator (e.g., for separating gas, oil, and water). The separator can include various features that help separate the components of the incoming fluid (e.g., diffusers, demisters, vanes, baffles, settlers, etc.).

[0062] As an example, a fluid may be a single-phase or multi-phase fluid, where "phase" refers to immiscible components (eg, consider two or more of oil, water, and gas).

[0063] As an example, the separator 142 can be used to substantially separate a multiphase fluid into its three phases of oil, gas, and water, as appropriate and as the case may be, wherein each phase emerging from the separator 142 can be referred to as a separated fluid. Such separated fluid can be routed away from the separator 142 to the fluid management section 124. In various cases, the separated fluid may not be completely homogeneous. For example, the separated gas leaving the separator 142 may include some residual amount of water or oil, and the separated water leaving the separator 142 may include some amount of oil or entrained gas. Similarly, the separated oil leaving the separator 142 may include some amount of water or entrained gas.

[0064] like Figure 1 As shown in the example of FIG. 1 , the fluid management section 124 includes flow control equipment such as various manifolds and pumps (generally represented by box 144) for receiving fluid from the separator 142 and delivering the fluid to other destinations, as well as additional manifolds 146-1 and 146-2 for routing fluid to and from fluid tanks 148-1 and 148-2. Figure 1 1 and 146 - 2 and two tanks 148 - 1 and 148 - 2 are depicted, but it should be noted that the number of manifolds and tanks may vary. For example, in one embodiment, the fluid management section 124 may include a single manifold and a single tank, while in other embodiments, the fluid management section 124 may include more than two manifolds and more than two tanks.

[0065] As for the manifolds and pumps 144, they may include various manifolds and pumps, such as gas manifolds, oil manifolds, oil transfer pumps, water manifolds, and water transfer pumps. In at least some embodiments, the manifolds and pumps 144 may be used to route the fluid received from the separator 142 to one or more fluid tanks 148-1 and 148-2 via one or more additional manifolds 146-1 and 146-2, and route the fluid between the tanks 148-1 and 148-2. As an example, the manifolds and pumps 144 may include features for routing the fluid received from the separator 142 directly to one or more burners 152 to burn gas and oil (e.g., bypassing the tanks 148-1 and 148-2) or for routing the fluid from one or more tanks 148-1 and 148-2 to one or more burners 152.

[0066] As described above, the components of the system 110 may vary between different applications (e.g., operations, etc.). As an example, the equipment within each functional group of the system 110 may also vary. For example, the heat exchanger 136 may be provided as part of the separation section 122 rather than as part of the well control section 120.

[0067] In certain embodiments, system 110 may be a surface well testing system capable of remote monitoring and control. Remote monitoring may be accomplished through sensors mounted on various components. In some cases, the monitoring system (e.g., sensors, communication systems, and human-machine interfaces) may be capable of monitoring one or more of segments 120, 122, 124, and 126.

[0068] Figure 2 An example of a system 250 is shown, which may be referred to as a surface well testing system. The system 250 may include Figure 1 Various features of system 110. Various equipment of system 250, such as separation equipment, may be present at one or more types of locations (eg, production locations, well testing locations, etc.).

[0069] exist Figure 2 In FIG. 2 , the multiphase fluid (here represented by arrows 252) enters flow head 254 and is routed to separator 270 through surface safety valve 256, steam-heat exchanger 260, throttling manifold 262, flow meter 264 and additional manifold 266. Figure 2 In the example of FIG. 2 , the system 250 includes a chemical injection pump 258 for injecting chemicals into the multiphase fluid flowing to the separator 270. For example, consider injecting one or more demulsifiers upstream of the separator 270 and / or in the separator 270.

[0070] exist Figure 2 In the illustrated embodiment, the separator 270 is a three-phase separator that generally separates the multiphase fluid 252 into gas, oil, and water components. The separated gas is routed downstream from the separator 270 through a gas manifold 274 to either of the burners 276-1 and 276-2 for combusting the gas and burning the oil. The gas manifold 274 includes valves that can be actuated to control the flow of gas from the gas manifold 274 to one or the other of the burners 276-1 and 276-2.

[0071] As shown, the oil separated from the separator 270 can be routed downstream to the oil manifold 280. The valves of the oil manifold 280 can be operated to allow the oil to flow to either of the burners 276-1 and 276-2 or to either of the tanks 282 and 284. The tanks 282 and 284 can be of any suitable form, but are preferably Figure 2280 as vertical surge tanks, each tank having two fluid compartments. This allows each tank to simultaneously contain different fluids, such as water in one compartment and oil in another compartment. An oil transfer pump 286 can be operated to pump oil through the well testing system 250 downstream of the separator 270. The separated water from the separator 270 can be similarly routed to the water manifold 290. Like the oil manifold 280, the water manifold 290 includes valves that can be opened or closed to allow water to flow to either of the tanks 282 and 284 or the water treatment and disposal device 294. The water transfer pump 292 can be used to pump water through the system.

[0072] In various cases, in the case of an emulsion in the production fluid from the reservoir, one or more chemicals can be added to break up the emulsion. For example, consider adding one or more demulsifiers. When separator 270 operates more efficiently, system 250 can operate more efficiently. For example, in the case of an emulsion, the separation of separator 270 may be adversely affected so that the separation of oil and water cannot occur effectively (e.g., with respect to time, etc.). As explained, demulsifiers can be added to help the separation by separator 270 to be more effective, for example, so that more water (e.g., brine) in the production fluid can be removed from oil (e.g., crude oil) in separator 270.

