System and method for separating water-in-oil emulsions by electro-energization in petroleum extraction, method for training a neural network, and corresponding computer-readable memory

Abstract

The present invention relates to a system and a method for separating water-in-oil emulsions by electro-energization of the water-in-oil emulsion extracted from an oil well (210), wherein the fluid petroleum (PF) is electro-energized by applying high voltage to an electron trap due to the electrical capacitive property of an assembly comprising an extraction pipe (220), electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6,..., 130-n) and spark gaps and / or diodes (110-1, 110-2, 110-3,..., 110-n), and wherein electro-energization of the fluid petroleum (PF) promotes partial and / or total separation of the water-in-oil emulsion upstream of the primary extraction separator (230), wherein the fluid petroleum (PF) is a water-in-oil emulsion. The present invention also relates to a method for training a neural network to evaluate physicochemical properties and take decisions for controlling a processing unit in order to promote the electro-acidification and / or electro-alkalization of the process fluid. Finally, the present invention also relates to a computer-readable memory comprising information and instructions which, when executed, carry out an electro-energization method according to the invention.

Description

SYSTEM AND METHOD FOR SEPARATING WATER-IN-OIL EMULSIONS BY ELECTROENERGIZATION IN PETROLEUM EXTRACTION, METHOD FOR TRAINING A NEURAL NETWORK AND CORRESPONDING MEMORY READ BY COMPUTER Field of application

[0001] The present invention belongs to the field of methods for obtaining petroleum, oil and gas from wells or mines, notably the electrocoalescence of water-in-oil emulsions by the use of electron traps. Introduction

[0002] The present invention relates to a system and a method for separating water-in-oil emulsions by means of electro-energizing the water-in-oil emulsion extracted from an oil well, wherein the electro-energizing of the fluid oil occurs by applying high voltage, in which the fluid oil is a water-in-oil emulsion.

[0003] The present invention also relates to a method for training a neural network to evaluate the physicochemical properties and make decisions for controlling a processing unit to promote the electroacidification and / or electroalkalization of the fluid in process.

[0004] Finally, the present invention also relates to a computer-readable memory comprising information and instructions which, when executed, perform an electro-energization method in accordance with the invention. Fundamentals of the invention

[0005] Oil exploration is basically done by pumping oil from a natural underground deposit using... one or more extraction wells, with one or more separators for oil, water, gas and solids.

[0006] The most commonly used petroleum treatments vary between the use of physical and chemical means for refining, purification, and removal of waste, water, and / or hydrocarbons.

[0007] Physical filters are known and used, such as classic physical barriers, sieves, gravel, activated carbon, polymers, nanoparticles, cavitation, nanocavitation, centrifuges, electrolysis systems, settling tanks, ozonizers, electromagnetic waves, among others, as well as chemical means, such as various chemicals, surfactants and polymers used to flocculate, agglomerate, polarize and purify the extracted oil.

[0008] It should be noted, however, that the state of the art does not provide solutions that utilize the properties and advantages of electron traps as electrocoalescence agents for emulsions containing oil and water in oil exploration. Therefore, in oil extraction from an oil well, the state of the art does not reveal solutions that utilize the electrical capacitive property established inside an extraction pipe and a high-voltage source arranged coaxially to the extraction pipe, which functions as an electron trap, capable of totally or partially separating the water-in-oil emulsion that forms the extracted oil. It is worth remembering that the water-in-oil emulsion can also be an emulsion of water, salts, sand, gases, ores, solid or fluid residues, and ions in oil.

[0009] It is also important to mention that the environment and conditions of oil wells are highly varied, with high temperatures and high salinity levels always prevailing, and often the presence of ions. divalent compounds, including calcium and magnesium, represent hard water with high surface tension. State of the art

[0010] Approximate solutions known in the state of the art for the separation of water-in-oil emulsions in petroleum extraction can be found in state-of-the-art documents such as US patent document US 2017 / 0021287, entitled "Systems and methods for unipolar separation of emulsions and other mixtures", which relates to systems and methods for separating two or more phases of an emulsion or other mixture, wherein said methods include providing the mixture with a unipolar and liquid charge (for example, so that adjacent droplets in the same acquire unipolar and liquid charges), thereby enhancing the coalescence of droplets of similar phase in the same and producing or enhancing the production of two or more consolidated phases, and collecting the two or more consolidated phases, according to its abstract.

[0011] It should be noted, however, that the teachings in document US 2017 / 0021287 do not provide conditions for the creation of an electron trap. On the contrary, it provides grounding (see, for example, Figure 18), which allows for the discharge and neutralization of the emulsion and prevents the trapping of free electrons and products overloaded with them. Furthermore, its teachings regarding its use in the oil field are scarce and superficial, with brief mentions of its applications being suitable for this technical area. Thus, it does not allow someone skilled in the art to find sufficient teachings regarding this specific application.

[0012] Another patent document whose solution can be mentioned is document PI 1002195-7, entitled "Process for increasing efficiency". The abstract describes a process for increasing the efficiency of electrocoalescence of water / oil (W / O) emulsions formed in petroleum treatment plants during the desalting and dehydration stages. This process comprises forming a composition including between 0.001% and 50% by weight of vegetable, animal, or synthetic triglycerides, petroleum, and water (between 2% and 40% water, preferably between 5% and 25% water by weight). The abstract also describes the composition used in the process.

[0013] However, PI 1002195-7 also does not offer conditions for the creation of an electron trap. It is also noted that its figures and descriptive report do not provide those skilled in the art with details about its vessel or voltage application device, revealing the use of devices already known in the prior art and the lack of suggestion or disclosure of a new device or system with advantages over what is known at the time of its publication. Additionally, and as evidenced by the abstract itself, this document uses a composition as a demulsifying agent to obtain the expected results. Besides being more expensive, the inclusion of such a composition makes the electrocoalescence process less optimized.

[0014] Finally, it is also worth mentioning patent document WO2013 / 082681, entitled "Equipment for electrostatic destabilization of fluid emulsions under pressure in a hermetic system and test method," which describes equipment for the electrostatic destabilization of conductive and non-conductive fluid emulsions under pressure in a hermetic system, where the emulsion separation occurs through a highly efficient electrocoalescence process used industrially, according to its summary. The equipment in WO2013 / 082681 comprises a separation vessel, a feed vessel to supply fluids to the vessel and gases for pressurizing said vessel, an agitator to emulsify the fluids, the vessel being kept hermetic by threaded connections adapted to the upper caps of the vessel. Furthermore, this document also describes the test method for evaluating the electrostatic stability of fluid emulsions using the equipment.

[0015] In addition to a sophisticated constructive configuration, the object of WO2013 / 082681 exhibits a very complex and diverse testing method compared to that sought by the present invention. By way of example only, the method claimed in the above WO document requires agitation of the emulsion and, consequently, the inclusion of a stirrer in the equipment. It is also noted that it exhibits a directly applied electrolysis system which, however, differs greatly from the electron trap proposed herein, since a simple electric current through two electrodes does not constitute an electron trap. Those skilled in the art will understand that to generate an electron trap, an electric current blocker in the medium is necessary, along with a system composed of high-voltage electric generators.

