WO2020111561A1 - 머신러닝 기반 태양광 발전 제어 시스템 및 방법 - Google Patents
머신러닝 기반 태양광 발전 제어 시스템 및 방법 Download PDFInfo
- Publication number
- WO2020111561A1 WO2020111561A1 PCT/KR2019/015018 KR2019015018W WO2020111561A1 WO 2020111561 A1 WO2020111561 A1 WO 2020111561A1 KR 2019015018 W KR2019015018 W KR 2019015018W WO 2020111561 A1 WO2020111561 A1 WO 2020111561A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- control
- unit
- module
- real
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/165—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values
- G01R19/16533—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values characterised by the application
- G01R19/16538—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values characterised by the application in AC or DC supplies
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/165—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values
- G01R19/16533—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values characterised by the application
- G01R19/16538—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values characterised by the application in AC or DC supplies
- G01R19/16547—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values characterised by the application in AC or DC supplies voltage or current in AC supplies
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R21/00—Arrangements for measuring electric power or power factor
- G01R21/06—Arrangements for measuring electric power or power factor by measuring current and voltage
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G08—SIGNALLING
- G08C—TRANSMISSION SYSTEMS FOR MEASURED VALUES, CONTROL OR SIMILAR SIGNALS
- G08C17/00—Arrangements for transmitting signals characterised by the use of a wireless electrical link
- G08C17/02—Arrangements for transmitting signals characterised by the use of a wireless electrical link using a radio link
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
- H02H3/38—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to both voltage and current; responsive to phase angle between voltage and current
- H02H3/385—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to both voltage and current; responsive to phase angle between voltage and current using at least one homopolar quantity
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
- H02H7/20—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for electronic equipment
- H02H7/205—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for electronic equipment for controlled semi-conductors which are not included in a specific circuit arrangement
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S40/00—Components or accessories in combination with PV modules, not provided for in groups H02S10/00 - H02S30/00
- H02S40/30—Electrical components
- H02S40/36—Electrical components characterised by special electrical interconnection means between two or more PV modules, e.g. electrical module-to-module connection
-
- G—PHYSICS
- G08—SIGNALLING
- G08C—TRANSMISSION SYSTEMS FOR MEASURED VALUES, CONTROL OR SIMILAR SIGNALS
- G08C2201/00—Transmission systems of control signals via wireless link
- G08C2201/40—Remote control systems using repeaters, converters, gateways
- G08C2201/42—Transmitting or receiving remote control signals via a network
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0061—Details of emergency protective circuit arrangements concerning transmission of signals
- H02H1/0084—Details of emergency protective circuit arrangements concerning transmission of signals by means of pilot wires or a telephone network; watching of these wires
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
- H02H3/24—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to undervoltage or no-voltage
- H02H3/243—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to undervoltage or no-voltage for DC systems
Definitions
- the present invention relates to a machine learning-based photovoltaic power generation control system and method, and more specifically, to collect voltage and power information of photovoltaic modules in real time, learn the collected information in a machine learning platform, and vary according to the learning results.
- Machine learning-based photovoltaic power generation that can control the photovoltaic modules based on the extracted control information by performing modeling for the service function, and analyze the devices connected to the photovoltaic module in real time to control the photovoltaic system in real time. It relates to a control system and method.
- Photovoltaic power generation is a power generation method that converts sunlight into direct current electricity to produce electric current.
- a solar panel with several solar cells is spread on a large scale to produce electricity using solar energy.
- Smart Grid Smart Grid
- the current maintenance management system of solar power only measures the inverter output of the entire system, and only performs on-site monitoring by a webcam. It is taking time. In addition, it is difficult to grasp the failure type, the failure pattern, and the change in power generation due to environmental changes due to the lack of data securing of solar power.
- Figure 1 is a schematic diagram showing a conventional general photovoltaic power condition diagnosis system, the conventional photovoltaic power generation technology when a failure occurs in any one of a plurality of photovoltaic cell modules, a failure or an abnormal state occurs in any module It is not easy to determine whether it has been done, and it is impossible to remotely determine whether the power generation capacity is low due to a small amount of solar radiation or the power generation capacity has fallen due to a module failure or an abnormal condition.
- the existing photovoltaic power generation system is affected by the characteristics of numerous photovoltaic modules and devices connected thereto, but since it is difficult to control all of these individual characteristics, a photovoltaic power generation control system using a machine learning technique is required. .
- the present invention has been devised to solve the conventional problems as described above, the purpose of which is to collect voltage and power information of photovoltaic modules in real time, learn the collected information in a machine learning platform, and provide various services according to the learning results.
- Machine learning-based photovoltaic power generation control capable of real-time control of the photovoltaic power generation system by real-time analysis of devices connected to the photovoltaic power generation module and control of the photovoltaic power generation module based on the extracted control information It is intended to provide a system and method.