[0073] A well testing area in which well testing system 250 (or other embodiments of a well testing system) is installed may be classified as a hazardous area. In some embodiments, the well testing area is classified as a Zone 1 hazardous area according to International Electrotechnical Commission (IEC) standard 60079-10-1:2015.

[0074] exist Figure 2 In the example of FIG. 2 , a cabin 296 at a well site may include various types of equipment to acquire data from the well testing system 250. The acquired data may be used to monitor and control the well testing system 250. In at least some cases, the cabin 296 may be located in a non-hazardous area separate from the well testing area having the well testing system 250. This is accomplished by Figure 2 The dashed line 298 in FIG. 29 is shown as a demarcation line generally used as a demarcation line between a hazardous area having the well testing system 250 and a non-hazardous area of ​​the compartment 296 .

[0075] The equipment of the system may be monitored during the process to verify proper operation and facilitate control of the process. Such monitoring may include taking a number of measurements, examples of which may include throttle manifold temperature and pressure (upstream and downstream), heat exchanger temperature and pressure, separator temperature and pressure (static and differential), oil flow and volume from the separator, water flow and volume from the separator, and liquid levels in the system's tanks.

[0076] As an example, a mobile monitoring system may be provided. In such an example, monitoring of a process may be performed on a mobile device (e.g., a mobile device suitable for use in a Zone 1 hazardous area, such as a well testing area). Various types of information may be automatically acquired by the sensors and then presented to an operator via the mobile device. The mobile monitoring system may provide various functions, such as sensor data display for quality assurance and quality control purposes, video display, interpretation of sensor or video information, and manual input screens (e.g., a digital tally book for recording measurements taken by an operator).

[0077] As an example, a site or facility may include one or more computing devices. For example, consider a programmable logic controller (PLC), a gateway device, etc. As for a gateway device or a simple gateway, it may include one or more features of an AGORA gateway (e.g., v.202, v.402, etc.) and / or another gateway. For example, consider using an INTEL ATOM E3930 or E3950 dual-core processor, equipped with DRAM and eMMC and / or SSD. Such a gateway may include a trusted platform module (TPM), which may provide secure and measured boot support (e.g., through hashing, etc.). The gateway may include one or more interfaces (e.g., Ethernet, RS485 / 422, RS232, etc.). As for power, the gateway may consume less than about 100W of power (e.g., consider less than 10W or less than 20W). As an example, the gateway may include an operating system (e.g., consider LINUX DEBIAN LTS). As an example, the gateway may include a cellular interface (e.g., 4G LTE with global modem / GPS, etc.). As an example, the gateway may include a WIFI interface (e.g., 802.11a / b / g / n). As an example, the gateway may operate using 100-240 volts AC, 50 / 60 Hz, or 24 volts DC. As for size, consider a gateway having a protective box with dimensions of approximately 10 inches by 8 inches by 4 inches (e.g., 25 cm by 20.3 cm by 10.1 cm). The gateway may be operably coupled to various devices at the site, as well as operably coupled to one or more networks, which may include cloud resources (e.g., cloud platform resources). As an example, the gateway may be an edge device that may be remotely controlled, accessed, etc.

[0078] As an example, a method can provide a correlation of actual field efficiencies of demulsifier intermediates and commercial products inferred from standard bottle tests with interfacial property measurements. For example, various interfacial properties at the oil / water interface, such as interfacial tension (IFT), viscoelastic modulus (E), elastic modulus (E'), and viscous modulus (E") can be correlated with the results of standard bottle tests to generate a database that facilitates rapid screening of existing and / or new intermediates for demulsifier efficiency. As explained, a demulsifier analysis framework can receive input regarding interfacial property measurements, and can generate output regarding demulsifier performance (e.g., demulsifier efficiency).

[0079] As an example, a method may include obtaining interfacial properties using a drop shape tensiometer that allows for measurement of oscillating droplets to infer the viscoelastic properties of an oil droplet in a bulk phase of an aqueous solution. As an example, such a method may estimate demulsifier efficiency, for example, from a regression model that relates the efficiency measured in a bottle test to interfacial properties measured at the oil / water interface. In one embodiment, the interfacial properties may include interfacial tension, viscoelastic modulus, elastic modulus, and viscous modulus measured at the oil / water interface before and after addition of a corresponding demulsifier chemical. In various embodiments, one or more alternative and / or additional interfacial and / or bulk properties may be used to estimate demulsifier efficiency.

[0080] As mentioned above, there are various types of tensiometers. For example, consider oscillatory spinning drop tensiometers (see, e.g., Marquez et al., "The Oscillatory Spinning Drop Technique. An Innovative Method to Measure Dilational Interfacial Rheological Properties of Crude Oil-Brine Systems in the Presence of Asphaltenes", Colloids Interfaces 5, 42, 2021, 1-22) and drop volume tensiometers (see, e.g., Gong et al., "Study of Dynamic Interfacial Tension for Demulsification of Crude Oil Emulsions", SPE International Symposium on Oilfield Chemistry, Houston, Texas, February 2001). In various examples, data from a hanging drop tensiometer is utilized, which has droplet oscillation capabilities, allowing optical drop shape analysis (see, e.g., Chen et al., "Pilot Performance of Chemical Demulsifier on the Demulsification of Produced Water from Polymer / Surfactant Flooding in the Xinjiang Oilfield", Water 10, 1874, 2018, 1-9). As an example, a system can include a hanging drop tensiometer. In addition to rheological properties of the oil / water interface, a hanging drop tensiometer can also provide accurate and sensitive interfacial tension measurements.

