System for virtualising the centraliser cabinet in single-phase transformer banks and / or single-phase reactor banks of digital substations, method for monitoring and switching defective phases, method for training a neural network for monitoring and switching defective phases, and corresponding computer-readable memory

Abstract

The present invention relates to a system for virtualising the centraliser cabinet in single-phase transformer banks and single-phase reactor banks of digital substations, wherein the virtualisation eliminates the need for the physical centraliser cabinet through the provision of merging units within the phase control cabinets of the respective equipment. The present invention also relates to a computer-implemented method for monitoring and switching defective phases (A, B, C) of the single-phase transformers and / or single-phase reactors of the digital substation without human manual intervention. The present invention further relates to a computer-implemented method for training a neural network for monitoring the normal or abnormal operating conditions of the phases (A, B, C) of transformer banks and / or reactor banks, and to computer-readable memories comprising a set of instructions which, when executed, carry out the method for monitoring and switching defective phases in a digital substation.

Description

Virtualization system for centralizing cabinets in single-phase transformer and / or reactor banks of digital substations, phase monitoring and switching method. DEFECTIVE PHASE SWITCHING METHOD FOR TRAINING A NEURAL NETWORK FOR MONITORING AND SWITCHING DEFECTIVE PHASE SYSTEMS AND MEMORY READ BY CORRESPONDING COMPUTER Field of application

[0001] The present invention pertains to the field of circuit arrangements and systems for the supply or distribution of electrical energy, more specifically to the supervision of devices, systems and methods for the operation and monitoring of electrical substations for the transformation of electrical current and electrical potential. Finally, the present invention also relates to the field of computer-implemented arrangements based on machine learning. Introduction

[0002] The present invention relates to a virtualization system for the centralizing cabinet in single-phase transformer banks and single-phase reactors of digital substations, wherein virtualization involves the elimination of the physical centralizing cabinet by arranging MUs within the phase control cabinets of the respective equipment.

[0003] The present invention also relates to a method for monitoring and switching defective phases of single-phase transformers and / or single-phase reactors in a digital substation without manual human intervention.

[0004] The present invention also relates to a method for training a neural network to monitor the normal or abnormal operating conditions of the phases of transformer banks and / or banks of reactors and also computer-readable memories containing a set of instructions that, when executed, perform the method of monitoring and switching defective phases in a digital substation. Fundamentals of the invention

[0005] The Protection, Control and Supervision Systems (PCSS) of electrical substations are being improved, undergoing a revolution in data acquisition and dissemination techniques, and due to the digitization or virtualization of one or more substation components, these new and modern substations are also called digital substations.

[0006] The electrical substations of the nature discussed here are, preferably, but without limitation to the invention, power generation, distribution, transmission, storage, industrial and substation substations, fixed and / or mobile, with skids of all types, especially digital substations comprising digital transformers and / or digital reactors. These equipment may be of the three-phase type, when the phase windings are installed in the same tank, or of the single-phase type, when each winding is installed in a tank, forming a bank of single-phase equipment. It should be noted that, in the case of single-phase equipment, there may be one or more spare phases, depending on the importance of the substation and / or the number of transformer and reactor banks installed.

[0007] In three-phase transformers and reactors, all signals for protection, control, and supervision are available at terminals in the respective control cabinet. In single-phase systems, the signals for each phase are also available at terminals in the respective control cabinet, but they are concentrated in so-called centralizing cabinets, which receive the signals. Signals from all phases that make up that bank, plus the reserve phase, if any. In all these cases, the LEDs installed in the control room or command center receive these signals through control cables.

[0008] Digital substations essentially comprise primary equipment and MUs at the process or yard level, LEDs and / or CPRs and / or VPRs for control and protection at the process or bay level, and supervision and control equipment at the station level. Yard-level equipment is connected to process-level equipment via the so-called process bus, while process-level equipment is connected to both the process bus and station-level equipment via the so-called station bus, as can be seen in Figure 1 attached.

[0009] In current state-of-the-art solutions, the station bus, represented by the blue line in Figure 1 attached, carries communication protocols between the LEDs and / or CPRs and / or VPRs via GOOSE messages, and between the LEDs and the supervisory system via MMS messages. These protocols are defined in the IEC-61850-8-1 standard.

[0010] The process bus, represented by the red line in Figure 1 attached, carries GOOSE protocols (IEC-61850-8-1 standard) and Sampled Values ​​or SVs (IEC-61850-9, IEC-61950-9-2LE and IEC-61869-9 standards), in addition to the PTP time synchronization protocol (defined in IEC-61588 and IEC-61850-9-3 standards) for LEDs, CPRs, VPRs and MUs.

[0011] The design of a transformer bank in a substation essentially comprises the design of the phase control cabinet (typical design for all phases, including the spare phase) along with The design of the centralizing cabinet, highlighted in red in Figure 2, attached, where Figure 2 represents a known architecture from the state of the art.

[0012] A three-phase transformer can be constructed from three identical single-phase transformers, forming a transformer bank for high-voltage applications, generally corresponding to values ​​above 100kV, but more preferably above 230kV, or also in special applications. Each single-phase transformer corresponds to one of the three phases, for example, phases A, B and C or phases 1, 2 and 3, depending on the conventional nomenclature, and there may also be a spare phase, for example, a phase R, depending on the conventional nomenclature.

[0013] Single-phase transformer and reactor banks are commonly installed in power transmission and generation substations, since it is easier to replace a single-phase machine than a three-phase machine in case of defects or malfunctions.

[0014] After the assembly and testing of each transformer that makes up a bank, including the reserve phase, plus the centralizing cabinet, these are transported to the construction site and the field allocation activity begins, commissioning of each phase of the bank, including the reserve, laying of control cables, point-to-point testing of these cables, commissioning with the SPCS with the activity of replacing the phases with the reserve phase.

[0015] In current state-of-the-art solutions, as shown in Figure 2 attached, the centralizing cabinet receives all the wiring from phases A, B, and C that make up a bank and, possibly, the wiring from a spare phase or phase R, with all the wiring coming from the Cabinet for the control room. Each phase is connected to a column of sockets by a set of three sockets per phase, where each socket is composed of a multipolar socket, which may be, for example, a female plug (fixed) and male plugs (movable).

[0016] When one of the phases needs to be replaced, the respective male plugs are manually disconnected from their original position and connected to the plugs of the spare phase. This replaces the signals originating from the replaced phase and utilizes the signals from the spare phase.

[0017] In these cases, it is evident that disconnecting the main phase busbar and connecting it to the spare phase busbar is a manual intervention, especially for the primary and secondary sides of the bank. Regarding the tertiary side, the intervention is generally carried out through busbars within the medium-voltage cubicles, but always manually. In both cases, the operator is exposed to inclement weather and the risk of electric shock when manually intervening in plugs on energized cables.

[0018] The challenge, therefore, lies in providing a digital substation comprising MUs installed as close as possible to the primary equipment of the substation, for example, in the phase control cabinets of the respective equipment, eliminating centralizing cabinets and enabling the replacement of one or more phases by one or more spare phases in a digital and automated manner, without the need for manual human intervention, where the MUs are responsible for acquiring the analog signals from the current transformers and temperature transducers, as well as the alarm, shutdown, supervision and command signals, making them available in communication protocols such as, but not limited to, SV and GOOSE protocols. State of the art

[0019] An example of solutions aimed at virtualizing procedures in transformer banks is the document entitled "Automatic Switching System for Rapid Energization of Reserve Phase in Single-Phase Transformer Banks," published on the website https: / / www.treetech.com.br / sistema-de-chaveamento-automatico-para-rapida-energizacao-de-fase-reserva-em-bancos-de-transformadores-monofasicos / . This document reveals, on page 4, that a "[...] central control unit performs all the interlocking logic as well as the switching sequences, both to replace a faulty phase with a reserve phase and to return to the original phase, always obeying the established preconditions, such as, for example, the bank must be isolated on the 230, 138, and 13.8kV sides."These sequences can be performed in automatic mode, where the entire replacement operation is done without the need for operator intervention, and in manual mode, where each opening or closing of disconnectors is supervised step-by-step by the operator [...].