- Machine learning-based solar power control system a plurality of solar modules for photovoltaic power generation are connected in series, the solar modules are arranged in at least two rows and columns; Node controllers for measuring voltage, current, and power data produced from the photovoltaic modules, respectively, and blocking switching of the connected photovoltaic module when the measured data is less than the set control current, voltage, and power control data; A gateway unit that receives voltage, current, and power measurement data measured from solar modules from the node controllers and parses the measurement data to store measurement data; A real-time control module that receives parsed data from the gateway, processes classification, collation, and analysis, stores it in a database, and transmits control commands to control the node controls to set control data to the gateway; And the real-time control module connected to the Internet or a communication network to monitor photovoltaic components and data including a solar module, learn data transmitted from the real-time control module by machine learning basis, and are required for solar power control. It is characterized by including a machine learning server that performs
- the node control unit includes: a measurement unit for measuring voltage, current, and power data produced by the solar module; Analysis control that compares the measured data and control data defined to control the current and voltage transmitted from the real-time control module to determine whether the measured data falls short of the defined control data and switches and controls the solar module operation. part; A data transmission management unit that transmits data stored in the node control unit to the outside and receives control data from the real-time control module and transmits the control data to the node control unit; And a switching unit for blocking or connecting a solar module connected to the node control unit according to the result determined by the analysis control unit.
- the gateway unit transmits measurement data received from the node control unit to the real-time control module and receives control command data transmitted from the real-time control module at any time to the node.
- a data collection management unit transmitting to the control unit;
- a data processor parsing the measurement data received from the node controller;
- a data storage unit that stores data parsed by the data processing unit.
- the real-time control module includes: an interface unit for real-time data transmission and reception between a gateway unit and a real-time control module and between a real-time control module and an integrated information server; A data analysis unit receiving the current, voltage, and power data measured by the solar module through the gateway unit to process classification, comparison, and analysis; A control determination unit that extracts reference data necessary for controlling the solar module using the processing result of the data analysis unit and determines a control command; A control command processing unit that defines control data to control the solar module and transmits it to the gateway in real time; And a service management unit that performs a management function to add or delete control functions for controlling the solar module inside the real-time control module.
- the machine learning server an interface unit for transmitting and receiving data in real time in the machine learning server;
- a monitoring unit which analyzes, analyzes, classifies, contrasts, and converts photovoltaic data transmitted from a photovoltaic power generation device including a photovoltaic module, a node control unit, a gateway unit, and a real-time control module according to characteristics;
- a determination unit to determine whether to process machine learning according to the profile information of the integrated processed data transmitted from the monitoring unit;
- a learning module that performs new learning on data requiring new learning according to the determination unit;
- a modeling unit that extracts data and performs modeling using the learned results;
- a learning database that stores the learning result and modeling performance result;
- a control unit that fetches modeling data for photovoltaic power generation control from the learning database and transmits to the service unit for photovoltaic power generation control along with data not determined by the determination unit;
- a service unit that selects among reference models transmitted from the control unit to define solar control service data
- the monitoring unit processes data transmission and reception with a real-time control module through a data interface roll
- the service unit is connected to an open interface unit to apply data received from the control unit It is characterized by providing data converted to a service or requesting data from various application programs.
- the machine learning-based photovoltaic power generation control method the node control step of measuring the voltage, current, power data produced from the solar modules, respectively;
- the gateway unit receives the measurement data from the node control unit, parsing and storing the data;
- the real-time control module processes parsed, classified, and analyzed parsed data in a database, and transmits it to a machine learning server;
- the machine learning server learns the data transmitted from the real-time control module based on the machine learning basis, extracts functional data necessary for solar power generation control, performs modeling, and provides control service data according to the modeling results to the real-time control module;
- the real-time control module transmits control command data for controlling the node controllers to the gateway unit;
- the gateway unit transmits the control command data received from the real-time control module to the node control unit;
- the node control unit is characterized in that it comprises a step of switching off the connected solar module when the data measured in the solar module is less than the current, voltage, and power control reference data set according
- the present invention by collecting voltage and power information in real time, learning the collected information on a machine learning platform, modeling various service functions according to the learning results, and controlling the solar modules by controlling the photovoltaic modules based on the extracted control information. It is effective to maintain the optimum power production by uniformly controlling the amount of power produced by the optical module.
- each measurement data connected to a solar module is transmitted to a machining learning platform in real time, thereby transmitting faulty device recognition information and type information to a machine learning server by a learning function, thereby enabling wide-area solar power generation equipment. It has the effect of analyzing the failure situation and the characteristics of the failure type in real time to service the user.
- FIG. 1 is a schematic diagram showing a conventional general photovoltaic state diagnosis system.
- FIG. 2 is a block diagram of an overall configuration of a machine learning-based solar power control system according to the present invention.
- FIG. 3 is a block diagram of a configuration of a node controller in the machine learning-based solar power control system shown in FIG. 2.
- FIG. 4 is a block diagram of a gateway unit in the machine learning-based photovoltaic power generation control system shown in FIG. 2.
- FIG. 5 is a block diagram of a real-time control module in the machine learning-based photovoltaic power generation control system shown in FIG. 2.