[0081] Figure 3 An example of a system 300 that can generate tensiometer data is shown. Figure 3 In the example of , the system 300 includes a light source 310, an optical camera 320, a droplet of liquid 330 rising from a U-shaped needle 350 in a cuvette 340, the U-shaped needle 350 is fluidically connected to a syringe 360, wherein the cuvette 340 can be supported on an adjustable workbench 370, and an oscillator 380 can control the oscillation.

[0082] System 300 can generate interface characteristic data from the droplet shape image, which can be in digital form or digitized. For example, cuvette 340 can be transparent so that light from light source 310 can pass through the liquid in cuvette 340, so that optical camera 320 can capture one or more shapes of droplet 330. Figure 3 As shown in the example of , the droplet 330 can be a droplet that rises due to buoyancy, wherein the droplet 330 includes a lighter phase of liquid (e.g., an oil phase composed of crude oil) surrounded by a larger, denser phase (e.g., an aqueous phase composed of salt water). As an example, to help mitigate the needle peristalsis effect, the U-shaped needle 350 can be a polymer coated needle (e.g., consider polytetrafluoroethylene (PTFE) and the like). The U-shaped needle 350 can be used to inject the lighter phase from the syringe 360. As an example, the height of the injection assembly can be adjusted relative to the optical path of the light source 310 and the optical camera 320 via the adjustable workbench 370, which helps ensure that the droplet shape is properly captured. As for oscillating the droplet 330 for measuring rheology, this can be achieved by controlling the syringe 360 ​​to increase and decrease the droplet volume. For example, the oscillator 380 can include one or more piezoelectric elements to transmit vibrations at one or more frequencies.

[0083] As an example, the system 300 may include one or more interfaces for control commands and / or data transmission. For example, consider an interface for control and / or data transmission of one or more of the oscillator 380, the adjustable workbench 370, the injector 360, the light source 310, and the optical camera 320. The system 300 may include one or more processors and at least one accessible memory of the one or more processors, wherein the memory may store processor executable instructions. As an example, the system 300 may include one or more databases and / or be operably connected to one or more databases and / or include one or more ML models and / or be operably connected to one or more ML models. As an example, the system 300 may be a demulsifier analysis framework system, which may output demulsifier performance data (e.g., demulsifier efficiency) in response to the generation of interface characteristic data, and the interface characteristic data may include rheological characteristic data (e.g., viscoelastic data, etc.).

[0084] As an example, a method may include using a crude oil phase to generate droplets, and obtaining measurements in the presence and absence of a demulsifier to infer differences in measured interfacial properties. This method may include comparing fluid systems (e.g., phases) with and without a demulsifier for one or more specific oils from a target oil field. As an example, one or more transparent model oils may be used as a bulk phase, and a hanging droplet of a heavier aqueous phase may be injected, for example, from a top position through a needle. As explained, the tensiometer may be operated in a top droplet mode, wherein a dense phase is introduced into a lighter phase from a top position, or in an opposite mode, wherein a lighter phase is introduced into a denser phase (e.g., a heavier phase) from a bottom position. In the top droplet mode, the droplets may extend downward under the influence of gravity, and in the bottom droplet mode, the droplets may extend upward under the influence of gravity due to buoyancy.

[0085] As an example, the tensiometer may include one or more features of a Krüss Scientific tensiometer, for example, consider a Drop Shape Analyzer tensiometer model DSA30R, which may produce IFT and rheological data in the range of 0.01 to 2000 mN / m, with a resolution of 0.01 mN / m, an accuracy of 0.2 mN / m, and utilize one or more models (e.g., Young-Laplace) for static and / or dynamic measurements. As for oscillating droplet measurements, consider data for one or more of E, E', E", and phase shift, where the oscillation may be performed according to one or more waveforms (e.g., a sine wave), and where one or more models (e.g., the Lucassen-van den Tempel diffusion-adsorption model) may be utilized.

[0086] For the Young-Laplace model, it can be used to determine the contact angle in drop shape analysis. For example, the drop shape can be analyzed based on the shape of an ideal stationary drop, whose surface curvature is generated only by the force balance between surface tension and weight. According to the Young-Laplace model, for a curved liquid surface, there is a relationship between the curvature radii r1 and r2, the surface tension σ, and the Laplace pressure p:

[0087] Δp=σ(r1 -1 +r2 -1 )

[0088] The Young-Laplace model can be used to determine the surface tension of a liquid or the interfacial tension of a two-phase system using a pendant drop, and to measure interfacial tension using a spinning drop tensiometer.

[0089] As for the Lucassen-van den Tempel diffusion-adsorption model, it can be used for frequency-dependent data because the real and imaginary parts of the dilatational elastic modulus have frequency dependence, which can be determined for various concentrations of demulsifier (e.g., 0 and greater). The real part of the modulus tends to increase with the droplet area oscillation frequency (ν) and can be described by fitting the Lucassen-van den Tempel (LvdT) diffusion-adsorption model:

[0090] E=E0(1+ξ+iξ) / (1+2ξ+2ξ 2 ),ξ=(ν D / 2ν) 0.5

[0091] where E0 is the Gibbs elasticity, ν D is the characteristic frequency of the diffusion transport mechanism, which can be determined by using the acquired data and appropriate fitting techniques; noting that E = E' + iE".

[0092] Figure 4 Example plots 410, 420, 430, and 440 of data from a tensiometer are shown, which include rheological data. As shown, plot 410 shows data for E, viscoelastic modulus (mN / m) versus frequency (Hz) for an oil and water system without a demulsifier, an oil and water system with 10 ppm of a demulsifier labeled Demulsifier A, and an oil and water system with 30 ppm of a demulsifier labeled Demulsifier A, while plot 420 shows data for E', viscous modulus (mN / m) versus frequency (Hz), and plot 430 shows data for E", elastic modulus (mN / m) versus frequency (Hz). Figure 4 , plot 440 shows data of IFT (mN / m) versus frequency (Hz).