[0016] It should also be noted that electroneutralization by simple electrolysis processes is already widely known, with reports of its use easily found throughout the literature of this technical area. However, reports on the use of electron traps for such purposes, as well as for providing physicochemical effects in a practical way, are very scarce and incomplete.

[0017] There are several advantages to using electron traps for a wide variety of applications, including energy savings (compared to simple electrolysis), ease of adaptation to various practical systems, both dynamically and in batches, simple and economical commercial application, high processing speed, and a clean and sustainable process.

[0018] Taking into account the aforementioned teachings of the prior art, there is a clear need for an emulsion separation solution that resolves the problems not overcome by the relevant prior art.

[0019] Thus, the material now disclosed aims to solve such problems through a system and method of electrocoalescence of emulsions, which uses the principle of electron trapping together with spark gaps or blockers of trivial electric current flow, thus providing the necessary potential differentials for the electron traps in question, in which electrical energy is applied to a water / oil emulsion in order to electro-energize the fatty acids and other radicals of the oil, forcing their separation from the water. Objectives of the invention

[0020] One of the objectives of the present invention is, therefore, to provide a system for separating water-in-oil emulsions by electroenergization. in the extraction of petroleum, according to the characteristics of claim 1 of the attached claims.

[0021] Another objective of the present invention is to provide a method for separating water-in-oil emulsions in electro-energization oil extraction, according to the characteristics of claim 11 of the attached claims.

[0022] Yet another objective of the invention is to provide a method for training a neural network, according to the characteristics of claim 12 of the attached claims.

[0023] Another objective of the present invention is to provide a computer-readable memory, according to the characteristics of claim 13 of the attached claims.

[0024] Other features and details of the features are represented by the dependent claims. Brief description of the figures

[0025] For a better understanding and visualization of the object of the present invention, it will now be described with reference to the attached figures, representing the technical effect obtained through an exemplary embodiment that is not limiting the scope of the present invention, in which, schematically: Figure 1: Shows a front view of an extraction platform comprising the system for separating water-in-oil emulsions by electro-energization in oil extraction, according to the invention; and Figure 2: Shows an enlarged view of detail A of Figure 1. 1. Detailed description of the invention

[0026] The following detailed description refers to the accompanying drawings in which embodiments of the present invention are represented, by way of non-limiting illustration. These embodiments are described in such a way as to allow a person skilled in the art to reproduce their results. Other embodiments resulting from structural, mechanical, logical, electrical and electronic changes are possible and can be carried out without departing from the spirit and scope of the present invention. The following detailed description should therefore not be understood in a restrictive or limiting manner.

[0027] The present invention relates to a system (100) and a method for separating water-in-oil emulsions by electro-energization in oil extraction in oil wells (200), to a method for training a neural network and to a corresponding computer-readable memory. System

[0028] The system (100), according to the invention, comprises: i. At least one extraction platform (200), offshore or onshore, for oil exploration from an oil well (210); ii. At least one extraction pipeline (220) for extracting fluid oil (PF) from the oil well (210); iii. At least one positive electrical power source (110) and at least one negative electrical power source (111); iv. At least one grounding means (220-1) electrically connected to the extraction pipeline (220) by means of a plurality of connecting cables (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and a plurality of connectors (113); V. At least one high-voltage cable (120) arranged coaxially to the extraction pipe (220) inside the extraction pipe (220) and electrically connected to the electrical power source (110, 111); vi. A plurality of power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); VU. A plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); VÜi. At least one processing unit (170) for system control (100); wherein the fluid petroleum (FP) extracted from the oil well (210), in the form of a water-in-oil emulsion, is subjected to electromagnetic pulses generated by the electrical power source (110, 111) and controlled by the processing unit (170) comprising a computer-readable memory and a trained convolutional neural network, based on the signals from the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), the electromagnetic pulses being transmitted to the fluid petroleum (FP) through the electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) along the entire path of the fluid petroleum (FP) through the extraction pipeline (220), from the oil well. (210) to a primary extraction separator (230) from the extraction platform (200), where the pulses Electromagnetic currents generated by the electrical power source (110, 111) are alternated with switching of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) controlled by the processing unit (170), wherein the extraction pipe (220), together with the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), forms an electron trap due to the electrical capacitive property of the assembly formed by the extraction pipe (220), the power supply terminals (130-1, 130-2, 130-3 ... 130-4, 130-5, 130-6, ..., 130-n) and by spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), and wherein the electro-energization of the fluid petroleum (PF) promotes the partial and / or total separation of the water-in-oil emulsion before the primary extraction separator (230).

[0029] The term extraction platform (200) refers to any and all extraction platforms (200), whether offshore - in the high seas or far from the coast - or onshore - on land -, for the exploration of oil from an oil well (210).

[0030] The term "petroleum", according to the invention, refers to petroleum in pumpable fluid form extracted from the oil well (210), this fluid petroleum (FP) being essentially a water-in-oil emulsion, this emulsion may also comprise, in addition to water and oil, but without limiting the invention, salts, sand, gases, ores, solid or fluid residues, ions and all kinds of impurities from the oil well (210) and usual in oil extraction processes, the fluid petroleum (FP) thus also being an emulsion of water, salts, sand, gases, ores, solid or fluid residues, ions and impurities in oil.

[0031] It is noteworthy that fluid petroleum (FP) is an emulsion comprising water, salts, sand, gases, ores, solid or fluid residues, ions, and oil impurities, and thus, the components of this emulsion have different electrical conductivities. Saltwater, ores, and ions, for example, have higher electrical conductivity than gases and the oil itself, where gases and oil, depending on their composition, can even be considered electrical insulators. It is precisely this difference in electrical conductivity that aids in the formation of differences in electrical potential between the phases that compose fluid petroleum (FP), which, when electro-energized, according to the invention, begins to separate especially from water and gases.

[0032] An oil well (210), in the context of the invention, refers to any and all oil wells (210) drilled in a permeable underground (onshore) or seabed (offshore) formation containing hydrocarbons, including units and installations known in the prior art and necessary for the extraction and / or production and / or processing of hydrocarbons.

[0033] The extraction pipe (220) is any and all extraction pipe (220) used to aspirate and / or receive the upward flow of oil extracted from the oil well (210) on an extraction platform (200), wherein this extraction pipe (220) preferably comprises, but without limiting the invention, an electrically insulated inner surface. It should be noted that this insulation may be insulation applied to the inner part of the extraction pipe (220) and / or insulation resulting from the formation of an oxidation layer on the inner surface of the extraction pipe (220), usual in the operation on extraction platforms (200).

[0034] Still within the context of the present invention, the term "electrical energy source" refers to an electrical energy source, with adjustable voltage and pulsed current, which may be direct current and / or alternating current, comprising at least one positive electrical energy source (110) for electron sequestration (electroacidulation) and at least one negative electrical energy source (111) for electron accumulation (electroalkalization).