[0020] Although this solution aims to automate the switching of the disconnectors themselves, it still requires the existence of a centralizing cabinet and the need for manual operator intervention, as described on page 4 of the aforementioned prior art document, where it reads "Before starting the transfer process, the operator must transfer the plugs from the faulty phase to the backup phase. This process is manual and is part of the transfer preconditions".

[0021] As can be inferred from the previous descriptions, there is room for a virtualization system of the centralizing cabinet in single-phase transformer banks and single-phase reactors in substations. digital and for a method of monitoring and switching faulty phases of single-phase transformers and / or single-phase reactors in the digital substation without manual human intervention. Objectives of the invention

[0022] One of the objectives of the present invention is, therefore, a virtualization system for the centralizing cabinet in single-phase transformer banks and single-phase reactors of digital substations, according to the characteristics of claim 1 of the attached claims.

[0023] Another objective of the present invention is a method for training a neural network for monitoring and switching defective phases, according to the characteristics of claim 8 of the attached claims.

[0024] Yet another objective of the present invention is a method for monitoring and switching defective phases, according to the characteristics of claim 12 of the attached claims.

[0025] Another objective of the present invention is to provide computer-readable memories containing a set of instructions that, when executed, perform the method of monitoring and switching defective phases in a digital substation, according to the characteristics of claim 14 of the attached claims.

[0026] Other features and details of the features are represented by the dependent claims. Brief description of the figures

[0027] 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 exemplary embodiments that are not limiting to the scope of the present invention, in which, schematically: Figure 1: presents a schematic view of a busbar in a state-of-the-art substation; Figure 2: presents a schematic view of a conventional connection scheme for transformer banks with centralizing cabinets in a state-of-the-art substation; Figure 3: presents a schematic view of a connection scheme for virtualized transformer banks according to the invention; and Figure 4: presents a logical scheme of the method for monitoring and switching defective phases by signals according to the invention. Detailed description of the invention

[0028] 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.

[0029] The present invention relates to a virtualization system for the centralizing cabinet in single-phase transformer banks and single-phase reactors of digital substations, wherein virtualization implies... Elimination of the physical centralizing cabinet, with the MUs being installed inside the phase control cabinets of the respective equipment.

[0030] The present invention also relates to a method for monitoring and switching defective phases of single-phase transformers and / or single-phase reactors in a digital substation, in the event of abnormal operation of at least one of the phases (A, B, C) and the spare phase (R) of at least one bank of single-phase transformers and / or single-phase reactors, without manual human intervention.

[0031] Furthermore, the present invention relates to a method for training a neural network for monitoring the operating conditions of at least one of the phases (A, B, C) and the reserve phase (R) and automatic replacement of defective phases of at least one transformer bank and / or reactor bank, and also to computer-readable memories containing a set of instructions that, when executed, perform the method of monitoring and switching defective phases in a digital substation.

[0032] In the context of the present invention, the term "substation" refers to any and all electrical power substations for the transformation of electrical current and / or electrical potential, whether or not they are part of an electrical power distribution system, according to the known definitions of the prior art. The term "digital substation" (DS) refers to a virtual substation according to the invention, in which the digitization of data exchanged between one or more components of the DS occurs as close as possible to the origin of the data.

[0033] In the context of the present invention, the term "current transformer" (CT) refers to any and all electromagnetic devices known in the prior art that transform a high-voltage electric current. The invention relates the current to a value in a low-value current and vice versa, maintaining the same phase relationship between the primary and secondary currents, performing tasks known in the prior art such as, but not limited to, current measurement, circuit protection, isolation between high-voltage and low-voltage circuits, etc.

[0034] In the context of the present invention, the term "potential transformer" (PT) refers to any and all electromagnetic devices known in the prior art that transform a high-value electrical voltage into a low-value voltage and vice versa, maintaining the same phase relationship between the primary and secondary voltages, performing tasks known in the prior art such as, but not limited to, current measurement, circuit protection, isolation between high-voltage and low-voltage circuits, etc.

[0035] In the context of the present invention, the term "reactor" refers to any and all devices used to control electric current, voltage, and other variables in power distribution systems. They can be used to improve power transmission efficiency, reduce energy losses, protect sensitive equipment, and maintain the stability of the electrical system. Reactors can be classified in different ways, such as bus reactors, line reactors, and compensating reactors, each with its specific function within the electrical system.

[0036] In the context of the present invention, the term "power transformer" refers to any and all interface devices between two different circuits, generally one high voltage and the other low voltage. Power transformers can be classified in different ways, such as step-up transformers, step-down transformers, etc. Phase-shifting transformers, each with its specific function within the electrical system.

[0037] In the context of the present invention, the term "IED" (abbreviation for Intelligent Electronic Devices) refers to any and all devices known in the prior art comprising an integrated microprocessor-based controller for power system equipment, such as, but not limited to, circuit breakers, transformers, capacitor banks, as well as protection and control relays and the like.

[0038] In the context of the present invention, the term "CPR" (abbreviation for Centralized Protection Relay) refers to any protection system composed of one or more protection devices applied, in a dedicated manner, to each line, busbar, transformer, reactor, etc. of the electrical substation. This relay continuously monitors electrical power and acts as an automatic safety device, detecting faults and opening circuit breakers to isolate the defective equipment and prevent damage to the entire system, where a centralized protection system consolidates the logic of individual relays. In short, a CPR results from the combination of several LEDs in a single hardware, allowing it to supervise several transmission lines, several transformers, several reactors, several busbars, several feeders, depending on the size of the substation.

[0039] In the context of the present invention, the term "VPR" (abbreviation for Virtualized Protection Relay) refers to a protection system in which dedicated physical relays are replaced by software running on a virtualized platform, consolidating the CPR logic in a virtual environment.

[0040] In the context of the present invention, the term "SKID" refers to any and all known prior art chassis-shaped structures upon which a power substation is disposed.

[0041] In the context of the present invention, the term "MU" (abbreviation for Merging Units) refers to any data acquisition unit for digitizing analog signals from current transformers (CTs) and potential transformers (PTs) and binary signals from substation yard equipment, also known in the prior art. It basically consists of an electrical circuit that receives analog voltage and current values, an analog-to-digital converter that samples the signals, and a processor that standardizes the digital signal to work in accordance with standards such as, but not limited to, the IEC 61850 standard. The internal processing uses distinct blocks, which causes a delay in the signal. To correct this delay, an external time synchronizer (GNSS or similar) must be used.

[0042] In the context of the present invention, the term "GOOSE" (abbreviation for Generic Object Oriented Substation Event) refers to generic substation events (which is a control model defined according to standards such as, but not limited to, the IEC 61850-8-1 standard), being a protocol that functions as a fast and reliable mechanism for transferring event data across entire electrical substation networks, including, but not limited to, binary supervisory messages, alarms, and commands, made available in message format for transmission over Ethernet networks.

[0043] In the context of the present invention, the term "SV" (abbreviation for Sampled Values) refers to the message protocols responsible for transmitting instantaneous values ​​of... primary voltage and current quantities of the power substation system, including, but not limited to, sampled analog values ​​made available in message format for transmission over Ethernet networks, as defined, for example, but not limited to, in standards IEC-61850-9, IEC-61950-9-2LE and IEC-61869-9.