- FIG. 6 is a block diagram of a machine learning server in the machine learning-based solar power control system shown in FIG. 2.
- FIG. 7 is another embodiment of a block diagram of a machine learning server in the machine learning-based solar power control system shown in FIG. 2.
- FIG. 8 is a flowchart of a method for controlling solar power generation based on machine learning according to the present invention.
- FIG. 2 is a block diagram of an overall configuration of a machine learning-based solar power control system according to the present invention.
- the machine learning-based photovoltaic power generation control system includes a plurality of photovoltaic modules 10, a node control unit (NBC, 30), an inverter 50, a gateway unit (G/W) , 100), a real-time control module (RTCU, 200), a machine learning server 300.
- the photovoltaic modules 10 are arranged in at least two or more rows and columns by connecting a plurality of photovoltaic modules that convert photovoltaic energy into electrical energy and perform photovoltaic power generation in series.
- the photovoltaic device is provided with one or more strings in which the photovoltaic modules 10, which are the smallest units for actually transmitting the generated electricity, are connected in series, and the strings are gathered again to form an array, in the case of large-scale photovoltaic power generation. At least dozens and as many as hundreds or more arrays are installed and operated.
- the power produced from the plurality of solar modules is connected to the inverter 50 in 6 or 12 channels and converted into AC power such as 220V.
- the node control unit 30 is connected to each of the photovoltaic modules 10, measures voltage, current, and power data produced from the photovoltaic modules 10, respectively, and measures measured current, voltage, and power. If the control data is not reached, the connected solar modules are switched off.
- the solar modules 10, A to I are connected in series, and the node controllers 30 are connected in series to the solar modules 10 connected in series with the same structure, respectively.
- the gateway unit 100 receives voltage, current, and power measurement data measured from the solar modules 10 from the node balancing control units 30 and parses the measurement data to store measurement data.
- the real-time control module 200 receives current, voltage, and power data measured from the photovoltaic modules 10 through the gateway 100, processes classification, contrast, and analysis, and stores them in a database, while controlling the node.
- a control command for setting control data for controlling the 30s is transmitted to the gateway unit 100.
- the machine learning server 300 is connected to the real-time control module 200 and the Internet or a communication network to monitor the photovoltaic power generation component and data including the solar module, and to receive data transmitted from the real-time control module 200. Learning based on machine learning, extracting functional data necessary for solar power generation control, and performing modeling, and provides control service data according to the modeling results to the real-time control module 200.
- the function of the machine learning server may be processed by the real-time control module 200.
- FIG. 3 is a block diagram of a configuration of a node controller in the machine learning-based solar power control system shown in FIG. 2.
- the node control unit 30 includes a measurement unit 31, an analysis control unit 32, a data transmission management unit 33, and a switching unit 34 It includes.
- the measurement unit 31 measures voltage, current, and power data produced by the solar module 10.
- the measurement unit 31 measures the voltage, current, and power produced by the photovoltaic module 10 composed of each power production panel and transmits it to the analysis control unit 32.
- the analysis control unit 32 is the measured current, voltage, power data of each photovoltaic module secured from the measurement unit 31 and the current and voltage control data values set and transmitted by the real-time control module 200 According to the operation level of the switching unit 34, the switching unit 33 operates the blocking switching of the solar module 10 connected to the node control unit 30 in the switching unit 33.
- the analysis control unit 32 controls the defined data by comparing the measured data with control data that is transmitted from the real-time control module 200 and is defined to control the current and voltage of the solar module. The operation of the solar module is switched to determine whether the data is insufficient.
- the data transmission management unit 33 transmits data stored in the node control unit 30 to the outside, receives control data from the real-time control module 200 through the gateway unit 100, and transmits the data to the node control unit 30. .
- the switching unit 34 blocks or connects the solar module connected to the node control unit 30 according to the result determined by the analysis control unit 32.
- FIG. 4 is a block diagram of a gateway unit in the machine learning-based photovoltaic power generation control system shown in FIG. 2.
- the gateway unit 100 includes a data collection management unit 110, a data processing unit 120, and a data storage unit 130.
- the data collection management unit 110 transmits the measurement data received from the node control unit 30 to the real-time control module 200 and receives the control command data transmitted from the real-time control module 200 at any time, so that the node control unit 30 Transfer to.
- the information transmitted to the data collection management unit 110 is information transmitted from the node control unit 30.
- the correct data is obtained through connection or disconnection of the solar module from the node control units 30 connected to 10X16 solar modules per gateway unit. Special management function is built in to deliver the data safely.
- the data processing unit 120 parses the measurement data received from the node control unit 30.
- the data is parsed (encoded/decoded), and the data processed through the parsing process is transferred to the data storage unit 130 for storage.
- the data storage management unit 130 stores the data in a file data system structure or a database structure for easy use of the stored information, and the stored data is transmitted to the real-time control module 200 through the data collection management unit 110. .
- FIG. 5 is a block diagram of a real-time control module in the machine learning-based photovoltaic power generation control system shown in FIG. 2.