[0035] The power sources (110, 111) are switchable and electrically connected to the high-voltage cables (220) by means of a set of switches or commutators. This connection also includes a set of diodes to ensure the correct direction of current flow according to the switched / selected source (110, 111) to energize the high-voltage cables (220), thus preventing reverse currents during the electro-energization process and enabling complete ionization as per parameterization. It should be noted that one or more of the diodes may eventually be replaced by contactless spark gap devices.

[0036] It should be noted that, regardless of whether a direct current or alternating current power source is used, practical tests complementary to the studies of the present invention make it clear that the higher the applied voltage, the better and more intense the harmonization of the resulting electron flow within the fluid. The choice of current intensity follows the same reasoning, that is, the higher the applied current, the more uniform the electron flow.

[0037] These considerations, however, should not be understood as limiting the applications of the present invention, since the choice of voltage and current levels will depend on the type of fluid, the conditions and characteristics of the fluid, of the extraction pipe (220) through which it is transferred, of any salts and / or gases and / or ions and / or water and / or sand and / or ores and / or impurities and other elements present in the fluid and other conditions that may influence the dielectric characteristics of the assembly.

[0038] That being said, the use of both low and high voltages and currents must be considered, with pulsed direct current being preferable, but without discarding the option of pulsed alternating current. For high-voltage generating sources, we have Van de Graaff sources or trivial sources, capable of generating pulsed or non-pulsed unilateral currents. The electrical voltages can vary within a range of 0.1 V to 1 GV, preferably following the range between 10 and 300 kV, more preferably a range around 15 kV. The frequency of the electrical pulses can be from 60 Hz to 1 x 10 15Hz, preferably between 60 and 1 kHz.

[0039] Power sources (110, 111) are suitable electrical power sources according to the invention, being direct current and / or pulsed alternating current sources that should enable electrical potential differences between 1 kV and 100 GV, preferably, but not limited to, a range between 0.1 V and 10 GV. The choice of voltage will depend essentially on the type of fluid to be energized, the intended energization time, and the presence or absence of any salts and / or gases and / or ions and / or water and / or sand and / or ores and / or impurities and other elements present in the fluids, as well as, of course, the dielectric properties of the equipment and its components and, possibly, the extraction piping (220). The values ​​cited here should not be understood as limiting the scope of the invention and may be Larger or smaller than indicated, according to the necessary electrical power conditions.

[0040] Suitable electrical power sources (110, 111) according to the invention are direct current or pulsed alternating current sources that should enable electrical currents between 1 pA and 1 kA, preferably, but not limited to, a range between 1 mA and 100 A. The choice of electrical current intensity will depend essentially on the type of fluid to be energized, the intended energization time, and the presence or absence of any salts and / or gases and / or ions and / or water and / or sand and / or ores and / or impurities and other elements present in the fluids. The electrical power sources (110, 111) can be powered by the existing power grid or by alternative sources such as solar panels, wind turbines, etc. The values ​​and quotations should not be understood as limiting the scope of the invention and may be higher or lower than indicated, according to the necessary electro-energization conditions.

[0041] A grounding means (220-1), according to the invention, is a grounding means electrically connected to the extraction pipeline (220) and which may be from the extraction platform itself (200), interconnecting all elements, parts, components and equipment that contain components or are conductors of electrical energy and which, for the safety of the extraction platform (200), need to all be at the same electrical potential, to avoid potential differences that allow the undesirable formation of leakage currents and / or sparks and / or flashes, which are highly dangerous in oil extraction.

[0042] The grounding medium (220-1) is electrically connected to at least one spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) by means of at least one connecting cable (112), which may be any cable. or usual busbar for interconnecting elements, parts, components and equipment to the grounding. The grounding medium (220-1) can be a grounding medium already existing on the extraction platform (200) or also a grounding medium added later.

[0043] A spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n), according to the invention, is any spark gap or similar device known in the prior art and capable of acting as a device for transmitting electrical energy without permanent physical electrical contact, while a diode is any semiconductor component known in the prior art that allows the passage of electric current in only one direction. Such a spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) must exist in order to obtain the necessary conditions for the creation of the electron trap, wherein the spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) is placed between the connecting cable (112) and the conductor (113), and the electrical contact between the connecting cable (112) and, consequently, the grounding medium (220-1), and the conductor (113) only occurs by the activation of the spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) and, as already described, without mechanical contact between the connecting cable (112) and the conductor (113). The switching of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) is done by means of the processing unit (170).

[0044] The connectors (113), in turn, are any electrically conductive elements connected to the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and to the inside of the extraction pipe (220), establishing electrical contact between the fluid oil (PF), the extraction pipe (220) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), each connector (113) being a cable or a busbar or a pin.

[0045] The spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) is connected between the connection cable (112) and the connector (113), wherein the electrical contact between the connection cable (112) and the connector (113) occurs by switching the (110-1, 110-2, 110-3, ..., 110-n) spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) controlled by the processing unit (170) based on signals received from the plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n).

[0046] A high-voltage cable (120), according to the invention, is any electrically insulated cable capable of transmitting electrical energy with high voltages and / or currents, the high-voltage cable (120) being arranged coaxially to the extraction pipe (220) inside the extraction pipe (220) and electrically connected to the electrical energy source (110, 111), the high-voltage cable (120) comprising a plurality of electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120).

[0047] The power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) are any terminals capable of transmitting electrical energy in a controlled manner to the fluid petroleum (FP) within the extraction pipeline (220). The power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n), therefore, are uninsulated terminals so that they can subject the fluid petroleum (FP) to pulses of electrical voltages and currents coming from the power source (110, 111) and transmitted by the high-voltage cable (120).

[0048] The switching of the electrical pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) are controlled by the processing unit (170) based on signals received from condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), where the number of switching or pulses per unit of time applied to the power terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) determine the number of switching or pulses per unit of time of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n).

[0049] That is, for each predetermined number of switching operations of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) there will be at least one switching operation of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n).

[0050] Between each of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) there are flexible tensioning spacers (131) slidingly fixed to the high-voltage cable (120) by means of sliding fasteners (132), wherein the tensioning spacers (131) and the sliding fasteners (132) maintain both the spacing between the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the positional and coaxial stability of the high-voltage cable (120) and the power supply terminals (130-1, 130-2, 130-3, 130-4, ..., 130-n). 130-5, 130-6, ..., 130-n) inside the extraction pipe (220).

[0051] In the context of the present invention, the term "condition detector" refers to any device, equipment, or system known in the prior art capable of detecting, from the measurement of the physicochemical characteristics of fluid petroleum (FP) in its non-energized and / or energized state. Physical quantities acquired continuously or intermittently at predetermined intervals from one or more condition detectors (160-1, 160-2, 160-3, 160-4, ..., 160-n) are transmitted. to the processing unit (170) in the form of input signals, preferably comprising, but not limited to, values ​​of electrical conductivity and / or electrical resistance, pH, time values, electrical voltage, electrical current, electrical power, impedance, reactance, fluid flow rate, fluid density, fluid temperature, pressure inside the extraction pipe (220), water-in-oil emulsion composition, electric field intensity and other physical quantities applicable to the invention, the physical quantities being able to be measured directly and / or indirectly by one or more of the condition detectors (160-1, 160-2, 160-3, 160-4, ..., 160-n).