[0044] In the context of the present invention, the term "MMS" (abbreviation for Manufacturing Message Specification) refers to an international standard protocol, defined in a standard, for example, but not limited to the invention, in the IEC61850 standard, which deals with messaging systems for real-time process data transfer and supervisory control information between networked devices or computer applications, including, but not limited to the invention, client / server type communication between IED and SCADA.

[0045] In the context of the present invention, the term "QPEX" (abbreviation for Operational Expenditure) refers to the operational expenses necessary for the functioning of a substation.

[0046] In the context of the present invention, the term "CAPEX" (abbreviation for Capital Expenditure) refers to expenses related to long-term investments, generally in physical or intangible assets that impact the structure of the SE and serve to generate future returns.

[0047] In the context of the present invention, the term "SCADA" (abbreviation for Supervisory Control and Data Acquisition), also called SCADA supervisory system, refers to systems that use software to monitor and supervise the variables and devices of industrial processes, not Industrial equipment and devices connected via specific communication servers / drivers.

[0048] In the context of the present invention, the term "Ethernet Network" refers to any and all wired computer networking technologies commonly used in local area networks, metropolitan area networks, and wide area networks.

[0049] In the context of the present invention, the term "PTP" (abbreviation for Precision Time Protocol) refers to any protocol used to synchronize the system time across multiple systems on a LAN to a common main clock on the LAN, especially important in latency-sensitive environments, such as in the case of a digital substation of the invention, being a high-precision time synchronization signal made available via Ethernet network, defined in standards such as, but not limited to, the IEEE 1588v2 standard.

[0050] In the context of the present invention, the term "GNSS" (abbreviation for Global Navigation Satellite System) refers to any and all equipment known in the prior art that uses a constellation of satellites orbiting the earth to provide positioning, time, and navigation information to users anywhere on the planet. In power substations, GNSS is used for various purposes, mainly for time synchronization and precise positioning, crucial information for the coordinated and efficient operation of electrical networks, ensuring that devices and equipment are operating within the same time interval, which is essential for the reliable transmission and distribution of energy.

[0051] In the context of the present invention, the term "phase selector switches" refers, in a broader sense, to any and all devices combined to perform sectioning and / or switching, generally without the presence of electric current, and therefore capable of connecting and disconnecting electrical circuits and also of switching between electrical circuits, such as, for example, but without limiting the invention, disconnecting and isolating the circuit of a faulty phase and connecting it to the circuit of a spare phase.

[0052] In the context of the present invention, the term "SNMP" (abbreviation for Simple Network Management Protocol) refers, in a broader sense, to any and all standard internet protocols for managing devices on IP networks, including, but not limited to, routers, computers, servers, workstations, printers, racks, etc., and, in a more restricted sense, to any and all protocols used to report what is happening on network assets, in this case GNSS and switches.

[0053] In the context of the present invention, the term "SCL" (abbreviation for Substation Configuration Description Language) refers to the language and representation format specified in standards such as, but not limited to, the IEC 61850 standard for configuring electrical devices in substations. Substation configuration language (SCL) is a standardized language used to describe and configure the components and systems within an electrical substation, allowing different protection, control, and monitoring devices to be configured and interoperable, facilitating integration and communication between them. SCL specifies a hierarchy among configuration files by defining four groups: Specification description file The invention relates to the System Interface Card (SSD), IED Capabilities Description File (ICD), Substation Configuration Description File (SCD), IED Configuration Description File (CID), Instantiated IED Description File (IID), and System Interchange Description File (SED). Files such as ICD, CID, SSD, SCD, IID, and SED, among others, are related to the SCL and are generally part of standards, such as, but not limited to, the IEC 61850 standard. The SCL file and its derivatives are essential to ensure the interoperability and proper functioning of devices within an electrical substation, facilitating the configuration, maintenance, and operation of the electrical system.SCL files can be created in various formats, such as UML (Unified Modeling Language), XML (Extensible Markup Language), IEC 61850 (a communication standard for substation automation), and other proprietary formats used by software in the field of electrical engineering.

[0054] In the context of the present invention, the term "SCD" (abbreviation for Substation Configuration Description) refers to files that can be created in various formats, such as XML (Extensible Markup Language), IEC 61850 (a communication standard for substation automation), among other proprietary software formats used in the field of electrical engineering, comprising a complete description of the substation details, containing, for example, information on all existing IEDs in the substation, communication configuration data, and a complete description of the substation, including information on: - The substation topology containing details about the physical structure of the substation, such as the arrangement of equipment, layout of circuit breakers, transformers, relays, disconnect switches, phase selector switches, busbars and other devices; - Equipment configuration, such as specific information about each component, including its technical characteristics, nominal capacities, manufacturing data, and electrical connections; - Diagrams and schematics such as drawings and diagrams that illustrate the arrangement of equipment and the interconnection between them, which may include single-line, three-line, logic diagrams and other schematics that represent the electrical configuration of the substation; and - Control and protection information, such as details about the control, protection, automation, and monitoring systems used in the substation, including operating logic, configuration of protection relays, SCADA systems, and other control devices.

[0055] In the context of the present invention, the term "Cl D" (abbreviation for Configured IED Description) refers to a protocol used for communication between an IED configuration tool and an IED, containing specific information about how each individual IED is configured within the substation, including operating parameters, communication settings, protection settings, among others.

[0056] In the context of the present invention, the term "IIP" (abbreviation for Instantiated IED Description) refers to a file that defines the configuration of an IED for a project and is used as a data exchange format from the IED configurator to the system configurator.

[0057] In the context of the present invention, the term "SEP" (abbreviation for System Exchange Description) refers to a file that must be exchanged between system configurators of different projects.

[0058] In the context of the present invention, the expression "operating condition" should be understood, broadly, as the current real-time operating state of one or more components of the substation and / or the communication between two or more components of the substation and / or the communication between at least one of the substation components and the SCADA system and, more narrowly, as the operating condition of the transformer and / or reactor banks, especially, but without limiting the invention, of the phases (A, B, C) and at least one spare phase (R), when present, which may include, but without limiting the invention, one or more normal operating conditions and / or one or more abnormal operating conditions of the phases (A, B, C, R).This comparison, performed by the processing unit, results in output data and / or instructions in the form of warnings and / or alarms of deviation and / or inconsistency and / or also in the form of suggestions and / or procedural instructions for maintaining the operation of the phases (A, B, C, R) and / or correction of the parameters of one or more components of the substation or the substation as a whole and, in the case of using a neural network for monitoring and switching defective phases, the output data and / or instructions will be generated in the form of warnings and / or alarms and also suggestions and / or. Instructions for maintaining the operation of phases (A, B, C, R) and / or correcting the parameters of one or more components of the substation or the substation as a whole.

[0059] In the context of the present invention, a normal operating condition, or simply normal condition, indicates that the transformer and / or reactor banks are operating normally with their respective phases (A, B, C) and that the reserve phase (R) is not connected to the power system. It should be noted that a normal operating condition is preferably, but without limiting the invention, a condition in which there is no exchange of fault and / or defect messages between at least one MU of the respective bank and at least one IED and / or CPR and / or VPR.

[0060] In the context of the present invention, an abnormal operating condition, or simply an abnormal condition, indicates that at least one of the phases (A, B, C) of the transformer and / or reactor bank is damaged and / or inoperative and / or operating with a defect and / or has suffered an accident and that it will not be able to return to operation in a short time, making it necessary to replace it with at least one spare phase (R).