[0093] As described above, the results of the interfacial tension and interfacial rheology measurements can be correlated with the demulsifier efficiency inferred from the bottle test. For example, it is contemplated that one or more correlations may be performed by one or more linear regression models, nonlinear regression models, ML models, and one or more other types of artificial intelligence algorithms.

[0094] Once correlations are revealed through statistical and / or probabilistic analysis of a database populated by measurements, new interfacial measurements for one or more demulsifier intermediates (e.g., new demulsifier intermediates) can be generated, allowing rapid screening and evaluation of these chemicals for effective demulsification potential.

[0095] Figure 5An example of a workflow 500 is shown that can generate a database 550 for a framework 560 that can be used to output demulsifier performance indicators 590 in response to receiving interfacial property measurements 570 of one or more potential demulsifiers 582 (e.g., candidate demulsifiers), crude oils 584 from various oil fields, and brine 586 from various oil fields. As shown, the database 550 can be constructed from interfacial property measurements 530 and bottle tests 540 of existing demulsifiers 512, crude oils 514 from various oil fields, and brine 516 from various oil fields. As explained, the interfacial property measurements can be generated using one or more tensiometers, where such measurements can include rheological property measurements (e.g., E, E', E"), etc.).

[0096] As an example, framework 560 can be a dynamic framework that can respond to receiving data such as interface property measurements 570 to generate outputs such as output 590. Such a framework can reduce the need for bottle testing. For example, output 590 can be generated for one or more potential demulsifiers 582 without having to perform bottle testing on the one or more potential demulsifiers 582. Since tensiometer measurements can be obtained faster than bottle test results, such a framework can be robust and fast.

[0097] As an example, a laboratory or other facility may include one or more tensiometers that are operably coupled locally and / or remotely to the framework. As an example, the framework may be dynamically updated in response to one or more database updates. For example, as additional bottle tests are performed, a database and one or more models, etc., may also be updated.

[0098] As an example, an oil field may be supplied with a reserve of one or more candidate demulsifiers, wherein one or more samples from the oil field may be transported to a facility for interface property measurements. In such an example, the framework may analyze the interface property measurements based on data included in a database to generate an output about the performance of one or more candidate demulsifiers. In such an example, one or more commands may be transmitted to the oil field to control the selection of one of the one or more candidate demulsifiers, and, for example, to inject it into a process stream for demulsification. This method may include maintaining sensor data about candidate demulsifier levels, separator performance, etc. For example, if the first-ranked candidate demulsifier is at a low level in the oil field, a second-ranked candidate demulsifier with a higher level in the oil field may be selected for injection, optionally with a request to obtain an additional supply of the top-ranked candidate demulsifier.

[0099] As explained, the rate determination step may be transporting the oilfield sample to a facility that includes a tensiometer. As explained, the analysis of one or more candidate demulsifiers may be accelerated by reducing the need to perform one or more bottle tests. By using a demulsifier analysis framework, this approach may improve field separations because performance data for candidate demulsifiers may be generated more quickly.

[0100] As explained, the framework can allow for rapid analysis of changes in one or more properties, such as by scanning oils from different reservoirs to analyze the suitability of a particular intermediate in different regions, and by faster concentration screening to optimize the final mixture.

[0101] Figure 6 An example workflow 600 is shown in which an emulsion breaker 604 is subjected to a bottle test to generate bottle test data 610, which may include time series data (e.g., water drop out (WDO) versus time), subjected to a tensiometer measurement to generate related tensiometer data 620, which may include time series and / or frequency series data (e.g., E, E', E", and / or IFT with respect to frequency), and subjected to a field implementation to generate field data 630, which may include time series data (e.g., from one or more field transients), wherein the bottle test data 610, the tensiometer data 620, and the field data 630 may be stored in one or more databases 640. Such a workflow may be a database generation workflow.

[0102] Figure 6 Also shown is model generation of one or more models 650, which may utilize data in one or more databases 640, optionally utilizing input from one or more experts 655, which may provide supervised learning and / or semi-supervised learning and / or one or more functions, which may involve unsupervised learning (e.g., considering hyperparameter tuning, etc.). As shown, a tensiometer measurement technique may be performed on a candidate demulsifier 608 using one or more tensiometers to generate tensiometer data 660 for the candidate demulsifier 608. For example, consider including, for example, Figure 4 As explained, the tensiometer data 660 may include data with and without demulsifiers, data on frequency, data on time, etc. The tensiometer data 660 may be used directly and / or indirectly as an input to at least one of the one or more models 650 to produce an output 670, which may then be used for field implementation 680 of one or more candidate demulsifiers 608, such as selected, ranked, etc., via at least one of the one or more models 650. Figure 6As shown, a workflow can be designed to receive one type of data (e.g., as a digitized image, raw data, etc.) for a candidate demulsifier and output another type of data (e.g., as a digitized image, raw data, etc.), where the output characterizes the performance of the candidate demulsifier (e.g., considering demulsifier efficiency).

[0103] As an example, candidate demulsifiers 608 may include associated data, such as metadata. For example, consider metadata that may be associated with field data 630, which may include specifications regarding one or more field parameters (e.g., reservoir type, fluid type, equipment type, residence time, etc.).