[0052] Condition detectors (160-1, 160-2, 1603, 160-4, ..., 160-n) comprise mechanical, luminous, pneumatic, analog, digital and similar sensors known from the state of the art, for measuring electrical conductivity and / or electrical resistance, pH, presence of gases, temperature, alkalinity, viscosity, fluid oil inlet flow rate (PF) and other physical quantities that assist in monitoring and decision-making by the processing unit (170) regarding the switching of electrical pulses from the power supply terminals (130-1, 130-2, 130-4, ..., 130-n) and, consequently, regarding the switching of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n). The signals generated by the plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) that will be received by the processing unit (170), are input signals corresponding to the condition and physical-chemical properties of the fluid petroleum (PF).

[0053] A processing unit (170), in the context of the present invention, refers to a computer system or a processing circuit configured to control the system (100), comprising A central processing unit, or CPU, that executes the instructions of a computer program, processing and performing arithmetic and logical operations, as well as data input and output. The computer program is stored on a computer-readable medium with memory for data storage, connection to one or more communication and data networks, and to one or more remote databases and / or a local and / or centralized and / or decentralized and / or cloud-based information storage and retrieval environment. It is also equipped with all the usual peripherals of the state of the art and is capable of exchanging information with electronic and physical media, interfaces, applications, mobile equipment, other memory devices, etc.

[0054] The processing unit (170) of the invention comprises at least one control unit and a human-machine interface (HMI) comprising information / instruction acquisition devices and information / instruction presentation devices and other devices and / or equipment connected to the system (100) operate together and may be, in groups or separately, interconnected by one or more communication and data networks.

[0055] A processing unit (170) of the invention may be part of a computer system or be divided into one or more modules of a processing circuit. The term module, according to the invention, refers to an application-specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated or group of processors) and a memory that executes one or more software programs or firmware. It also refers to a combinational logic circuit and / or other suitable components capable of providing the functionalities in question.

[0056] A "computer program" according to the invention is a program executable on a processor of the invention and thus on a processing unit (170) of the system (100) of the invention, for example, in the form of an application.

[0057] A "processing circuit" according to the invention is configured to determine a trained neural network according to the invention. This processing circuit may therefore include a processor, such as a central processing unit (CPU), a microcontroller, a microprocessor, a field programmable gate array (FPGA), a graphics card, or special hardware for convolutional neural networks such as the trained convolutional neural network of the invention.

[0058] A "computer-readable memory," in the context of the present invention, is any memory or storage device, remote or local, volatile or non-volatile, transient or non-transient (permanent), that stores information and instructions and, in particular, a computer-readable memory that stores instructions capable of executing an electro-energization method according to the invention.

[0059] A "trained neural network," in the context of the present invention, is a machine learning model configured to provide output instructions in response to the reception of input information, wherein the output instructions are unambiguously and in real time associated with each reading of input information provided by at least one condition detector (160-1, 160-2, 160-3, 160-4, ..., 160-n) of fluid petroleum (FP). A trained neural network, according to the invention, may comprise interconnected groups of artificial neurons (e.g., neuron models), and may also be a computational device or be represented as a method to be executed by a device. computational. A neural network trained according to the invention is a neural network trained with an architecture such that it delivers results quickly and accurately and that it can be run on processors embedded in the system (100) and also, alternatively, on portable processors such as, for example, but not limited to, processors in cell phones, smartphones, tablets and the like, with high processing speed and concomitant accuracy, and may also comprise layers of neurons that can be configured in a receptive field arranged side by side.

[0060] In the context of the present invention, the term "train" refers to adjusting the parameters of the machine learning model so that, from a number of input information associated with previously known values ​​of the physical-chemical conditions and characteristics of fluid petroleum (FP) to values ​​corresponding to predetermined ideal conditions, it is able to provide, as output instructions, instructions for the processing unit (170) unequivocally and in real time associated with each reading of input information provided by at least one condition detector (160, 165) of the fluid petroleum (FP), to correct at least one physical-chemical condition and characteristic of the fluid petroleum (FP) by selecting the electrical power source (110, 111), switching instructions for the electrical pulses of the power supply terminals (130-1, 130-2, 130-4, ..., 130-n) and, consequently, switching the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 130-n) ..., 110-n), so that the fluid oil (FO) conditions resemble at least one predetermined ideal condition expected for the fluid in the vicinity of the condition detectors (160-1, 160-2, 160-3, 160-4, ..., 160-n).

[0061] A primary extraction separator (230), according to the invention, is a separator for water, oil, solids and gases known in the prior art, possibly already existing on the extraction platform (200), performing the separation of the elements described above.

[0062] The primary extraction separator (230) may preferably, but without limiting the invention, be a gravity separator for separating the components of the mixture produced in oil wells (210), which generally includes oil, water, gas and solid waste. This equipment operates based on the density difference between the components, allowing them to separate naturally under the influence of gravity, where the mixture of fluids (oil, gas and water) and waste enters the separator, within which the density difference causes the gas to rise to the top, while the water and other waste, being denser, sink to the bottom, while the oil generally remains between these two layers.The gas, oil, solids and water are then extracted through separate outlets for further processing or disposal, wherein the primary extraction separator (230) comprises a primary water pipe (231), a primary oil outlet (232), a primary gas outlet (233) and a primary solids outlet (234).

[0063] The system (100), according to the invention, may further comprise a second electron trap (140) disposed downstream of the primary extraction separator (230) for additional electro-energization of both the water from a primary water pipe (231) and the oil from a primary oil pipe (232) resulting from the separation performed by the primary extraction separator (230), wherein the second electron trap (140) may further comprise a secondary water outlet (231- 1), a secondary oil outlet (232-2), a secondary gas outlet (233-1) and a secondary solids outlet (233-1).

[0064] The system (100) may also comprise at least one reel (120-1) for lowering or raising the high-voltage cable (120) inside the extraction pipe (220). System operation

[0065] Fluid petroleum (FP) extracted from the oil well (210), in the form of a water-in-oil emulsion, is subjected to electromagnetic pulses generated by the electrical power source (110, 111) and controlled by the processing unit (170) comprising a computer-readable memory and a trained convolutional neural network, based on signals from condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), with the electromagnetic pulses being transmitted to the fluid petroleum (FP) via the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) along the entire path of the fluid petroleum (FP) through the extraction pipeline (220), from the oil well. (210) to a primary extraction separator (230) of the extraction platform (200), wherein the electromagnetic pulses generated by the electrical power source (110, 111) are alternated with switching of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) controlled by the processing unit (170), in which the extraction pipe (220), together with the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), forms an electron trap due to the electrical capacitive property of the assembly formed by the extraction pipe (220), the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), ..., 110-n), and in which a. Electro-energization of fluid petroleum (PF) promotes partial and / or total separation of the water-in-oil emulsion before the primary extraction separator (230).