[0061] At the low-voltage signal level, these are switched through blocks and releases in programmed logic within the LEDs and / or CPRs and / or VPRs, by means of message protocols such as, for example, but without limiting the invention, GOOSE messages and / or SVs, which are described, for example, but without limiting the invention, in SCL type files of the SE.

[0062] It should be noted that the operating conditions of the invention take into account the result of the comparison between the expected and ideal parameters for each of the phases (A, B, C) and reserve (R) of at least one of the component banks of the SE and evaluated by the processing unit and / or LEDs and / or CPRs and / or VPRs (or by a neural network of the processing unit). trained according to the machine learning model of the invention), in particular, but without limiting the invention, of one or more phases (A, B, C) of transformer and / or reactor banks, and may include, in addition to monitoring the integrity, completeness and synchronization of the exchanged messages, additional parameters relating to the values ​​contained in the messages, i.e., the physical quantities read from the LEDs and / or CPRs and / or VPRs versus the expected values.

[0063] As described above, it is possible, by associating the parameters of the received messages with the expected values ​​and ranges, in addition to identifying a normal operating condition, to obtain information indicative of one or more abnormal operating conditions of one or more phases (A, B, C) of one or more transformer and / or reactor banks. This information allows the monitoring of the phases (A, B, C) and helps to accurately determine the need to replace at least one of the defective phases (A, B, C) with at least one spare phase (R). It should be noted that, even after replacing one of the defective phases (A, B, C) with at least one spare phase (R), the system of the invention will continue to monitor the operating conditions of the phases (A, B, C) and the spare phase (R), in particular, blocking the use of the spare phase (R) in operation by one or more other transformer and / or reactor banks of the substation.It should also be noted that, under normal operating conditions of the phases (A, B, C), the normal operating condition of the reserve phase (R) will be the condition of being switched off or inactive. Furthermore, as already mentioned above, a normal operating condition is preferably, but without limiting the invention, a condition in which there is no exchange of fault and / or defect messages between at least one MU of the respective bank and at least one IED and / or CPR and / or VPR.

[0064] In the context of the present invention, the term "monitor" refers to one or more tasks of monitoring an operating condition of one or more phases (A, B, C) and the backup phase (R), also including terms related to monitoring and their synonyms and / or equivalent terms such as, but not limited to, supervision, tracking, surveillance, verification, evaluation, measurement, etc.

[0065] Also in the context of the present invention, the term "neural network" refers to a machine learning model that can be a convolutional neural network, where the term "train" refers to adjusting the parameters of the machine learning model so that, from a number of reference parameters associated with the ideal and / or expected values ​​for the phases (A, B, C) and reserve phase (R), it is able to, from one or more input data from the LEDs and / or CPRs and / or VPRs, provide at least one output data or instruction and, in case of an abnormal operating condition, command the replacement of a defective phase (A, B, C) with a reserve phase (R). It should be noted that, in addition to a convolutional neural network, it is also possible to use other learning algorithms such as, for example, Q-learning or even LLM (abbreviation for the English term Large Language Model) and other appropriate networks and / or algorithms.

[0066] In the context of the invention, a central server is a computer or computing system or computing circuit, or even a set of executable instructions on a pre-existing SE computer, with one or more centralized computing systems or one or more data processing centers, that makes services and resources available or stores them on and through a communication and data network. It comprises one or more electronic processors capable of executing tasks from a A computer program or set of instructions stored in a computer-readable medium, preferably a computer equipped with a processor, 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, and also equipped with all the usual peripherals of the state of the art, being capable of exchanging information with the electronic and physical medium, interfaces, applications, mobile equipment, other memory devices, etc.

[0067] A server can be at least a web server that delivers or serves web pages, an application server that handles application operations between users and applications or databases, a cloud server, a database server, a file server, a service server, a game server implementing games or services for a game, and a media server providing media such as streaming video or audio.

[0068] The invention's server can also be a remote control center for the user / operator / client / maintainer of the SE.

[0069] A computer program according to the invention is a program that is executable on a processor of the invention and, thus, also executable on a processor of the portable equipment of the invention, for example, in the form of an application.

[0070] A database, according to the invention, is any set of data, files, information, instructions, and records that form organized collections of interrelated data, hosted in one or more memories or storage devices. of the invention's system and which can be accessed, fed and managed by the invention's data processing centers.

[0071] Both the servers and the processors and other equipment and devices of the system of the invention may comprise one or more databases, independent and / or interconnected.

[0072] A communication and data network, in the context of the present invention, comprises a centralized or decentralized network that interconnects one or more active or passive components of the system according to the invention. The servers, processors, data processing center, database, and portable equipment of the invention may be connected to one or more communication and data networks, Ethernet networks, physical memories, cloud storage and the like, the internet, one or more data and / or program clouds, computer terminals, mobile devices, telephone devices, barcode readers, QR codes, DataMatrix codes and the like, credit cards, NFC or BLE devices, gas stations and service stations, in short, to any equipment or interface necessary, directly or indirectly, for the execution of a method according to the invention.

[0073] The communication and data network may comprise any wired or wireless connection, the internet, or any other form of communication, and may include any number of different communication and data networks between any server, device, resource, and system, and / or other servers, devices, resources, and systems described in this document. The communication and data network may enable communication between various computing resources or devices, servers, and systems, and may employ different types of networks, for example, but not limited to, computer networks, telecommunications networks (e.g., Cell phones), mobile wireless data networks, cable, radio and similar networks, and any combination of these and / or other networks.

[0074] A processor is, therefore, a processing unit and / or IED and / or CPR and / or VPR that executes the instructions of a computer program or application, processing and executing 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.

[0075] A processing unit according to the invention may be, form part of, or be divided into one or more modules. 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.

[0076] A processing circuit according to the invention can also be configured to determine a neural network according to the invention.

[0077] This processing circuit can 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.

[0078] In the context of the present invention, a computer-readable memory or computer-readable medium is any memory or storage device, remote or local, volatile or non-volatile, transient or non-transient (permanent). System according to the invention

[0079] A virtualization system for the centralizing cabinet in single-phase transformer banks and / or single-phase reactor banks of digital substations, according to the present invention, presupposes the existence of at least one power substation (PS) comprising at least one single-phase transformer bank and / or at least one single-phase reactor bank, at least one SCADA system, at least one Ethernet network, at least one GNSS system, one or more switches, one or more circuit breakers, one or more LEDs and / or CPRs and / or VPRs, one or more actuators, transformers in general, capacitor banks, storage batteries, compensators, generators, possibly one or more electric motors and / or fans and / or heat exchangers, smoke detectors, fire alarms, audible and / or visual alarms, and all other components necessary for the operation of a substation.

[0080] The system of the invention comprises at least one processor or processing unit, independent or not of the SE SCADA, this processing unit comprising at least one memory that stores computer-executable information and instructions and at least one trained convolutional neural network, in which the processing unit and / or IED and / or CPR and / or VPR is configured to provide Output data and / or output instructions regarding one or more operating conditions of one or more single-phase transformer banks and / or single-phase reactor banks, in response to receiving, as input information, one or more messages and / or input information from each of the phases (A, B, C) and the reserve phase (R), obtained from at least one MU of the bank and transmitted to one or more LEDs and / or CPRs and / or VPRs, wherein the execution of the instructions performs one or more steps of the monitoring and switching method of the phases (A, B, C, R) of the single-phase transformer banks and / or single-phase reactors of the digital substation.

[0081] A system of the invention comprises at least one MU for each bank of single-phase transformers and / or for each bank of single-phase reactors, at least one protection IED and / or CPR and / or VPR for each bank of single-phase transformers and / or for each bank of single-phase reactors, at least one control IED and / or CPR and / or VPR for each bank of single-phase transformers and / or for each bank of single-phase reactors, at least one phase selector switch for each phase (A, B, C) and spare phase (R) of a bank of single-phase transformers and / or for each bank of single-phase reactors for selecting the busbar to which each phase (A, B, C) and spare phase (R) will be connected.