[0104] As an example, the one or more models 650 can be trained (e.g., machine learning) using data from the one or more databases 640 so that various inputs about the candidate demulsifiers 608 can be received to produce outputs about the predicted effectiveness of the candidate demulsifiers 608 without having to perform bottle tests on multiple top-ranked candidate demulsifiers 608. Alternatively, the need for bottle testing can be greatly reduced, for example, by narrowing the list of candidate demulsifiers to a few top-ranked candidate demulsifiers.

[0105] As described above, the method may include simulation. For example, consider an output 670 including predicted time performance of a demulsifier, which may be used to construct a water separation versus time curve. In such an example, a separation process, for example, may be simulated, wherein a certain amount of a selected demulsifier is injected, which is performed according to a water separation versus time curve. The results from such a simulation may help customize an injection plan, injection rate, etc., regarding one or more field parameters (e.g., flow rate, residence time, desired separation performance, etc.). Thus, the output 670 may include information for outputting one or more performance curves of one or more candidate demulsifiers 608. As an example, the output 670 may present information in a format similar to bottle test data and / or in a format similar to field data, which may include performance versus time, concentration versus performance, performance versus time based on concentration, etc.

[0106] As an example, data in one or more databases may include relationships that can be understood, revealed, etc. using one or more techniques. For example, consider regression, correlation, etc. As an example, data may be organized by tensiometer measurement parameters. For example, consider the generation of a two-dimensional plot of bottle test dynamics versus viscoelasticity and / or IFT. While such a two-dimensional plot may be viewed visually to understand some relationships, such a plot or combination of dimensions may involve more complex relationships.

[0107] As an example, the relationship can be linear and / or non-linear. As an example, a neural network model and a linear regression model can be implemented for a linear relationship, noting that a linear regression can be computationally less demanding (e.g., with respect to training) than a neural network model. However, for a non-linear relationship, the model can include, for example, a polynomial regression model, a support vector machine (SVM), a random forest, etc., or, for example, a neural network model. Depending on the amount of data, a relatively modest neural network model may be computationally more efficient.

[0108] As an example, data can be represented visually so that the problem can be viewed as an image-based problem. For example, consider an image based on a combination of bottle test data and tensiometer data, where a trained neural network can receive the tensiometer data and predict what its bottle test data looks like in the image. This problem is similar to filling in the missing part of the image (e.g., bottle test data) when a portion of the image (e.g., tensiometer data) is given. In this approach, the missing part can be synthetic, which can be performed through a generative process, and / or actual, which can be performed through a matching process of information from a database. As for image-based methods, the model can include a convolutional neural network (CNN). As an example, if a problem is viewed as a time series problem, it can be processed using one or more recurrent neural networks and LSTM (long short-term memory). For example, as described above, the performance of a demulsifier can be characterized using time series data, such as separated water versus time, which can be obtained from bottle tests, field trials, etc. As an example, the framework can receive tensiometer data and associate it with separated water versus time data, which can be output in a manner that provides a ranking of candidate demulsifiers.

[0109] As for regression and classification, these can be handled as separate tasks or combined tasks. Various techniques can be implemented to convert regression-type problems into classification-type problems and vice versa. As an example, the efficiency of a candidate demulsifier can be predicted, or the efficiency of a candidate demulsifier can belong to a category that can be determined by classification.

[0110] As mentioned earlier, the framework can implement clustering or grouping, which may be a problem of identifying similarities. As an example, a combined regression (prediction) and classification ML model can be constructed. For example, consider an architecture with an input layer, a hidden layer, and multiple output layers. In such an example, the regression and classification output layers can be connected to the common last hidden layer of the model. Given two output layers, two loss functions can be used to train the model, for example, considering the mean squared error (MSE) loss of the regression output layer and the sparse classification cross entropy of the classification output layer. An example of a combined ML model for regression (prediction) and classification can be used to determine the age of abalone (e.g., a fish) from physical details, where the number of annual rings predicted for abalone is a proxy for the age of the abalone (e.g., age can be predicted as a numerical value (in years) or a class label (as an ordinal year of the class)). In various examples, the trained ML model can output probability information. For example, consider the probability that an input belongs to a specific class.

[0111] As for the types of machine learning models that the system can utilize, consider one or more of a support vector machine (SVM) model, a k-nearest neighbor (KNN) model, an ensemble classifier model, a neural network (NN) model, etc. As examples, the machine learning model can be a deep learning model (e.g., a deep Boltzmann machine, a deep belief network, a convolutional neural network, a stacked autoencoder, etc.), an ensemble model (e.g., a random forest, a gradient boosting machine, bootstrap aggregation, adaptive boosting (AdaBoost), stacked summarization, gradient boosting regression trees, etc.), a neural network model (e.g., a radial basis function network, a perceptron, back propagation, a Hopfield network, etc.), a regularization model (e.g., a ridge regression, a least absolute shrinkage and selection operator, an elastic net, a least angle regression), a rule system model (e.g., a cubist, a one rule rule), zero rule, repeated incremental pruning to produce error reduction), regression models (e.g., linear regression, ordinary least squares regression, stepwise regression, multivariate adaptive regression splines, local estimation scatter plot smoothing, logistic regression, etc.), Bayesian models (e.g., naive Bayes, average dependency estimator, Bayesian belief network, Gaussian naive Bayes, multinomial naive Bayes, Bayesian network), decision tree models (e.g., classification and regression trees, iterative bisection3, C4.5, C5.0, chi-square automatic interaction detection, decision tree Stakes, conditional decision trees, M5), dimensionality reduction models (e.g., principal component analysis, partial least squares regression, Sammon map, multidimensional scaling, projection pursuit, principal component regression, partial least squares discriminant analysis, mixed discriminant analysis, quadratic discriminant analysis, regularized discriminant analysis, flexible discriminant analysis, linear discriminant analysis, etc.), instance models (e.g., k-nearest neighbors, learning vector quantization, self-organizing maps, local weighted learning, etc.), clustering models (e.g., k-means, k-median, expectation-maximization, hierarchical clustering, etc.), and so on.