- the real-time control module 200 includes an interface unit 210, a data analysis unit 220, a control determination unit 230, and a control command processing unit 240, a service management unit 250.
- the interface unit 210 is for transmitting and receiving data in real time between the gateway unit 100 and the real-time control module 200 and between the real-time control module 200 and the machine learning server 300.
- the node control unit 30 transmits control command data through the interface unit 210 through analysis, determination, and determination, and the node control unit 30 includes current, voltage, and power production values of each solar module. Make it possible to function as a short circuit or connection.
- the data analysis unit 220 receives current, voltage, and power data measured by the solar module 10 through the gateway unit 100 to process classification, comparison, and analysis.
- the data analysis unit 220 receives the current, voltage, and power data measured for each photovoltaic module as solar power generation information through the interface unit 210 and classifies, contrasts, and analyzes them as a pre-processing process to determine control To the department 230.
- the control determination unit 230 extracts reference data necessary for controlling the solar module using the processing result of the data analysis unit 220 to determine a control command.
- the control command processing unit 240 defines control data to control the solar module 10 and transmits it to the gateway unit 100 in real time.
- the service management unit 250 performs a management function to add or delete control functions for controlling the solar module 10 inside the real-time control module 200.
- the function of the machine learning server may be processed by the real-time control module 200.
- the configuration of the machine learning server described later without separately placing the machine learning server is in the real-time control module.
- the design can be changed to include it.
- FIG. 6 is a block diagram of a machine learning server in the machine learning-based solar power control system shown in FIG. 2.
- the machine learning server 300 includes an interface unit 310, a monitoring unit 320, a determination unit 330, and a learning module 340. , A modeling unit 350, a learning database 360, a control unit 370, a service management module 380, and a service unit 390.
- the interface unit 310 is for transmitting and receiving data in real time in a machine learning server.
- the machine learning server 300 is connected to the real-time control module 200 and the Internet or a communication network, and the interface unit 310 processes data transmission and reception in real time between the real-time control module 200 and the machine learning server 300.
- the data processed by the machine learning server 300 may provide various services or data through the interface unit 310.
- the monitoring unit 320 analyzes the photovoltaic data transmitted from the photovoltaic component including the solar module 10, the node control unit 30, the gateway unit 100, and the real-time control module 200, Classification, contrast, and transformation are monitored by integrated processing according to characteristics.
- the monitoring unit 320 receives the equipment and measurement data constituting the photovoltaic power generation, analyzes and processes the profile information of components such as power generation functions, measurement values, changes, features, and transmits the processed information to the determination unit 330.
- the data monitored by the analysis, classification, collation, and conversion processing by the monitoring unit 320 are data for machine learning learning, for example, connection device protocol information, connection device failure type information, module-specific voltage, current, It may include power data, module failure type information, weather information, solar altitude information, power amount for each module angle, user power information, KEPCO supply power information, and the like.
- the determination unit 330 determines whether machine learning is processed according to the profile information of the integrated processed data transmitted from the monitoring unit 320.
- the determination unit 330 determines whether the data transmitted from the monitoring unit 320 is necessary for machine learning processing, and transmits the new learning necessary to the learning module 340, and controls the data determined that no new learning is necessary. (370).
- the learning module 340 performs new learning on data requiring new learning according to the determination unit 330, and the learning results are stored in the learning database 360.
- the modeling unit 350 extracts data using the learned result and performs modeling, and the modeling result is stored in the learning database 360.
- connection device recognition modeling API modeling
- device-specific protocol modeling connection device failure modeling
- power production prediction modeling module failure (disability)
- module failure (disability) Modeling
- weather state/time power generation modeling module angle power generation modeling
- user power usage modeling and supply power modeling.
- the controller 370 fetches modeling data for solar power generation control from the learning database 360 and transmits the data to the service unit 390 for solar power generation control along with data not determined by the determination unit.
- the service unit 390 selects among reference models transmitted from the control unit 370 to define solar power control service data and transmits it to the real-time control module 200 through the interface 310.
- the control services processed in the service part include photovoltaic power generation balancing control using a node control unit, multi-device protocol connection support function, device-specific failure type diagnosis function, voltage, current, and power amount monitoring module failure type function for each solar module, device Star PV module Weather power generation and production forecast function, information service, solar altitude power production forecast and data service, solar module up and down and left and right angle production data service, user power consumption information provision function, KEPCO supply power And information providing functions.
- FIG. 6 a plurality of modeling lists that are modeled and delivered by the modeling unit are displayed on the service unit 390 by a plurality of red circles.
- the service management module 380 is for managing by deleting, adding, and updating the reference modeling list transmitted to the service unit 390.
- FIG. 7 is another embodiment of a configuration block diagram of a machine learning server in a machine learning based solar power control system.
- the interface unit 310 is divided into a data interface 311 and an open interface unit (open API, 312), as compared to FIG. 6.
- the monitoring unit 320 processes data transmission and reception with the real-time control module through the data interface 311 roll.