[0082] It should be noted that the virtualization of the centralizing cabinet of a single-phase bank, according to the invention, can be performed using only one MU installed in the control cabinet of each phase (A, B, C). However, since the MU will digitize any and all types of signals existing in the equipment, such as, but not limited to, current signals, intrinsic protections, forced ventilation, tap changer, supervisions, and any other type of signal, it is important that there be... Redundancy, that is, at least two MUs per phase control cabinet. These MUs must be connected to a communication bus, which may be the station bus, but preferably, without limiting the invention, the process bus, making the digitized information available in multicast communication protocols such as GOOSE and SV, but not limited to these protocols or this type of protocol. In this way, the LEDs and / or CPRs and / or VPRs can use the signals in their logic and virtualize the centralizing cabinet.

[0083] Thus, in the event of the detection of an abnormal operating condition, through a command originating from the SCADA and / or processing unit, for example, the IED and / or CPR and / or VPR will verify if the equipment bank is de-energized, will execute the commands to open and close the phase selector switches to isolate and replace the defective phase (A, B, C), and the LEDs and / or CPRs and / or VPR will block the messages received from the MUs of the defective phase (A, B, C), replacing them with those of the reserve phase (R).

[0084] In the context of the present invention, a phase selector switch is preferably, but without limitation to the invention, an electrical device used to disconnect and isolate parts of an electrical circuit and switch between circuits, wherein these phase selector switches must be able to react to an instruction from the processing unit and / or SCADA and / or LEDs and / or CPRs and / or VPRs, whose main function is to ensure safety during maintenance, repair operations or in case of faults in the electrical circuit, and may be, for example, but without limitation to the invention, motorized phase selector switches for remote control.

[0085] In addition to at least one memory, at least one processing unit, and at least one trained neural network, the system of the invention may also comprise: - At least one human-machine interface (HMI) and / or processing equipment for acquiring and presenting information / instructions; - At least one database; - At least one communication and data network; - One or more sets of input information from the equipment; - One or more sets of data and / or output instructions from LEDs and / or CPRs and / or VPRs; and Other devices and / or equipment necessary for the operation of the system and execution of a method in accordance with the invention.

[0086] The MUs, LEDs, CPRs, VPRs, memory, processors, processing equipment for information / instruction acquisition and presentation, servers, databases, communication and data networks, and other devices and / or equipment eventually and additionally included in the system, are all interconnected by the communication buses of the SE and / or by one or more communication and data networks. Images and data are stored as one or more electrical signals, and the processing of these signals is done by one or more components of the system of the invention.

[0087] With the installation of multi-phase modules (MUs) in the control cabinets of each phase (A, B, C) that make up a bank of single-phase equipment, including the reserve phase (R), all signals are digitized (current, alarms, shutdowns, supervision, commands, etc.) and made available on the bus or process network through multicast protocols such as, for example, but without limiting the invention, GOOSE messages and SVs (according to IEC-61850 standard).

[0088] All MUs of the phases (A, B, C), including the reserve phase (R), will publish GOOSE messages and SVs with the digitized information, and the protection and / or control LEDs and / or CPRs and / or VPRs will sign the messages of the respective bank and of the reserve phase (R).

[0089] Under normal conditions, the IED and / or CPR and / or VPR will use the GOOSE and SV messages provided by the phases (A, B, C) of the bank. In the case of a replacement, the IED and / or CPR and / or VPR will block the message of the replaced phase (A, B, C) and consider the message of the reserve phase (R). The phase replacement command (A, B, C) may occur on the HMI of the IED and / or CPR and / or VPR and / or via the substation SCADA system and / or via the processing unit.

[0090] In terms of design, the virtualization that is the subject of the present invention provides a time saving in terms of schedule, since it will no longer be necessary to prepare the functional and constructive designs of the centralizing cabinet. Only the cabinet for the single-phase equipment will be designed, which will be typical for the other equipment.

[0091] Without a centralizing cabinet, there are field gains for assembling the power banks, as there will be no need to lay and commission control cables, only AC and DC power cables, which improves the substation's CAPEX.

[0092] Regarding operation, the substation maintenance technician will no longer need to operate from a panel in the yard or external area of ​​the substation, reducing the risk of accidents. In addition to safety, phase replacement will be faster, as programmed and tested logic will be activated, thus reducing the client's operating expenses.

[0093] The GOOSE and SV information, in the MUs of each phase including the reserve, is configured in SCL type files, to then be loaded into the LEDs and / or CPRs and / or VPRs.

[0094] These files are loaded into the computer-readable memory of the invention, which then has all the information about how the communication network between the system elements is formed, for example, how the MUs and LEDs and / or CPRs and / or VPRs communicate with each other and what information is being made available on the network (GOOSE messages and SVs).

[0095] By applying at least one MU to the cabinets of each piece of equipment, whether three-phase or single-phase with or without backup, the system of the invention applies to three possible scenarios.

[0096] In the case of three-phase equipment, all signals are transmitted from the equipment directly to the MU(s) via the wired or wireless communication and data network and / or Ethernet network, and no longer to the terminals, since the virtualization, the subject of the present invention, eliminates the centralizing cabinet. From the phase control cabinet, the signals go to the LEDs and / or CPRs and / or VPRs in the control room via the process bus and / or wired or wireless communication and data network and / or Ethernet network. It should be noted that, because the signals are digitized, fiber optic cables, wireless networks, etc. can be used.

[0097] In the case of single-phase equipment without a spare phase (R), all signals are transmitted from the equipment directly to the MU(s) via the wired or wireless communication and data network and / or Ethernet network, and no longer to the terminals, since virtualization, the subject of the present invention, eliminates the centralizing cabinet. From the cabinet, the signals go to the LEDs and / or CPRs and / or VPRs in the control room via the process bus and / or wired or wireless communication and data network. Wired and / or Ethernet network. It should be noted that, because the signals are digitized, fiber optic cables, wireless networks, etc. can be used;

[0098] In the case of single-phase equipment with a spare phase (R), all signals are transmitted from the equipment directly to the MU(s) via the wired or wireless communication and data network and / or Ethernet network, and no longer to the terminals, since virtualization, the subject of the present invention, eliminates the centralizing cabinet. From the cabinet, the signals go to the LEDs and / or CPRs and / or VPRs in the control room via the process bus and / or wired or wireless communication and data network and / or Ethernet network. It should be noted that, because the signals are digitized, fiber optic cables, wireless networks, etc., can be used. In this case, the spare phase will also have at least one MU.

[0099] Since the signals are digitized in the phase control cabinets of the respective phases (A, B, C) and reserve phase (R), it is necessary to create mechanisms so that the LEDs and / or CPRs and / or VPRs acquire the signals from the phases that make up a bank even when the reserve phase (R) replaces one of the phases (A, B, C) that has malfunctioned. This is achieved by virtualizing the centralizing cabinet of the respective bank, remembering that the virtualization of the centralizing cabinet corresponds to the elimination of the centralizing cabinet, improving the state of the art in a novel and inventive way. Method for training a neural network

[0100] The invention's method for training a convolutional neural network for monitoring and switching defective phases, wherein a neural network refers to a machine learning model, is a computer-implemented method capable of adjusting the parameters of the machine learning model to, from a number of reference parameters associated with ideal and / or expected values ​​for the phases (A, B, C) and reserve phase (R), be able to, from one or more input data from the LEDs and / or CPRs and / or VPRs and / or the processing unit, provide at least one output data or instruction and, in case of abnormal operating condition, command the replacement of a defective phase (A, B, C) with a reserve phase (R).