[0112] As an example, the TENSORFLOW framework (Google LLC, Mountain View, California) can be implemented, which is an open source software library for data flow programming, including a symbolic math library, which can be implemented for machine learning applications that can include neural networks. As an example, the CAFFE framework can be implemented, which is a DL framework developed by Berkeley AI Research (BAIR) (University of California, Berkeley, California). As another example, consider the SCIKIT platform (e.g., scikit-learn) using the PYTHON programming language. As an example, a framework such as the APOLLO AI framework can be utilized (APOLLO.AI GmbH, Germany). As described above, a framework such as the PYTORCH framework can be utilized.

[0113] As an example, a training method may include various actions that may be operated on a dataset to train an ML model. As an example, a dataset may be divided into training data and test data, where the test data may provide an evaluation. A method may include cross-validation of parameters and optimal parameters, which may be provided for model training.

[0114] The TENSORFLOW framework can run on multiple CPUs and GPUs (with optional CUDA (NVIDIA Corp., Santa Clara, California) and SYCL (The Khronos Group Inc., Beaverton, Oregon) extensions for general-purpose computing on graphics processing units (GPUs). TENSORFLOW is available on 64-bit LINUX, MACOS (Apple Inc., Cupertino, California), WINDOWS (Microsoft Corp., Redmond, Washington), and mobile computing platforms, including those based on the ANDROID (Google LLC, Mountain View, California) and IOS (Apple Inc.) operating systems.

[0115] TENSORFLOW computations can be represented as stateful data flow graphs; note that the name TENSORFLOW comes from the operations that this neural network performs on multidimensional arrays of data, which can be called "tensors".

[0116] As an example, the device can utilize TENSORFLOW LITE (TFL) or another type of lightweight framework. TFL is a set of tools that support machine learning on devices, where models can run on mobile, embedded, and IoT devices. TFL is optimized for machine learning on devices, addressing latency (no round trip to the server), privacy (no personal data leaves the device), connectivity (requires an Internet connection), size (reduced model and binary size), and power consumption (e.g., effective reasoning and lack of network connection). TFL includes multi-platform support, covering ANDROID and iOS devices, embedded LINUX and microcontrollers, and multiple language support, including JAVA, SWIFT, Objective-C, C++, and PYTHON. TFL can provide high performance through hardware acceleration and model optimization. Machine learning tasks that can be performed using TFL can include, for example, simulation, regression, classification, object detection, posture estimation, question answering, text classification, etc. As an example, a gateway can be installed at a site, and provide execution of a framework such as a TFL framework. In such an example, the gateway can provide determination of one or more control actions and release of one or more control actions at the site.

[0117] Figure 7 An architecture 700 of a framework such as the TENSORFLOW framework is shown. As shown, the architecture 700 includes various features. As an example, in the terminology of the architecture 700, a client may define a computation as a dataflow graph, and, for example, a session may be used to initiate graph execution. As an example, a distributed host may remove a particular subgraph from a graph, as defined by the parameters of "Session.run()"; split a subgraph into multiple parts running in different processes and devices; distribute graph blocks to worker services; and initiate graph block execution via a worker service. As for the worker services (e.g., one per task), as an example, they may schedule execution of graph operations using kernel implementations appropriate for the available hardware (CPU, GPU, etc.) and, for example, send and receive operation results to and from other worker services. As for the kernel implementations, for example, they may perform computations for separate graph operations.

[0118] Figure 7 Also shown are some examples of types of machine learning models 710, 720, 730, and 740, one or more of which may be utilized. As explained, the ML model-based method may include receiving data that may be multi-dimensional spatial and / or temporal data. As an example, time may be a dimension such that the data may be spatial and temporal. As previously explained, frequency may be a dimension. As an example, a convolutional neural network and / or one or more other types of neural networks may be used for spatial and / or spatiotemporal analysis.

[0119] Figure 8 An example of a method 800 and an example of a system 890 are shown. As shown, the method 800 includes a receiving block 810 for receiving, by a demulsifier analysis framework, tensiometer data of an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency, and wherein the demulsifier analysis framework includes a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions; a determining block 820 for determining performance characteristics of the candidate demulsifier relative to the oil and water field sample using the tensiometer data as input to a model of the demulsifier analysis framework without performing a bottle test on the oil and water field sample; and an output block 830 for outputting, by the demulsifier analysis framework, performance characteristics of the candidate demulsifier relative to breaking an emulsion of the oil and water field sample.

[0120] Method 800 is shown as including various computer readable storage media (CRM) blocks 811 , 821 , and 831 , which may include processor-executable instructions that may instruct a computing system (which may be a control system) to perform one or more actions described with respect to method 800 .

[0121] exist Figure 8 In the example of , system 890 includes one or more information storage devices 891, one or more computers 892, one or more networks 895, and instructions 896. With respect to one or more computers 892, each computer may include one or more processors (e.g., or processing cores) 893 and a memory 894 for storing, for example, instructions 896 executable by at least one of the one or more processors 893 (e.g., see blocks 811, 821, and 831). As an example, a computer may include one or more network interfaces (e.g., wired or wireless), one or more graphics cards, a display interface (e.g., wired or wireless), and the like.