- the service unit 390 converts the modeling data received from the control unit 370 and the data not determined by the determination unit into application functions, and responds to various requirements or requests information from various application programs in an open interface. It can be provided through the unit 312.
- FIG. 8 is a flowchart of a method for controlling solar power generation based on machine learning according to the present invention.
- the machine learning-based solar power control method according to the present invention proceeds as follows.
- the node control unit 30 measures voltage, current, and power data produced from the solar modules 10, respectively (S10).
- the gateway unit 100 receives the measurement data from the node control unit 30, parses it and stores it (S20).
- the real-time control module 200 which receives the information transmitted to the gateway unit, processes the parsed data by sorting, matching, and analyzing it, stores the parsed data in a database, and transmits the parsed data to the machine learning server 300 (S30).
- the machine learning server 300 learns the data transmitted from the real-time control module 200 based on machine learning, extracts functional data necessary for solar power generation control, performs modeling, and controls the control service data according to the modeling results in real time.
- a module 200 S40.
- the real-time control module 200 transmits control command data for controlling the node control units 30 to the gateway unit (S50).
- the gateway unit 100 transmits the control command data received from the real-time control module 200 to the node control unit 30 (S60).
- the node controller 30 cuts off and controls the connected solar module (S70). .
- the node control unit 30 is configured to display the current measured voltage, power, and power data if the current measured values of the current, voltage, and power data are not set as reference data according to a control command.
- the present invention collects the amount of power generated by the photovoltaic module in real time, learns the collected information from the machine learning platform, performs modeling for various service functions according to the learning results, and generates the amount of power generated by the photovoltaic module by the extracted control information.
- By uniformly controlling the power it is possible to maintain the optimal power production, and by transmitting device recognition information and type information to the machine learning server, it analyzes the device failure situation and type characteristics of wide-area photovoltaic power generation in real time and serves the user. can do.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Automation & Control Theory (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Photovoltaic Devices (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
- Control Of Electrical Variables (AREA)
Abstract
Description
Claims (7)