[0101] A trained convolutional neural network, according to the invention, can comprise interconnected groups of artificial neurons (e.g., neuron models), and can also be a computational device or be represented as a method that will be executed by a computational device.

[0102] A convolutional neural network trained according to the invention may be, for example, but without limiting the invention, a deep feedforward trained convolutional neural network, of such architecture that it delivers results quickly and accurately and that it can also be run, especially on portable processors such as, for example, but not limited to the present invention, processors of 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.

[0103] The preferred neural network, without limiting the scope of the invention, is a convolutional neural network, but it could be any other suitable similar architecture provided that the model is trained accordingly, just as the aforementioned convolutional neural network has been. Furthermore, the trained neural network of the invention may comprise a combination of two or more architectures.

[0104] The training serves to configure the neural network to provide diagnoses and instructions unambiguously associated with each phase (A, B, C) and reserve phase (R) of at least one bank of transformers and / or reactors of the substation, in response to receiving, as input, the measured values ​​of each component and / or the messages and / or input information received from the LEDs and / or CPRs and / or VPRs and / or processing unit.

[0105] Once the convolutional neural network has been trained and the system of the invention, as a whole, has been fed with the information and instructions described so far, the processing units and / or LEDs and / or CPRs and / or VPRs will be ready to monitor the phases (A, B, C) and the reserve phase (R) and, in case of abnormal operation, command the replacement of at least one defective phase (A, B, C), for example, the first phase (A), by at least one reserve phase (R).

[0106] For training the convolutional neural network of the invention, at least one database in the form of at least one file is used.

[0107] The training database is a historical database collected online by the inventor between 2021 and 2024 from 73 transformer equipment using its own monitoring system and offline between 2020 and 2024 from 1,302 transformer equipment using its own reports. The historical data on abnormal operation may include, for example, but not limited to, data such as transformer gas pressure and / or quantity, absence of current or voltage, current or voltage above or below the expected ideal, etc.

[0108] The training database thus generated comprises electrical and thermal data monitored by IEDs and centralized systems. Data in transformers. Anomalies are detected after analyzing at least one year of training data, based on parametric and non-parametric statistical metrics. This data is re-evaluated daily to refine the database and learning algorithm, thus enabling better accuracy in future diagnoses.

[0109] Therefore, the training database includes practical experience resulting from several years of monitoring transformer banks, comprising historical series of abnormal operation information such as, but not limited to, information collected and recorded as indicative and / or conclusive of abnormal operation and / or failure and / or defect of at least one phase (A, B, C) and / or reserve phase (R). The training database also includes the ideal and / or expected reference data for the phases (A, B, C) of each of the transformer banks and / or at least one file or database containing the expected physical quantities for the phases (A, B, C). It should also be noted that the database may also include, but not limited to, data from an SCL file of the substation and / or other additional files or databases.

[0110] This information is unequivocally assigned to at least one phase (A, B, C) and / or reserve phase (R), where the trained convolutional neural network is fed with this information that will assist in detecting at least one abnormal operating condition as input data and in the subsequent output instruction in the form of a warning and / or alarm and, mainly, for carrying out the replacement of at least one defective phase (A, B, C) by at least one reserve phase (R).

[0111] With this database, a deep convolutional feedforward neural network can be created, which can include layers of neurons that can be configured in a side-by-side receptive field, where the input data are read from the LEDs and / or CPRs and / or VPRs and / or processing unit, and the output data comprise at least one command for at least one phase selector switch to disconnect the faulty phase (A, B, C) and connect the spare phase (R).

[0112] A method for training a neural network for monitoring and switching defective phases, according to the invention, comprises the following method steps: (i) Provide at least one database in the form of at least one file with the ideal and / or expected historical and / or reference data for the phases (A, B, C) of each of the transformer banks and / or at least one file or database comprising the expected physical quantities for the phases (A, B, C); (ii) Configure a neural network to receive, as input, the input dataset, where each training input dataset corresponds to the respective historical and expected datasets, according to at least one database in the form of at least one file with the historical data and with the ideal and / or expected reference data for the phases (A, B, C) and reserve phase (R) of each of the transformer banks; (iii) Receive training input data; (iv) Configure the neural network to assign to each phase (A, B, C) and reserve phase (R) at least one expected value, according to at least one database in the form of at least one file with the ideal and / or expected reference data for the phases (A, B, C) and reserve phase (R) of each of the transformer banks and / or at least one file or database that includes the expected physical quantities for the phases (A, B, C) and reserve phase (R); (v) Configure the trained convolutional neural network so that, in response to an input representing a normal operating condition of phases (A, B, C) and reserve phase (R), it does not generate any blocking signal for any of the circuit breakers associated with phases (A, B, C) and reserve phase (R) and does not generate any alarm; (vi) Configure the trained convolutional neural network to, in response to an input representing an abnormal operating condition of the phases (A, B, C), generate a blocking signal to the IED and / or CPR and / or VPR to block the circuit breaker associated with the faulty bank, generate an alarm and, in response, as output data, generate an automatic instruction and / or wait for the substation operator to command the replacement of the faulty phase (A, B, C) with the spare phase (R), where this command can be given through the SCADA (via MMS message) or through the IED (buttons and / or keys and / or HMI) and / or CPR and / or VPR and / or processing unit; (vii) Configure the trained convolutional neural network to remove the internal blocking signal from the LEDs and / or CPRs and / or VPRs to unlock the operation of the circuit breakers; (viii) Configure the trained convolutional neural network to block the use of phase (R) by other parallel banks to the bank that is using the reserve phase (R); and (ix) Produce a neural network by repeatedly training the trained convolutional neural network with each of the training datasets for each of the phases (A, B, C) and the backup phase (R).

[0113] It should be noted that the command to replace the defective phase (A, B, C) with the spare phase (R) can also be performed by the substation operator, as a way of confirming and / or supervising the trained convolutional neural network. In this case, the output data from the trained convolutional neural network may be, in addition to the alarm, at least a suggested procedure to be confirmed by the operator. This confirmation can be done by the operator from an HMI and / or SCADA control panel, etc. Method for monitoring and switching defective phases

[0114] A method for monitoring and switching defective phases, according to the invention, is a computer-implemented method for monitoring the phases (A, B, C) and reserve (R) of single-phase transformers and / or single-phase reactors of the digital substation and replacing at least one defective phase (A, B, C) with at least one reserve phase (R) without manual human intervention, the method being executed by one or more components of the system of the invention.

[0115] The IED and / or CPR and / or VPR for protection and control of a single-phase equipment bank receives, from at least one MU of at least one of the phases (A, B, C) that make up at least one bank of transformers and / or reactors, the digitized information of the phases (A, B, C) of said bank (A, B, C) plus that of the reserve phase (R), including the states of the phase selector switches, if applied, which will be made available through multicast protocols, such as, for example, but without limiting the invention, GOOSE and SV messages, from MU type unit(s). Internally, the IED and / or CPR and / or VPR is configured to execute a logic for selecting the messages it will use in its functions, such that, under normal operating conditions, even though it is receiving information from all four phases (A, B, C, R), only the information from phases (A, B, C) will be used, and the signals from the reserve phase (R) are blocked internally in the IED and / or CPR and / or VPR.It should be noted, once again, that a normal operating condition is preferably, but without limiting the invention, a condition in which there is no exchange of fault and / or defect messages between at least one MU of the respective bank and at least one IED and / or CPR and / or VPR.