[0066] The electron trap formed by the extraction pipe (220) together with the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110- (2, 110-3, ..., 110-n), it works by switching the electromagnetic pulses generated by the electrical power source (110, 111) and transmitted to the fluid petroleum (PF) through the electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) followed by the switching of the spark gaps and / or diodes (110-1, 110-2, 110- 3, ..., 110-n) which, when switched, connect the electron trap to the extraction platform grounding (200), generating an electron flow which, after the switching is complete and depending on the instruction generated by the processing unit (170) and / or the neural network from the input signals of the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), and the consequent selection of the positive electrical power source (110) or the negative electrical power source (111), generates an accumulation of positive or negative charges in the fluid oil (FO) in the vicinity of the electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n), in response to conditions and physical-chemical properties read.In this way, the electron trap corrects the physical and chemical conditions and properties of fluid petroleum (FP) to adapt them to predetermined ideal values, thus enabling the separation of water, gases, solid elements, and oil from the water-in-oil emulsion that forms fluid petroleum (FP).

[0067] It should be noted that the grounding switching can occur between a certain number of switching operations of the electrical power sources (110, 111) and even before the switching operations of the electrical power sources (110, 111) to promote an initial flow of electrons.

[0068] However, switching the grounding is also important in situations where, according to the input signals from the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), there is a need to switch between an electron sequestration condition and an electron accumulation condition or vice versa, that is, when there is a need to reverse between positive and negative charges or vice versa.

[0069] Furthermore, the grounding may also be switched when there is interference in the operation of the system (100), and thus the switched grounding will be a way to reset the system (100) and / or restart the execution of the method according to the invention.

[0070] The processing unit (170) and / or the neural network will decide and control the ideal moments for switching according to each situation and, therefore, the importance of switched grounding along the extraction pipeline (220). The more grounding and power supply points, the more efficient and effective the separation of the water-in-oil emulsion will be.

[0071] If the input signals generated by the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) indicate a 50% water-in-oil emulsion, then there will be, for example, but without limiting the invention, the generation of two signals from the sensors of the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n). Since water is more conductive than oil, there will be an input signal referring to Conductivity and permittivity, in addition to an input signal referring to pH values, where water is more alkaline and oil is essentially neutral. In the case of a 50% water in 50% petroleum emulsion (W / O 50 / 50), the sensors will indicate that the emulsion has medium conductivity, that is, a medium capacity for accumulating electrical charges, and is slightly alkaline (pH of 7.5).

[0072] For situations involving emulsions with less water and more oil, such as a W / O ratio of 30 / 70, and since oil has a neutral pH and lower electrical conductivity than water, the sensors will indicate that the emulsion has low conductivity, meaning a low capacity to accumulate electrical charges, and a pH around 7.

[0073] In the case of emulsions with a higher saltwater content, for example, a 70 / 30 W / O emulsion, the sensors will indicate that the emulsion has high electrical conductivity, meaning a greater capacity to accumulate electrical charges, with a pH tending towards 8.

[0074] Based on the above examples, which are illustrative and not limiting of the invention, it is possible to determine which output instructions the processing unit (170) and / or the neural network will send to the system (100).

[0075] When the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) indicate a 50 / 50 W / O emulsion, the processing unit (170) and / or the neural network will command the selection of the positive electrical power source (110) and the alternating switching of the power terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) in order to promote the sequestration of electrons from the emulsion, for example, with electromagnetic pulses at least 1 second apart, and which may have their frequency increased or decreased, but always within a same direction of electron sequestration. It should be noted that the same applies to the presence of gases, where the decrease in the surface tension of water caused by electroacidulation will promote the separation of the liquid and gaseous phases.

[0076] In the case of 30 / 70 W / O emulsions, the processing unit (170) and / or the neural network will control the selection of the positive electrical power source (110) and the alternating switching of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) in order to promote an even greater sequestration of electrons from the emulsion, for example, with electromagnetic pulses with a minimum interval of 1 second between them, and which may have their frequency increased or decreased, but always in the same direction of electron sequestration. In this way, there will be even greater acidification of the emulsion and, thus, the formation of positive water ions that will force repulsion between oil / fatty acids and water, in which this "injection of electrons" will decrease the surface tension of the water, which will facilitate the release and detachment of oil from the water, as well as the solids and gases trapped in the water.It is worth noting that the same applies to the presence of gases, where the decrease in the surface tension of water caused by electroacidulation will promote the separation of the liquid and gaseous phases.

[0077] For emulsions with a higher saltwater content, for example, the 70 / 30 W / O emulsion mentioned in the example above, the processing unit (170) and / or the neural network will control the selection of the negative electrical power source (111) and the alternating switching of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) in order to promote an accumulation of electrons (electroalkalization), for example, with Electromagnetic pulses with a minimum interval of 1 second between them, and whose frequency may be increased or decreased, but always in the same direction of electron accumulation. In this way, there will be an alkalization or saponification of the emulsion, facilitating the separation of oil from water, due to the decrease in the surface tension of the water.

[0078] Regarding the pH of the different phases of a water-in-oil emulsion according to the invention, it is important to consider that there are well-defined pH values ​​for gases in general, where there is a clear difference between the pH values ​​of hydrocarbon gases versus carbon dioxide gases, the same being true for seawater versus crude oil.

[0079] That being said, measuring the pH of an emulsion is a significant way to identify four elements in particular: hydrocarbon gases (since these gases are not ionizable, pH sensors cannot identify them); carbon dioxide gases with high acidity (low pH); crude oil (neutral pH); and underwater water (high pH).

[0080] Hydrocarbon gases, such as methane (CH4) and propane (C3H8), are gases that, when liquefied, do not have a well-defined pH. This is because methane is a chemical compound that does not ionize in water, meaning it does not dissociate into hydrogen ions (H+) and methane ions (CH4) or propane ions (C3H8-). Therefore, there is no concentration of hydrogen ions to measure the pH. Furthermore, methane and propane are gases that do not chemically react with water, meaning there is no chemical reaction that could affect the pH of the solution. Therefore, it is not possible to determine a specific pH for liquefied methane or propane.

[0081] Carbon dioxide (CO2), in turn, is a weak acid that, when dissolved in water, forms carbonic acid (H2CO3). The pH of carbon dioxide The pH of carbonic acid dissolved in water depends on the concentration of CO2 in the solution. In general, the pH of carbonic acid is slightly acidic, with a pH between 3.5 and 5.5. Some examples of pH of carbonic acid solutions at different concentrations are: 1% CO2 in water: pH 4.5; 5% CO2 in water: pH 3.8; and 10% CO2 in water: pH 3.4. It is also important to note that the pH of carbonic acid can vary depending on factors such as temperature, pressure, and the presence of other chemical compounds in the solution.

[0082] The pH of petroleum can vary depending on its chemical composition and the presence of impurities. In general, petroleum is a non-polar liquid that does not ionize in water, meaning it doesn't have a well-defined pH. However, some studies suggest that the pH of petroleum can range from 5.5 to 8.5, depending on its chemical composition and the presence of impurities. Some examples of pH values ​​for different types of petroleum are: Light petroleum: pH 6.5 - 7.5; Heavy petroleum: pH 5.5 - 6.5; and Crude oil: pH 6.0 - 8.0. It is important to note that these values ​​are only estimates and may vary depending on the source and methodology used to measure the pH.