- 태양광 발전을 하는 복수 개의 태양광 모듈이 직렬로 연결되어 적어도 2 이상의 행과 열로 배열되는 태양광 모듈들;상기 태양광 모듈들로부터 생산되는 전압, 전류, 전력 데이터를 각각 측정하고 측정된 데이터가 제어 설정된 전류, 전압, 전력 제어 데이터에 미달되면 연결된 태양광모듈을 차단 스위칭하는 노드 제어부들;태양광 모듈들로부터 측정되는 전압, 전류, 전력 측정 데이터를 상기 노드 제어부들로부터 전송받아 상기 측정 데이터를 파싱 처리하여 측정 데이터를 저장하는 게이트웨이부;상기 게이트웨이로부터 파싱된 데이터를 전송받아 분류, 대조, 분석 을 처리하여 데이터베이스에 저장하고, 노드 제어부들을 제어하기 위한 제어 데이터를 설정하는 제어 명령을 상기 게이트웨이부로 전송하는 실시간 제어 모듈; 및상기 실시간 제어 모듈과 인터넷 또는 통신망으로 연결되어 태양광 모듈을 포함하는 태양광 발전 구성 장치 및 데이터를 모니터링하며 상기 실시간 제어모듈로부터 전송된 데이터를 머신러닝 기반에 의하여 학습하고 태양광 발전 제어에 필요한 기능 데이터를 추출하여 모델링을 수행하며 모델링 결과에 따른 제어 서비스 데이터를 실시간 제어 모듈로 제공하는 머신 러닝 서버를 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 제 1 항에 있어서,상기 노드 제어부는,태양광 모듈에서 생산되는 전압, 전류, 전력 데이터를 측정하는 측정부;상기 측정된 데이터와 상기 실시간 제어모듈로부터 전송된 전류, 전압을 제어하기 위하여 정의되는 제어 데이터를 비교하여 측정된 데이터가 정의된 제어 데이터에 미달여부를 판단하여 태양광 모듈 동작을 스위칭 제어하는 분석 컨트롤부;상기 노드 제어부 내부에 저장된 데이터를 외부로 전송하고 실시간 제어모듈로부터 제어 데이터를 받아서 노드 제어부로 전송하는 데이터 전송 관리부; 및상기 분석 컨트롤부에서 판단된 결과에 따라서 상기 노드 제어부에 연결되는 태양광모듈을 차단하거나 연결하는 스위칭부를 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 제 1 항에 있어서,상기 게이트웨이부는,상기 노드 제어부로부터 수신된 측정 데이터를 실시간 제어 모듈로 송신하고 실시간 제어 모듈로부터 전송되는 제어 명령 데이터를 수시하여 상기 노드 제어부로 전송하는 데이터 수집 관리부;노드 제어부로부터 수신된 측정 데이터를 파싱 처리하는 데이터 처리부; 및상기 데이터 처리부에 의하여 파싱 처리된 데이터를 저장하는 데이터 저장부를 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 제 1 항에 있어서,상기 실시간 제어 모듈은,게이트웨이부와 실시간 제어 모듈 사이와 실시간 제어 모듈과 통합 정보 서버 사이에서 실시간으로 데이터 송수신을 위한 인터페이스부;태양광 모듈에서 측정된 전류, 전압, 전력 데이터를 상기 게이트웨이부를 통하여 전송받아 분류, 대조, 분석을 처리하는 데이터 분석부;상기 데이터 분석부의 처리 결과를 이용하여 태양광 모듈 제어에 필요한 기준 데이터를 추출하여 제어 명령을 판단하는 제어 판단부;태양광 모듈을 제어하기 위하여 제어 데이터를 정의하고 게이트웨이로 실시간으로 전송하는 제어 명령 처리부; 및실시간 제어 모듈 내부에서 태양광 모듈을 제어하기 위한 제어 기능을 추가하거나 삭제하는 관리 기능을 수행하는 서비스 관리부를 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 제 1 항에 있어서,상기 머신 러닝 서버는,머신 러닝 서버에서 실시간으로 데이터 송수신을 위한 인터페이스부;태양광 모듈, 노드 제어부, 게이트웨이부, 실시간 제어 모듈을 포함하는 태양광 발전 구성 장치로부터 전송되는 태양광 발전 데이터를 분석, 분류, 대조, 변환을 특성에 따라 통합 처리하여 감시하는 모니터링부;모니터링부에서 전송되는 상기 통합 처리된 테이터의 프로파일 정보에 따라 기계 학습 처리 필요 여부를 판단하는 판단부;상기 판단부에 따라 신규 학습이 필요한 데이터에 대하여 신규학습을 수행하는 학습 모듈;상기 학습된 결과를 이용하여 데이터를 추출하고 모델링을 수행하는 모델링부;상기 학습 결과와 모델링 수행 결과를 저장하는 학습 데이터베이스;상기 학습 데이터베이스에서 태양광 발전 제어를 위해서 모델링 데이터를 가져오고 상기 판단부에서 학습 판단되지 않은 데이터와 함께 태양광 발전 제어를 위한 서비스부로 전송하는 제어부;상기 제어부로부터 전달되는 기준 모델들 중에서 선택하여 태양광 제어 서비스 데이터를 정의하고 인터페이스를 통하여 실시간 제어 모듈로 전송하는 서비스부; 및상기 서비스부에 전달된 모델링 데이터 리스트를 삭제하거나 추가하고 업데이트하여 관리하기 위한 서비스 관리 모듈을 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 제 5 항에 있어서,상기 모니터링부는 데이터 인터페이스롤 통하여 실시간 제어모듈과 데이터 송수신을 처리하며,상기 서비스부는 개방형 인터페이스부와 연결되어 제어부에서 전달 받은 데이터를 응용 서비스로 변환하여 제공하거나 다양한 응용프로그램에서 요청하는 데이터를 제공하도록 한 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 시스템.
- 노드 제어부가 태양광 모듈들로부터 생산되는 전압, 전류, 전력 데이터를 각각 측정하는 단계;게이트웨이부는 측정 데이터를 노드 제어부들로부터 전송받아 파싱 처리하고 저장하는 단계;실시간 제어 모듈은 파싱된 데이터를 분류, 대조, 분석을 처리하여 데이터베이스에 저장하고 머신 러닝 서버로 전송하는 단계;머신 러닝 서버는 실시간 제어모듈로부터 전송된 데이터를 머신러닝 기반에 의하여 학습하고 태양광 발전 제어에 필요한 기능 데이터를 추출하여 모델링을 수행하며 모델링 결과에 따른 제어 서비스 데이터를 실시간 제어 모듈로 제공하는 단계;실시간 제어모듈은 노드 제어부들을 제어하기 위한 제어 명령 데이터를 게이트웨이부로 전송하는 단계;게이트웨이부는 실시간 제어 모듈로부터 수신된 제어 명령 데이터를 노드 제어부로 전송하는 단계;노드 제어부는 태양광 모듈에서 측정된 데이터가 제어 명령 데이터에 따라 설정된 전류, 전압, 전력 제어 기준 데이터에 미달되면 연결된 태양광모듈을 차단 스위칭하는 단계를 포함하는 것을 특징으로 하는 머신러닝 기반 태양광 발전 제어 방법.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP19889719.1A EP3890185B1 (en) | 2018-11-27 | 2019-11-07 | System and method for controlling solar photovoltaic power generation on basis of machine learning |
| AU2019388525A AU2019388525B2 (en) | 2018-11-27 | 2019-11-07 | System and method for controlling solar photovoltaic power generation on basis of machine learning |
| US17/296,965 US11953870B2 (en) | 2018-11-27 | 2019-11-07 | System and method for controlling solar photovoltaic power generation on basis of machine learning |