[0116] The input data received from the MUs is submitted to the processing unit and thus to the trained neural network of the invention, and in the case of normal operation, no alarm signal and / or suggestion and / or switching instruction will be generated.

[0117] In case of abnormal operation, when at least one MU of at least one transformer and / or reactor bank sends an abnormal operation signal for at least one of the phases (A, B, C), that is, when there is a need to replace one of the phases (A, B, C), such as, for example, the defective phase (A) of at least one first bank, here Referred to as Bank 1, the sequence of maneuvers, which correspond to the method steps of the invention, comprises the following steps.

[0118] The neural network sends a command to the IED and / or CPR and / or VPR of bank 1 of the defective phase (A), through the processing unit, to replace the defective phase (A) of bank 1, for example, with the spare phase (R). It should be noted that the trained convolutional neural network can also send an alarm and procedure suggestion to the substation operator, where the substation operator, either through the SCADA (via MMS message) or through the IED (buttons and / or keys and / or HMI) and / or CPR and / or VPR, commands the replacement of the defective phase (A) of bank 1, for example, with the spare phase (R).

[0119] The processing unit and / or IED and / or CPR and / or VPR checks if the corresponding bank 1 is isolated, i.e., if the circuit breakers are open and bank 1 is de-energized and switched off. At this point, the circuit breakers are interlocked in the IED and / or CPR and / or VPR so that they are not operated until the phase replacement (A) is completed. While the replacement is not completed, an internal signal is generated to the IED and / or CPR and / or VPR to block the operation of the circuit breakers.

[0120] The processing unit and / or IED and / or CPR and / or VPR sends commands via protocol, GOOSE for example, to the MUs installed in the yard to execute the opening of the switches of each winding of the equipment that connects to the main busbar of the substation and closing of the phase selector switches to connect the spare phase (R) to the main busbar of the substation. If it is a transformer bank, this switching will occur in each winding (primary, secondary, tertiary, etc.) and the neutral(s). If it is a reactor bank, it will occur in the primary winding and the neutral.

[0121] With the completion of the maneuvers of all phase selector switches, the IED and / or CPR and / or VPR blocks the information from the faulty phase (A), originating from the MU(s) of this equipment, in its logic and starts using the information published by the MU(s) of the reserve phase (R). It should be noted, once again, that while the replacement is not finalized, an internal signal is generated in the IED and / or CPR and / or VPR to block the operation of the circuit breakers (interlocking).

[0122] The circuit breaker interlock will be deactivated by removing the blocking signal, releasing the command for the operator to energize bank 1 with the reserve phase (R).

[0123] While the reserve phase (R) is in use by some bank, as in the example bank 1, the other banks parallel to it will be blocked from using the switched reserve phase (R) as a replacement for the defective phase (A). This is because the IED and / or CPR and / or VPR will generate a blocking signal and distribute it to the other IEDs and / or CPRs and / or VPRs of the other banks.

[0124] At any point in the sequence of steps described above, if the processing unit and / or IED and / or CPR and / or VPR has not received feedback, via MU(s), that the maneuver was successful, such as if the processing unit does not receive confirmation of the closing of a phase selector switch, the system will trigger an alarm and the circuit breakers will be locked. It should be noted that, if a switching operation is not completed due to a mechanical failure of the phase selector switch, manual intervention will be necessary to unlock and / or complete the switching operation.

[0125] It should also be noted that, although the example above referred to the LEDs of the equipment, this solution is also valid for the use of CPRs and / or VPRs.

[0126] Therefore, a method for monitoring and switching defective phases, according to the invention, comprises the following method steps: I. Obtain, by means of at least one IED and / or CPR and / or VPR, from at least one MU of at least one of the phases (A, B, C) of a transformer bank, at least one input information from at least one of the phases (A, B, C); II. Apply a machine learning model, preferably, but not limited to, a neural network model, preferably a trained convolutional neural network, to the input information in a processing unit; III. Obtain, in the processing unit, as a result of applying the machine learning model, at least one output piece of information equivalent to an operating condition of at least one of the phases (A, B, C); and IV. In case of normal operating conditions, do not generate any alarm signal and / or switching suggestion and / or instruction; or V. In case of an abnormal operating condition, generate, through the processing unit and / or IED and / or CPR and / or VPR, an alarm signal and an instruction in the form of a command to replace at least one defective phase (A, B, C) with at least one spare phase (R) from the bank; VI. Verify, using the processing unit and / or IED and / or CPR and / or VPR, if the corresponding bank is isolated, if the circuit breakers are open, and if the bank is de-energized; VII. Interlock the circuit breakers, through the processing unit and / or IED and / or CPR and / or VPR, generating an internal signal to the IED and / or CPR and / or VPR to block the operation of the circuit breakers; VIII. Send, through the processing unit and / or IED and / or CPR and / or VPR, commands via multicast protocol to at least one MU to execute the opening of the switches of each winding of the equipment that connects to the main busbar of the substation and closing of the phase selector switches to connect the spare phase (R) to the main busbar of the substation; IX. Block, through the processing unit and / or IED and / or CPR and / or VPR, the information from the defective phase (A, B, C), originating from at least one MU, in its logic and use the information published by at least one MU from the backup phase (R); X. Receive, through the processing unit and / or IED and / or CPR and / or VPR, from at least one MU, a signal indicating the completion of the switching maneuver; and XI. Deactivate, through the processing unit and / or IED and / or CPR and / or VPR, the interlocking of the circuit breakers by removing the blocking signal, releasing a command to energize the corresponding bank with at least one spare phase (R).

[0127] The logic behind the monitoring and switching of faulty phases (A, B, C) using at least one spare phase (R), illustrated in Figure 5, represents a method for selecting the signal (which can be digital, GOOSE, for example, or analog, SV, for example) between the phases that make up a bank (A, B, C) and the spare phase (R). The results of the logic, the outputs, are shown on the right side of the figure. The first three outputs are the result of the logic selection. Each of the three results comes from an "OR" gate that receives the signal from two "AND" gates. These "AND" gates provide the released signal to be used at the output, either from the phase itself (A, B, C) or from the spare phase (R). The blocking and release of the phase (A, B, C) to be used or not is determined by the state of the "LATCH" ("SET-S / RESET-R"). When the IED and / or CPR and / or VPR receives the phase replacement command, a signal release logic is performed, the "AND" gate before the "LATCH".To release this signal, no other phase (A, B, C) of the bank can be using the reserve phase (R), and it also receives a signal indicating that no other parallel bank is using the reserve phase (R). Therefore, if these conditions are met, the logic will act on the "SET" of the "LATCH," and the result is that the signal from the reserve phase (R) will be used. When the return signal acts on the "RESET" of the "LATCH," the system blocks the signal from the reserve phase (R) and uses the signal from phase (A, B, C). Note that when the bank uses the reserve phase (R), the last signal on the right side is generated, which serves to block the substitution in the other banks, if they exist. This logic can be implemented using other types of logic gates and connections, but the reasoning will remain the same, and the outputs will be the same, virtualizing the centralizing cabinet, the object of this invention. Final considerations

[0128] 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. Such modifications should be considered as falling within the scope of the present invention. Consequently, the particular embodiments described in detail above are merely 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. Virtualization system for the centralizing cabinet in single-phase transformer and / or reactor banks of digital substations, characterized by comprising at least one MU installed in the control cabinet of each phase (A, B, C) of at least one single-phase transformer and / or reactor bank.

2. System, according to claim 1, characterized in that it further comprises at least one protection IED and / or CPR and / or VPR for each bank of single-phase transformers and / or single-phase reactors, at least one control IED and / or CPR and / or VPR for each bank of single-phase transformers and / or single-phase reactors, at least one phase selector switch for each phase (A, B, C) and spare phase (R) of at least one bank of single-phase transformers and / or single-phase reactors for selecting the busbar to which each phase (A, B, C) and spare phase (R) will be connected.