[0083] Finally, the pH of seawater is slightly alkaline, with an average value of around 8.1. This is because seawater contains a large amount of calcium and magnesium ions, which are alkaline. However, it is important to note that the pH of seawater can vary depending on factors such as: depth, as the pH may be lower in deep waters due to the presence of dissolved carbon dioxide; temperature, as the pH can vary with temperature, with higher values ​​in warm waters; salinity, as the pH can vary with salinity, with higher values ​​in saltier waters; and the presence of living organisms, as the pH can be affected by the presence of living organisms, such as Algae and plankton, which can produce or consume carbon dioxide. Some examples of seawater pH under different conditions are: Surface seawater: pH 8.1-8.3; Deep seawater: pH 7.8-8.0; Warm seawater: pH 8.2-8.4; and Saltwater: pH 8.3-8.5. It is important to note that these values ​​are only an estimate and may vary depending on the specific location and conditions.

[0084] The conditions for electro-energization are essentially determined by the type of source (110, 111), the voltage and current applied by the source (110, 111) to the high-voltage cable (120), and the duration of each pulse. The choice of these three parameters is made according to the type and intensity of the electro-energization which, in the case of the present invention, occurs automatically and in real time by submitting, to each reading of input information provided by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) of fluid petroleum (FP), to the machine learning model configured to provide output instructions in response to the reception of this input information, wherein the output instructions are unambiguously and in real time associated with each reading of input information provided by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) of fluid petroleum (PF), altering the intensity of the electric field of the electron trap according to predetermined ideal values, so that the electron trap promotes the electroacidulation or electroalkalization of the fluid petroleum (PF).

[0085] The selection of the source (110, 111), the command for the voltage and current values ​​of the sources (110, 111) and the control of the operating time of the sources (110, 111) are functions executed and commanded by the processing unit (170), which assigns to each operating instruction a predetermined triple protocol of source / voltage-current / time, according to the Output instructions from the processing unit (170) and / or the trained neural network. Each instruction corresponds to a suitable ionization condition, in real time, for the fluid oil (FO) conditions. It should be noted that the choice of parameters can be unique for one or more electro-energization cycles, for example, when the physical-chemical conditions and characteristics of the fluid oil (FO) are known in advance. It should also be noted that the choice of parameters can be made manually by the user through a human-machine interface (HMI).

[0086] It should also be noted that, due to the continuous real-time correction of the fluid oil (FO) ionization conditions, there may be intense variation in these ionization conditions, especially according to the fluid oil (FO) conditions during extraction, due to the constant updating of the fluid oil (FO) conditions sent by the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n).

[0087] In the invention system, electroenergization can be used both for electron sequestration (positive direction - electroacidulation) with the selection of the positive source (110) and for electron accumulation (negative direction - electroalkalization) with the selection of the negative source (111), making it possible to obtain the exact quantity of ions with the desired charges (positive or negative direction) or, also, to promote eventual adjustments and corrections of the ion levels of the fluid petroleum (FP) in process (mixed or alternating direction) to obtain a final fluid (fluid petroleum (FP)) with the desired characteristics, predetermined according to the application and intended purpose for its energization.

[0088] In the context of the present invention, the term "electron sequestration" means that, in the case of the energized fluid, negative ions migrate to the positive pole of the constant polarity electric current immersed in the fluid, causing a desired excess of hydrogen ions (H+). + ) or cations and the consequent increase in the acidity of the fluid, here called electroacidulation. The source selected in this case is the positive electrical energy source (110).

[0089] In the context of the present invention, the term "electron accumulation" means that, in the case of the energized fluid, the positive ions migrate to the negative pole of the constant polarity electric current immersed in the fluid, causing a desired excess of hydroxyl ions (OH-) or anions and the consequent increase in the alkalinity of the fluid, here called electroalkalization. The source selected in this case is the negative source (111).

[0090] According to the invention, the intensity of the electroacidulation is determined by means of the output instructions of the processing unit (170) which will instruct the selection of one or more of two or more possibilities that will be assigned by the processing unit (170) to the corresponding triple protocol(s).

[0091] It is noteworthy that, for the separation of water contained in fluid petroleum (PF), a favorable condition occurs through the electroacidification of the water-in-oil emulsion, that is, by selecting a positive electrical energy source (110), in which case there is acidification of the water contained in the emulsion and the consequent repulsion of the oil molecules, resulting in partial or total separation of the water-in-oil emulsion. On the other hand, the electroalkalinization of the water-in-oil emulsion, which occurs through the selection of a negative electrical energy source (111), can also generate a favorable condition for separation, since it causes a reduction in the surface tension of the water and promotes the saponification of this water, similarly separating the water from the oil. Since these conditions can vary over the time of fluid oil (FO) extraction, it may be necessary to perform electroacidification and / or electroalkalization alternately at the same time in different locations of the extraction pipeline (220), according to the information provided by different condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), for example, at different elevations of the extraction pipeline (220), since the physicochemical characteristics of the fluid oil (FO) can vary within the oil well (210) and the fluid oil (FO), rising through the extraction pipeline (220), is subjected to several electron traps as it ascends through the extraction pipeline (220), with the water-in-oil emulsion having its physicochemical characteristics altered with each new electron trap to which it is subjected.

[0092] The number of electrical pulses and the intervals between electrical pulses transmitted to the fluid oil (FO) through the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n), in addition to the grounding switching through the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), performed at each predetermined number of electrical pulses, will depend essentially on the physical-chemical conditions and characteristics detected continuously and in real time by the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), with the number and intervals of electrical pulses and the grounding switching being controlled by the processing unit. (170).

[0093] Electro-energization, especially of the water contained in the water-in-oil emulsion, promotes a reduction in the surface tension of the water and a consequent reduction in friction between the water molecules, where the water Its fluidity is then increased, reducing the energy expenditure for pumping fluid petroleum (FP).

[0094] The operating time of the electron trap can vary according to the output instructions of the processing unit (170) and thus depending on the instructions for executing one or more energizing switches interspersed with grounding switches according to the readings of the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n). Therefore, the time can vary from a few milliseconds to a few seconds, with the fluid flow being a direct function of the time equivalent to the triple protocol instructed by the processing unit (170) or even the triple protocol chosen by the operator via the human-machine interface (HMI).

[0095] It is especially important to emphasize that the electrical voltage applied to the fluid petroleum (FP) must be consistent with the materials used in the electron trap, and such that it overcomes the dielectric strength of the insulation in the desired locations, to allow the flow and subsequent capture and / or accumulation of electrons, after the grounding is removed, promoting the capture and / or accumulation of electrons inside the fluid petroleum (FP) and, thus, the electro-energization of the fluid petroleum (FP).

[0096] In the case of positive direction or electron sequestration, a positive differential is created, resulting in the acidification of the fluid petroleum (FP). In this process modality, the electrostatic sensitivity of the electro-energized fluid occurs between the positive charges of the fluid and the electrons of the fluid.