| CN201980078435.9A CN113169709A (zh) | 2018-11-27 | 2019-11-07 | 基于机器学习的太阳能发电控制系统及方法 |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020180148821A KR101958941B1 (ko) | 2018-11-27 | 2018-11-27 | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 |
| KR10-2018-0148821 | 2018-11-27 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020111561A1 true WO2020111561A1 (ko) | 2020-06-04 |
Family
ID=65949061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/KR2019/015018 Ceased WO2020111561A1 (ko) | 2018-11-27 | 2019-11-07 | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US11953870B2 (ko) |
| EP (1) | EP3890185B1 (ko) |
| KR (1) | KR101958941B1 (ko) |
| CN (1) | CN113169709A (ko) |
| AU (1) | AU2019388525B2 (ko) |
| WO (1) | WO2020111561A1 (ko) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101958941B1 (ko) * | 2018-11-27 | 2019-03-18 | 주식회사 나눔에너지 | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 |
| CN110072208B (zh) * | 2019-05-16 | 2024-06-11 | 信息产业电子第十一设计研究院科技工程股份有限公司 | 一种光伏发电为基础的无线智能网络通信装置 |
| CN110658761A (zh) * | 2019-09-27 | 2020-01-07 | 北京四达时代软件技术股份有限公司 | 太阳能供电控制方法及装置 |
| KR102823503B1 (ko) * | 2019-12-09 | 2025-06-20 | (주)링크제니시스 | 인공지능 모듈을 구비하는 데이터 수집 게이트웨이 |
| KR102686512B1 (ko) * | 2019-12-23 | 2024-07-19 | (주)링크제니시스 | 인공지능 모듈을 구비하는 산업용 통신 프로토콜 변환 장치 및 데이터 수집 게이트웨이 |
| KR102228512B1 (ko) | 2020-11-05 | 2021-03-16 | 에이펙스인텍 주식회사 | 머신러닝을 이용한 태양광 발전설비의 고장 및 접속함 화재 감시기능과 원격제어기능을 구비한 태양광 발전장치 |
| KR102272948B1 (ko) | 2020-11-18 | 2021-07-06 | 센트리닉스 주식회사 | 비중앙적 무선 메쉬 네트워크 기반 분산형 모듈 레벨 태양광 발전 감시제어 시스템 |
| KR102420957B1 (ko) * | 2021-08-31 | 2022-07-14 | 주식회사 대양이엔씨 | IoT를 이용한 태양광 발전 모니터링 시스템 |
| KR20240088486A (ko) * | 2022-12-13 | 2024-06-20 | 엘지이노텍 주식회사 | 광 발전 시스템 |
| KR102901056B1 (ko) * | 2024-02-13 | 2025-12-18 | 한국에너지기술연구원 | 분산자원 건전성 관리 플랫폼 및 그 제어방법 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130001389A (ko) * | 2011-06-27 | 2013-01-04 | 최석준 | 효율적 태양광 발전 시스템 |
| JP2014155271A (ja) * | 2013-02-06 | 2014-08-25 | Hitachi Advanced Digital Inc | 太陽光発電設備のための監視システム |
| KR20180008820A (ko) * | 2018-01-09 | 2018-01-24 | 홍익대학교세종캠퍼스산학협력단 | 태양광 발전의 모니터링 시스템 및 그 방법 |
| KR20180112496A (ko) * | 2017-04-04 | 2018-10-12 | 엘에스산전 주식회사 | 전력 전송 경로 제어가 가능한 태양광 발전 시스템 |
| KR20180123951A (ko) * | 2017-05-10 | 2018-11-20 | 주식회사 알스피릿 | 기계학습 기반 실시간 태양광 셀 또는 어레이 상태 진단/예측 장치 및 방법 |
| KR101958941B1 (ko) * | 2018-11-27 | 2019-03-18 | 주식회사 나눔에너지 | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8264195B2 (en) * | 2008-10-01 | 2012-09-11 | Paceco Corp. | Network topology for monitoring and controlling a solar panel array |
| WO2010096783A1 (en) * | 2009-02-20 | 2010-08-26 | The Trustees Of Columbia University In The City Of New York | Dynamic contingency avoidance and mitigation system |
| TW201212251A (en) * | 2010-09-13 | 2012-03-16 | Wen-Bin Huang | Solar power generation apparatus and solar power generation system |
| KR101457643B1 (ko) | 2013-02-20 | 2014-11-10 | 한국 전기안전공사 | 태양광 발전설비 진단시스템 |
| KR101664953B1 (ko) | 2015-11-10 | 2016-10-11 | (주)대은 | 태양광 모듈 진단시스템 |
| JP6546213B2 (ja) * | 2017-04-13 | 2019-07-17 | ファナック株式会社 | 回路構成最適化装置及び機械学習装置 |
| CN108647716B (zh) * | 2018-05-09 | 2021-08-06 | 北京理工大学 | 一种基于复合信息的光伏阵列故障诊断方法 |