[0122] As an example, method 800 can be a workflow that can be implemented using one or more frameworks in a framework environment. As an example, system 890 can include local and / or remote resources. For example, consider that a browser application executed on a client device is a local resource relative to a user of the browser application, and a cloud-based computing device is a remote resource relative to the user. In such an example, a user can interact with a client device via a browser application, wherein information is transmitted to a cloud-based computing device (or multiple devices), and wherein the information can be received as a response and presented to a display operably connected to the client device (e.g., via a service, API, etc.).

[0123] As an example, a method may include receiving, by a demulsifier analysis framework, tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency, and wherein the demulsifier analysis framework includes a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions; using the tensiometer data as an input to the model of the demulsifier analysis framework, determining a performance characteristic of the candidate demulsifier relative to the oil and water field sample without performing a bottle test on the oil and water field sample; and outputting, by the demulsifier analysis framework, a performance characteristic of the candidate demulsifier relative to breaking an emulsion of the oil and water field sample. In such an example, the oil and water field sample may include crude oil and brine.

[0124] As an example, a method may include determining an amount of a candidate demulsifier to inject into a well fluid of an oilfield process, such as an oilfield process including an oil and water separation process.

[0125] As an example, an oil and water field sample may include an emulsion, where the performance characteristic indicates, for example, the efficiency of the candidate demulsifier in breaking the emulsion.

[0126] As an example, the performance characteristic may be demulsifier efficiency, where, for example, demulsifier efficiency is defined as the volume of separated water divided by the initial volume of water in the emulsion of oil and water oilfield sample.

[0127] As an example, the performance characteristic of a candidate demulsifier can be a rate, where, for example, the rate includes a concentration-dependent rate at which the candidate demulsifier separates water from an emulsion of oil and water in an oilfield sample with respect to time.

[0128] As an example, a method may include performing a process separation simulation using performance characteristics of a candidate demulsifier, wherein, for example, the process separation simulation simulates fluid behavior in the presence of the candidate demulsifier for a separator in an oil field from which oil and water field samples are collected.

[0129] As an example, the rheological data may include one or more moduli, wherein the one or more moduli include one or more of a viscoelastic modulus, an elastic modulus, and a viscous modulus.

[0130] As an example, the model may include a pixel image based model.

[0131] As an example, the model may include one or more of a regression model and a classification model.

[0132] As an example, the model may correlate received tensiometer data with at least a portion of previously acquired bottle test data (eg, possibly present in a database).

[0133] As an example, the demulsifier analysis framework may include a tensiometer that generates tensiometer data.

[0134] As an example, a method may include determining performance characteristics of a candidate demulsifier relative to an oil and water field sample using tensiometer data as an input to a model without performing bottle testing of the oil and water field sample, the model occurring in response to receiving the tensiometer data based at least in part on bottle test data of demulsifiers in oil and water emulsions. For example, the laboratory apparatus may include a tensiometer, wherein the framework is operably coupled to the tensiometer to receive the tensiometer data therefrom, and automatically generating an output based on at least a portion of the tensiometer data using the model.

[0135] As an example, a system may include a processor; a memory operably coupled to the processor; and processor-executable instructions stored in the memory to direct the system to: receive tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data relative to frequency; use the tensiometer data as input to a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions to determine performance characteristics of the candidate demulsifier relative to the oil and water field sample without performing bottle testing of the oil and water field sample; and output performance characteristics of the candidate demulsifier relative to the oil and water field sample.

[0136] As an example, one or more computer-readable storage media may include processor-executable instructions to instruct a computing system to: receive tensiometer data for an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency; determine performance characteristics of the candidate demulsifier relative to the oil and water field sample using the tensiometer data as input to a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions without performing bottle testing of the oil and water field sample; and output performance characteristics of the candidate demulsifier relative to the oil and water field sample.

[0137] A computer-readable storage medium (or computer-readable storage medium) is non-transitory, neither a signal nor a carrier wave. Instead, a computer-readable storage medium is a physical device that can be considered as a circuit or hardware.

[0138] Fig. 9Components of an example of a computing system 900 and an example of a networked system 910 with a network 920 are shown. As an example, the demulsifier analysis framework can be implemented within the computing system 900, within the networked system 910, and the like. The system 900 includes one or more processors 902, a memory and / or storage component 904, one or more input and / or output devices 906, and a bus 908. In an example embodiment, instructions can be stored in one or more computer-readable media (e.g., memory / storage component 904). Such instructions can be read by one or more processors (e.g., processor 902) via a communication bus (e.g., bus 908), which can be wired or wireless. One or more processors can execute such instructions to implement (in whole or in part) one or more properties (e.g., as part of a method). A user can view and interact with the output from the process through an I / O device (e.g., device 906). In an example embodiment, the computer-readable medium can be a storage component, such as a physical memory storage device, such as a chip, a chip on a package, a memory card, etc. (e.g., a computer-readable storage medium).

[0139] In an example embodiment, the components may be distributed, such as in a network system 910. The network system 910 includes components 922-1, 922-2, 922-3, ... 922-N. For example, component 922-1 may include processor 902, and component 922-3 may include a memory accessible by processor 902. In addition, component 922-2 may include an I / O device for display and optionally interacting with the method. The network may be or include the Internet, an intranet, a cellular network, a satellite network, etc.