[0097] In the case of negative direction or electron accumulation, a negative differential is created, resulting in the alkalization of the fluid petroleum (FP). In this process modality, sensitivity The electrostatic interaction of an electrified fluid occurs between the negative charges of the fluid and the electrons of the fluid. Method

[0098] A method for separating water-in-oil emulsions by electro-energization in oil extraction, according to the invention, is a method implemented by a system (100) comprising at least one extraction platform (200), offshore or onshore, for oil exploration from an oil well (210); at least one extraction pipeline (220) for extracting fluid oil (PF) from the oil well (210); at least one positive electrical power source (110) and at least one negative electrical power source (111); at least one grounding means (220-1) electrically connected to the extraction pipeline (220) by means of a connecting cable (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and connectors (113); at least one high-voltage cable (120) arranged coaxially to the extraction pipe (220) inside the extraction pipe (220) and electrically connected to the electrical power source (110, 111); a plurality of power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); a plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); and at least one processing unit (170) for system control (100).

[0099] The method for separating water-in-oil emulsions by electro-energization in petroleum extraction is a computer-implemented method comprising: a. Install, on an oil extraction platform (200), at least one positive electrical power source (110) and at least one negative electrical power source (111); b. Insert, inside at least one extraction pipe (220) of the extraction platform (200), at least one high-voltage cable (120) arranged coaxially to the extraction pipe (220) and electrically connected to the electrical power source (110, 111), wherein the high-voltage cable (120) comprises: a plurality of electrical energizing terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); and a plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); c.Install, on the extraction platform (200), at least one grounding means (220-1) electrically connected to the extraction pipeline (220) by means of a plurality of connecting cables (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and a plurality of connectors (113); d. Start the extraction of oil in the form of fluid oil (FO) through the extraction pipeline (220); e. Receive in the processing unit (170), from the plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), input signals. corresponding to the physical-chemical condition and properties of fluid petroleum (PF); f. Submit the input signals to a trained neural network comprised of a computer-readable memory of the processing unit (170); g. Compare, using the trained neural network, the values ​​of the input signals to predetermined ideal physical-chemical condition and characteristic values; h. Generate, using the neural network, as a response to an input signal, at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) representing a value different from the predetermined ideal physical-chemical condition and characteristic values, at least one output signal in the form of at least one output instruction for the processing unit (170), wherein the output instruction comprises instructions for selecting at least one of the electrical power sources (110, 111), for switching the electromagnetic pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and for switching instructions for the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n); i. Repeat the reading of the input signals generated by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) until each input signal has a value similar to and / or equal to at least one predetermined ideal condition assigned to each of the input signals; and. j. Repeat the steps Method for training a neural network

[0100] A method for training a neural network, according to the invention, is a method for training a neural network for evaluating the physical-chemical conditions and properties of water-in-oil emulsions in a system (100) for separating water-in-oil emulsions by electro-energization in oil extraction, and generating instructions for a processing unit (170) that controls the system (100).

[0101] The system (100) comprises at least one extraction platform (200), offshore or onshore, for oil exploration from an oil well (210); at least one extraction pipeline (220) for extracting fluid oil (PF) from the oil well (210); at least one positive electrical power source (110) and at least one negative electrical power source (111); at least one grounding means (220-1) electrically connected to the extraction pipeline (220) by means of a plurality of connecting cables (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and a plurality of connectors (113); at least one high-voltage cable (120) arranged coaxially to the extraction pipeline (220) inside the extraction pipeline (220) and electrically connected to the electrical power source (110, 111); a plurality of electrical power terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); a plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); and at least one processing unit (170) for controlling the system (100).

[0102] Training the neural network involves adjusting the machine learning model parameters so that, from a number of input information comprising the physical-chemical conditions and characteristics of fluid petroleum (FP), derived from a plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) connected to the processing unit (170), associate the input information with previously known values ​​of the physical-chemical conditions and characteristics of the fluid petroleum (FP) and / or with values ​​corresponding to predetermined ideal conditions, so that the neural network can provide, as output instructions, instructions for the processing unit (170) unequivocally and in real time associated with each reading of input information provided by at least one condition detector (160) of the fluid petroleum (FP), to correct at least one physical-chemical condition and characteristic of the fluid petroleum (FP) by selecting the electrical power source (110, 111), switching instructions for the electrical pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and, consequently, switching the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), so that the fluid oil (FO) conditions resemble at least one predetermined ideal condition expected for fluid oil (FO) in the vicinity of the condition detectors (160, 165).

[0103] The method for training a neural network is a computer-implemented method comprising: A. Collect a set of input signals from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) in the form of physicochemical characteristics of the petroleum fluid (PF), in particular, but not limited to, invention, input signals relating to the electrical conductivity and / or electrical resistance of the fluid (PF); B. Configure a neural network to receive, as input, a set of predetermined ideal values ​​for different conditions and physicochemical characteristics of fluid petroleum (PF); C. Configure the neural network to receive, as input, from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), a set of input signals relating to the physical-chemical characteristics of fluid petroleum (PF); D. Unambiguously assign to each input signal at least one predetermined ideal value for different condition values ​​and physicochemical characteristics of fluid petroleum (PF) and generate a training set comprising the collected set of input signals; E. Train the neural network at a training stage using the training set; F. Create a second training set for a second training stage comprising the training set and incorrectly detected input signals and input signals with values ​​diverging from the predetermined ideal values; G. Train the neural network in a second training stage using the second training set; H. Configure the neural network to generate, in response to an input signal, at least one condition detector. (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), at least one output signal in the form of at least one output instruction for the processing unit (170), wherein the output instruction comprises instructions for switching the electromagnetic pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and instructions for switching the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n); I. Repeat the reading of the input signals generated by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) until each input signal has a value similar to and / or equal to at least one predetermined ideal condition assigned to each of the input signals; and J. To produce a neural network by repeatedly training the trained neural network with each of the training datasets. Memory read by computer

[0104] A computer-readable memory, according to the invention, is a memory comprising a set of information and instructions that, when executed, effect a method for separating water-in-oil emulsions by electro-energization in petroleum extraction, according to the invention. Conclusion

[0105] It will be readily understood by those skilled in the art that modifications can be made to the present invention without departing from the concepts set forth in the description above. These modifications must to be considered as encompassed by the scope of the present invention. Consequently, the particular embodiments described in detail above are only illustrative and exemplary and not limiting as to the scope of the present invention, to which the full extent of the appended claims and any and all equivalents thereof should be given.