| US11009389B2 (en) * | 2018-07-09 | 2021-05-18 | International Business Machines Corporation | Operating re-configurable solar energy generators for increasing yield during non-ideal weather conditions |
-
2018
- 2018-11-27 KR KR1020180148821A patent/KR101958941B1/ko active Active
-
2019
- 2019-11-07 WO PCT/KR2019/015018 patent/WO2020111561A1/ko not_active Ceased
- 2019-11-07 EP EP19889719.1A patent/EP3890185B1/en active Active
- 2019-11-07 US US17/296,965 patent/US11953870B2/en active Active
- 2019-11-07 CN CN201980078435.9A patent/CN113169709A/zh not_active Withdrawn
- 2019-11-07 AU AU2019388525A patent/AU2019388525B2/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130001389A (ko) * | 2011-06-27 | 2013-01-04 | 최석준 | 효율적 태양광 발전 시스템 |
| JP2014155271A (ja) * | 2013-02-06 | 2014-08-25 | Hitachi Advanced Digital Inc | 太陽光発電設備のための監視システム |
| KR20180112496A (ko) * | 2017-04-04 | 2018-10-12 | 엘에스산전 주식회사 | 전력 전송 경로 제어가 가능한 태양광 발전 시스템 |
| KR20180123951A (ko) * | 2017-05-10 | 2018-11-20 | 주식회사 알스피릿 | 기계학습 기반 실시간 태양광 셀 또는 어레이 상태 진단/예측 장치 및 방법 |
| KR20180008820A (ko) * | 2018-01-09 | 2018-01-24 | 홍익대학교세종캠퍼스산학협력단 | 태양광 발전의 모니터링 시스템 및 그 방법 |
| KR101958941B1 (ko) * | 2018-11-27 | 2019-03-18 | 주식회사 나눔에너지 | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP3890185A4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| US11953870B2 (en) | 2024-04-09 |
| EP3890185A4 (en) | 2022-08-31 |
| KR101958941B1 (ko) | 2019-03-18 |
| EP3890185A1 (en) | 2021-10-06 |
| AU2019388525B2 (en) | 2022-09-01 |
| AU2019388525A1 (en) | 2021-06-17 |
| EP3890185C0 (en) | 2023-07-19 |
| US20220029423A1 (en) | 2022-01-27 |
| CN113169709A (zh) | 2021-07-23 |
| EP3890185B1 (en) | 2023-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020111560A1 (ko) | 태양광 발전 밸런싱 제어 시스템 및 방법 | |
| KR101958941B1 (ko) | 머신러닝 기반 태양광 발전 제어 시스템 및 방법 | |
| CN106054672A (zh) | 基于rt‑lab的真实微电网运行动态仿真测试平台 | |
| WO2013032062A1 (ko) | 태양광 모듈 모니터링 시스템 | |
| WO2022222507A1 (zh) | 一种mlpe光伏系统及其mlpe设备检测方法 | |
| CN111799840A (zh) | 一种离并网一体化电源控制方法及系统 | |
| WO2013094838A1 (ko) | 단위 그룹별 최대전력점 추종을 수행하는 태양광 발전 시스템 | |
| WO2024019474A1 (ko) | 태양광 인버터 기능을 포함한 양방향 인버터 | |
| WO2023128031A1 (ko) | 머신러닝을 이용한 발전 저하 상황 분류가 가능한 태양광 발전 모니터링 시스템 및 그 방법 | |
| WO2020027538A1 (ko) | 발전 시스템의 이상 유무 판단 장치 및 방법 | |
| WO2018062624A1 (ko) | 스카다 시스템의 데이터 처리 장치 | |
| CN106971526A (zh) | 光伏应用系统远程监测系统 | |
| WO2024034758A1 (ko) | 원격지에서 태양광 발전시스템의 출력을 제한하는 장치 및 그 방법 | |
| WO2014046328A1 (ko) | 계통 데이터 축약 시스템 및 그 방법 | |
| CN110022130A (zh) | 一种光伏阵列故障检测设备和方法 | |
| CN110190990B (zh) | 一种低压配电台区网络拓扑结构的自动辨识方法及装置 | |
| CN110460155A (zh) | 具备图像处理及比对功能的馈线终端及其故障监测方法 | |
| CN114866031A (zh) | 分布式光伏发电状态监测系统 | |
| Ming et al. | Research on panoramic monitoring of distributed photovoltaic power station based on power Iot platform | |
| CN210468895U (zh) | 具备图像处理及比对功能的馈线终端 | |
| WO2014046315A1 (ko) | 분산전원이 연계된 전력계통의 통합 해석 장치 및 방법 | |
| CN121618700B (zh) | 重要负载备电管理器 | |
| CN220673456U (zh) | 一种光伏发电系统远程数据采集装置 | |
| CN219833840U (zh) | 配电自动化主站监测装置 | |
| WO2019225771A1 (ko) | 태양전지의 전압측정 시스템 및 이를 이용한 태양전지의 개별 고장진단 방법 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19889719 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 2019388525 Country of ref document: AU Date of ref document: 20191107 Kind code of ref document: A |
|
| ENP | Entry into the national phase |
Ref document number: 2019889719 Country of ref document: EP Effective date: 20210628 |