[0140] As an example, the device can be a mobile device that includes one or more network interfaces for information communication. For example, the mobile device can include a wireless network interface (e.g., operable via IEEE 802.11, ETSI GSM, Bluetooth, satellite, etc.). As an example, the mobile device can include components such as a main processor, a memory, a display, a display graphics circuit (e.g., optionally including touch and gesture circuits), a SIM slot, an audio / video circuit, a motion processing circuit (e.g., an accelerometer, a gyroscope), a wireless LAN circuit, a smart card circuit, a transmitter circuit, a GPS circuit, and a battery. As an example, the mobile device can be configured as a mobile phone, a tablet computer, etc. As an example, a method can be implemented (e.g., in whole or in part) using a mobile device. As an example, a system can include one or more mobile devices.

[0141] As an example, the system can be a distributed environment, such as a so-called "cloud" environment, in which various devices, components, etc., interact for the purpose of data storage, communication, computing, etc. As an example, a device or system can include one or more components for communicating information via one or more of the Internet (e.g., in the case of communicating via one or more Internet protocols), a cellular network, a satellite network, etc. As an example, the method can be implemented in a distributed environment (e.g., in whole or in part as a cloud-based service).

[0142] As an example, information can be input from a display (e.g., consider a touch screen), output to a display, or both. As an example, information can be output to a projector, a laser device, a printer, etc. so that the information can be viewed. As an example, information can be output stereoscopically or holographically. As for printers, consider 2D or 3D printers. As an example, a 3D printer can include one or more substances that can be output to construct a 3D object. As an example, data can be provided to a 3D printer to construct a 3D representation of an underground stratum. As an example, layers can be constructed in 3D (e.g., horizons, etc.), three-dimensionally constructed geological bodies, etc. As an example, holes, cracks, etc. can be constructed in 3D (e.g., as positive structures, as negative structures, etc.).

[0143] Although only a few example embodiments have been described in detail above, those skilled in the art will readily appreciate that many modifications are possible in the example embodiments. Therefore, all such modifications are intended to be included within the scope of the present disclosure as defined by the following claims. In the claims, means-plus-function clauses are intended to cover structures described herein that perform the described function, not only structural equivalents, but also equivalent structures. Thus, although a nail and a screw may not be structurally equivalent in that a nail employs a cylindrical surface to fasten wooden parts together while a screw employs a helical surface, in the context of fastening wooden parts, a nail and a screw may be equivalent structures.

Claims

1. A method comprising: receiving, by a demulsifier analysis framework, tensiometer data of an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency, and wherein the demulsifier analysis framework includes a model derived at least in part based on bottle test data of demulsifiers in oil and water emulsions; using the tensiometer data as input to a model of a demulsifier analysis framework to determine performance characteristics of candidate demulsifiers relative to oil and water field samples without bottle testing the oil and water field samples; and Through the demulsifier analysis framework, performance characteristics of candidate demulsifiers relative to breaking emulsions of oil and water oilfield samples are output.

2. The method of claim 1, wherein the oil and water field sample comprises crude oil and brine.

3. The method of claim 1, comprising determining an amount of a candidate demulsifier to be injected into a well fluid of an oilfield process.

4. The method of claim 3, wherein the oilfield process comprises an oil-water separation process.

5. The method of claim 1, wherein the oil and water field sample comprises an emulsion.

6. The method of claim 5, wherein the performance characteristic is indicative of the efficiency of the candidate demulsifier in breaking an emulsion.

7. The method of claim 1, wherein the performance characteristic comprises demulsifier efficiency.

8. The method of claim 7, wherein the demulsifier efficiency is defined as the volume of separated water divided by the initial water volume in the emulsion of oil and water oilfield sample.

9. The method of claim 1, wherein the performance characteristic of the candidate demulsifier comprises rate.

10. The method of claim 9, wherein the rate comprises a concentration-dependent rate of a candidate demulsifier for separating water from an emulsion of the oil and water field sample with respect to time.

11. The method of claim 1 , comprising performing a process separation simulation using the performance characteristics of the candidate demulsifier.

12. The method of claim 11, wherein the process separation simulation simulates fluid behavior in the presence of a candidate demulsifier for use in a separator in an oil field from which oil and water oil field samples are collected.

13. The method of claim 1, wherein the rheological data comprises one or more moduli, wherein the one or more moduli comprise one or more of a viscoelastic modulus, an elastic modulus, and a viscous modulus.

14. The method of claim 1, wherein the model comprises a pixel image based model.

15. The method of claim 1, wherein the model comprises one or more of a regression model and a classification model.

16. The method of claim 1, wherein the model relates the received tensiometer data to at least a portion of the bottle test data.

17. The method of claim 1, wherein the demulsifier analysis framework includes a tensiometer that generates tensiometer data.

18. The method of claim 1, wherein the determining occurs in response to receipt of tensiometer data.

19. A system comprising: processor; a memory operably coupled to the processor; as well as Processor-executable instructions stored in the memory for instructing the system to: receiving tensiometer data of an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency; using the tensiometer data as input to a model derived at least in part based on bottle testing data of demulsifiers in oil and water emulsions to determine performance characteristics of candidate demulsifiers relative to oil and water field samples without performing bottle testing of the oil and water field samples; and Output the performance characteristics of candidate demulsifiers relative to oil and water field samples.

20. One or more computer-readable storage media comprising processor-executable instructions to instruct a computing system to: receiving tensiometer data of an oil and water field sample including an added candidate demulsifier, wherein the tensiometer data includes rheological data with respect to frequency; using the tensiometer data as input to a model derived at least in part based on bottle testing data of demulsifiers in oil and water emulsions to determine performance characteristics of candidate demulsifiers relative to oil and water field samples without performing bottle testing of the oil and water field samples; and Output the performance characteristics of candidate demulsifiers relative to oil and water field samples.

Images