Claims

CLAIMS 1. System for separating water-in-oil emulsions by electro-energization in oil extraction, characterized by comprising: i. At least one extraction platform (200), offshore or onshore, for oil exploration from an oil well (210); u. At least one extraction pipe (220) for extracting fluid petroleum (PF) from the oil well (210); iii. At least one positive electrical power source (110) and at least one negative electrical power source (111); iv. At least one grounding means (220-1) electrically connected to the extraction pipe (220) by means of a plurality of connecting cables (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and a plurality of connectors (113); V. At least one high-voltage cable (120) arranged coaxially to the extraction pipe (220) inside the extraction pipe (220) and electrically connected to the electrical power source (110, 111); vi. A plurality of power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); vii. A plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); and VÜi. At least one processing unit (170) for system control (100); 2. System according to claim 1, characterized in that the fluid petroleum (FP) extracted from the oil well (210), in the form of a water-in-oil emulsion, is subjected to electromagnetic pulses generated by the electrical power source (110, 111) and controlled by the processing unit (170) comprising a computer-readable memory and a trained convolutional neural network, based on the signals from the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), the electromagnetic pulses being transmitted to the fluid petroleum (FP) by means of the electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) along the entire path of the fluid oil (PF) through the extraction pipeline (220), from the oil well (210) to a primary extraction separator (230) of the extraction platform (200), wherein the electromagnetic pulses generated by the electrical power sources (110, 111) are alternated with switching of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) controlled by the processing unit (170), wherein the extraction pipeline (220), together with the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) ..., 110-n), forms an electron trap due to the electrical capacitive property of the assembly formed by the extraction pipe (220), the electro-energization terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n), and in which the electro-energization of the fluid petroleum (PF) promotes the partial and / or total separation of the water-in-oil emulsion before the primary extraction separator (230).

3. System according to claim 1, characterized in that the fluid petroleum (FP) is a fluid and pumpable water-in-oil emulsion extracted from the oil well (210), the emulsion also possibly comprising salts, sand, gases, ores, solid or fluid residues, ions and all kinds of impurities from the oil well (210).

4. System, according to claim 1, characterized in that the electrical energy source (110,111) is of adjustable voltage with pulsed current, which may be direct current and / or alternating current, comprising at least one positive electrical energy source (110) for electron sequestration (electroacidulation) and at least one negative electrical energy source (111) for electron accumulation (electroalkalization).

5. System according to claim 1, characterized in that the grounding means (220-1) is a grounding means electrically connected to the extraction pipe (220) and is electrically connected to at least one spark gap and / or diode (110-1, 110-2, 110-3, ..., 110-n) by means of at least one connecting cable (112), wherein the connectors (113) are electrically conductive elements connected to the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and to the interior of the extraction pipe (220).

6. System according to claim 1, characterized in that the switching of the electrical pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and the switching of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) are controlled by the processing unit (170) and / or a trained neural network, based on the signals received from the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), wherein the number of switchings or pulses per unit of time applied to the terminals of The electrical energization (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) determines the number of switching operations or pulses per unit of time of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n).

7. System, according to claim 1, characterized in that for each predetermined number of switching operations of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) there is at least one switching operation of the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n).

8. System according to claim 1, characterized in that the condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) are sensors for measuring electrical conductivity and / or electrical resistance, pH, presence of gases, temperature, alkalinity, viscosity and fluid oil inlet flow rate (PF).

9. System according to claim 1, characterized in that the processing unit (170) is a computer system or a processing circuit configured to control the system (100), comprising a CPU, at least one control unit and a human-machine interface (HMI).

10. System according to claim 1, characterized in that it may further comprise a second electron trap (140) disposed downstream of the primary extraction separator (230) for additional electro-energization of both the water from a primary water pipe (231) and the oil from a primary oil pipe (232) resulting from the separation performed by the primary extraction separator (230), wherein the second electron trap (140) may further comprise a secondary water outlet (231-1), a secondary oil outlet (232-2), a secondary gas outlet (233-1) and a secondary solids outlet (233-1).

11. Method for separating water-in-oil emulsions by electro-energization in oil extraction, characterized by being a computer-implemented method executed on a system (100) as described in any one of claims 1 to 11, the method comprising: a. Installing, on an oil extraction platform (200), at least one electrical power source (110, 111), being a positive electrical power source (110) and a negative electrical power source (111); b. Insert, inside at least one extraction pipe (220) of the extraction platform (200), at least one high-voltage cable (120) arranged coaxially to the extraction pipe (220) and electrically connected to the electrical power source (110, 111), wherein the high-voltage cable (120) comprises: a plurality of power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) electrically connected to the high-voltage cable (120) and arranged coaxially to the extraction pipe (220) and spaced along the entire length of the high-voltage cable (120); and a plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n); c. Install, on the extraction platform (200), at least one grounding means (220-1) electrically connected to the extraction pipe (220) by means of a plurality of connecting cables (112), a plurality of spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n) and a plurality of connectors (113); d. Initiate the extraction of petroleum in the form of fluid petroleum (FP) through the extraction pipeline (220); e. Receive in the processing unit (170), from the plurality of condition detectors (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), input signals corresponding to the condition and physicochemical properties of the fluid petroleum (FP); f. Submit the input signals to a trained neural network; g. Compare, using the trained neural network, the values ​​of the input signals to predetermined ideal condition and physicochemical characteristics; h. Generate, using the neural network, as a response to an input signal from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) representing a value different from the predetermined ideal physical-chemical condition and characteristic values, at least one output signal in the form of at least one output instruction for the processing unit (170), wherein the output instruction comprises instructions for switching the electromagnetic pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and instructions for switching the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n); i. Repeat reading the input signals generated by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) until each input signal has a value. similar to and / or equal to at least one predetermined ideal condition assigned to each of the input signals; and j. Repeat the steps 12. Method for training a neural network, characterized by being a computer-implemented method for training a neural network for evaluating the physical-chemical conditions and properties of water-in-oil emulsions in a system (100) as described in any of claims 1 to 11, the method comprising: A. Collect a set of input signals from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) in the form of physical-chemical characteristics of the petroleum fluid (PF), in particular, but without limiting the invention, input signals relating to the electrical conductivity and / or electrical resistance of the fluid (PF); B. Configure a neural network to receive, as input, a set of predetermined ideal values ​​for different conditions and physicochemical characteristics of fluid petroleum (PF); C. Configure the neural network to receive, as input, from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), a set of input signals relating to the physical-chemical characteristics of fluid petroleum (PF); D. Unambiguously assign to each input signal at least one predetermined ideal value for different conditions and physicochemical characteristics of the fluid petroleum. (PF) and generate a training set comprising the collected set of input signals; E. Train the neural network at a training stage using the training set; F. Create a second training set for a second training stage comprising the training set and incorrectly detected input signals and input signals with values ​​diverging from the predetermined ideal values; G. Train the neural network in a second training stage using the second training set; H. Configure the neural network to generate, in response to an input signal from at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n), at least one output signal in the form of at least one output instruction to the processing unit (170), wherein the output instruction comprises instructions for switching the electromagnetic pulses of the power supply terminals (130-1, 130-2, 130-3, 130-4, 130-5, 130-6, ..., 130-n) and instructions for switching the spark gaps and / or diodes (110-1, 110-2, 110-3, ..., 110-n); I. Repeat the reading of the input signals generated by at least one condition detector (160-1, 160-2, 160-3, 160-4, 160-5, 160-6, ..., 160-n) until each input signal has a value similar to and / or equal to at least one predetermined ideal condition assigned to each of the input signals; and J. To produce a neural network by repeatedly training the trained neural network with each of the training datasets.

13. Computer-readable memory, characterized by comprising a set of information and instructions which, when executed, perform a method for separating water-in-oil emulsions by electro-energization in petroleum extraction, as described in claim 12.

Images