3. System, according to any one of claims 1 to 2, characterized in that it further comprises at least one processor or processing unit, wherein the processing unit comprises at least one memory that stores computer-executable information and instructions and at least one trained convolutional neural network, wherein the processing unit and / or IED and / or CPR and / or VPR and the trained convolutional neural network are configured to provide output data and / or output instructions on one or more operating conditions of at least one of the phases (A, B, C), in response to the reception of one or more input messages and / or information from each of the phases (A, B, C) and the backup phase (R), obtained from at least one MU installed in the phase control cabinet of each of the phases (A, B, C). B, C) and the reserve phase (R) of the bank and transmitted to one or more LEDs and / or CPRs and / or VPRs, where the execution of the instructions performs one or more steps of a monitoring and switching method of the phases (A, B, C, R) of the single-phase transformer banks and / or the single-phase reactors of the digital substation.

4. System according to claim 1, characterized in that it further comprises: - At least one human-machine interface (HMI) and / or processing equipment for acquiring and presenting information / instructions; - At least one database; - At least one communication and data network; - One or more sets of input information from the equipment; and - One or more sets of data and / or output instructions from the LEDs and / or CPRs and / or VPRs.

5. System, according to claim 1, characterized in that with the installation of the MUs in the control cabinets of each phase (A, B, C), including the reserve phase (R), all signals are digitized and made available on the substation process bus or network via multicast protocols such as GOOSE messages and SVs.

6. System, according to any one of claims 1 to 5, characterized in that all MUs of phases (A, B, C), including the reserve phase (R), will publish GOOSE messages and SVs with the digitized information and the protection and / or control LEDs and / or CPRs and / or VPRs will sign the messages of the respective bank and of the reserve phase (R).

7. System, according to any one of claims 1 to 6, characterized in that the IED and / or CPR and / or VPR, under normal operating conditions, will use the GOOSE and SV messages made available by the phases (A, B, C) of the bank and, in the event of a replacement of at least one defective phase (A, B, C) by at least one reserve phase (R), the IED and / or CPR and / or VPR will block the message of the replaced phase (A, B, C) and will consider the message of the reserve phase (R).

8. Method for training a neural network for monitoring and switching defective phases, implemented on a computer, characterized by comprising the following method steps: (i) Provide at least one database in the form of at least one file with the ideal and / or expected historical and / or reference data for the phases (A, B, C) of each of the transformer banks and / or at least one file or database comprising the expected physical quantities for the phases (A, B, C); (ii) Configure a neural network to receive, as input, the input dataset, where each training input dataset corresponds to the respective historical and expected datasets, according to at least one database in the form of at least one file with the historical data and with the ideal and / or expected reference data for the phases (A, B, C) and reserve phase (R) of each of the transformer banks; (iii) Receive training input data; (iv) Configure the neural network to assign to each phase (A, B, C) and reserve phase (R) at least one expected value, according to at least one database in the form of at least one file with the ideal and / or expected reference data for the phases (A, B, C) and reserve phase (R) of each of the transformer banks and / or at least one file or database that includes the expected physical quantities for the phases (A, B, C) and reserve phase (R); (v) Configure the trained convolutional neural network so that, in response to an input representing a normal operating condition of phases (A, B, C) and reserve phase (R), it does not generate any blocking signal for any of the circuit breakers associated with phases (A, B, C) and reserve phase (R) and does not generate any alarm; (vi) Configure the trained convolutional neural network to, in response to an input data representing an abnormal operating condition of the phases (A, B, C), generate a blocking signal to the IED and / or CPR and / or VPR to block the circuit breaker associated with the faulty bank, generate an alarm and, in response, as output data, generate an instruction and / or wait for the substation operator to command the replacement of at least one faulty phase (A, B, C) with at least one spare phase (R), where this command can be given through the SCADA (via MMS message) or through the IED (buttons and / or keys and / or HMI) and / or CPR and / or VPR and / or processing unit; (vii) Configure the trained convolutional neural network to remove the internal blocking signal from the LEDs and / or CPRs and / or VPRs to unlock the operation of the circuit breakers; (viii) Configure the trained convolutional neural network to block the use of phase (R) by other parallel banks to the bank that is using the reserve phase (R); and (ix) Produce a neural network by repeatedly training the trained convolutional neural network with each of the training datasets for each of the phases (A, B, C) and the backup phase (R).

9. Method, according to claim 8, characterized in that a normal operating condition indicates that the transformer and / or reactor banks are operating normally with their respective phases (A, B, C) and that the reserve phase (R) is not connected to the power system.

10. A method, according to any one of claims 8 or 9, characterized in that a normal operating condition is a condition in which there is no exchange of fault and / or defect messages between at least one MU of the respective bank and at least one IED and / or CPR and / or VPR.

11. Method, according to claim 8, characterized in that an abnormal operating condition indicates that at least one of the phases (A, B, C) of the transformer and / or reactor bank is damaged and / or inoperative and / or operating with a defect and / or has suffered an accident and that it will not be able to return to operation in a short time, requiring its replacement by at least one spare phase (R).

12. A computer-implemented method for monitoring and switching defective phases, characterized by comprising the following method steps: I. Obtain, by means of at least one IED and / or CPR and / or VPR, from at least one MU of at least one of the phases (A, B, C) of a transformer bank, at least one input information from at least one of the phases (A, B, C); II. Apply a machine learning model, preferably, but not limited to, a neural network model, preferably a trained convolutional neural network, to the input information in a processing unit; III. Obtain, in the processing unit, as a result of applying the machine learning model, at least one output piece of information equivalent to an operating condition of at least one of the phases (A, B, C); and IV. In case of normal operating conditions, do not generate any alarm signal and / or switching suggestion and / or instruction; or V. In case of an abnormal operating condition, generate, through the processing unit and / or IED and / or CPR and / or VPR, an alarm signal and an instruction in the form of a command to replace at least one defective phase (A, B, C) with at least one spare phase (R) from the bank; VI. Verify, using the processing unit and / or IED and / or CPR and / or VPR, if the corresponding bank is isolated, if the circuit breakers are open, and if the bank is de-energized; VII. Interlock the circuit breakers, through the processing unit and / or IED and / or CPR and / or VPR, generating an internal signal to the IED and / or CPR and / or VPR to block the operation of the circuit breakers; VIII. Send, through the processing unit and / or IED and / or CPR and / or VPR, commands via multicast protocol to at least one MU to execute the opening of the switches of each winding of the equipment that connects to the main busbar of the substation and closing of the phase selector switches to connect the spare phase (R) to the main busbar of the substation; IX. Block, through the processing unit and / or IED and / or CPR and / or VPR, the information from the defective phase (A, B, C), originating from at least one MU, in its logic and use the information published by at least one MU from the backup phase (R); X. Receive, through the processing unit and / or IED and / or CPR and / or VPR, from at least one MU, a signal indicating the completion of the switching maneuver; and XI. Deactivate, through the processing unit and / or IED and / or CPR and / or VPR, the interlocking of the circuit breakers by removing the blocking signal, releasing a command to energize the corresponding bank with at least one spare phase (R).

13. Virtualization system for the centralizing cabinet in single-phase transformer and / or reactor banks of digital substations, characterized by the fact that it performs at least one, preferably all the steps of the method for monitoring and switching defective phases defined in claim 12.

14. Computer-readable memory, characterized in that it comprises a set of instructions which, when executed, perform the method of monitoring and switching defective phases defined in claim 12.

Images