System and method for operating generator
By constructing a learning model that considers regional and operational factors, the optimal backup generator is selected for failed generators, addressing the instability in power backup and improving operational efficiency and grid stability.
Patent Information
- Application Number
- PCT/KR2024/006779
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-23
- Filing Date
- 2024-05-20
- Publication Date
- 2024-11-28
AI Technical Summary
In power trading systems, when a generator fails due to an accidental breakdown, the backup generator selection is often based on factors like maintenance schedules and efficiency, leading to unstable grid frequency, as it doesn't consider regional information, energy source characteristics, and transmission network conditions, resulting in inefficient power backup.
A learning model is built using real-time data on generator capacity, energy source characteristics, maintenance periods, and accident history to optimize the selection of the backup generator, ensuring the optimal power backup generator is chosen based on constructed learning models for each failure scenario.
This approach improves the efficiency of generator operations by minimizing backup charging time and enhancing the overall versatility of the generator system by selecting the most suitable backup generator, leading to a more stable power grid.
Smart Images

Figure KR2024006779_28112024_PF_FP_ABST
Abstract
Description
Generator operating system and method
[0001] The present invention relates to a system and method for operating a generator, and more particularly, to a technology for backing up power from an optimized backup generator in the event of a generator failure due to an accident.
[0002] In a typical power trading system, when a random generator fails due to an accidental failure, power is backed up from an adjacent generator based on the judgment of the power exchange staff or a preset accident scenario.
[0003] Here, the power backup generator is selected based on an accident scenario determined by taking into account the generator for which the planned preventive maintenance period has arrived, the generator efficiency at the time of the accident, and the unit cost of generation.
[0004] Therefore, since the power backup generator is selected regardless of the power backup generator's generation method, energy source, and capacity, there was a problem of unstable grid frequency of the power backup generator.
[0005] Accordingly, the applicant proposes a method of constructing a learning model by optimizing backup generators by reflecting regional information including the distance between adjacent generators and transmission network conditions for each of multiple generators, operational information including generator capacity and energy source characteristics, maintenance period, and maintenance and accident history, and selecting an optimal backup generator based on the constructed learning model for input generators with accident failures.
[0006] Meanwhile, the background technology described above is technical information that the inventor possessed for the purpose of deriving the present invention or acquired during the process of deriving the present invention, and cannot necessarily be said to be publicly known technology disclosed to the general public prior to the application for the present invention.
[0007] [Prior Art Literature]
[0008] [Patent Document]
[0009] (Patent Document 1) Korean Patent No. 10-2105194 (Registration Date: April 21, 2020)
[0010] The technical problem to be achieved by the present invention is to construct a learning model for optimizing backup generators by reflecting regional information including the distance between adjacent generators and transmission network conditions for each of a plurality of generators, operation information including the characteristics of the generator capacity and energy source, maintenance period, and maintenance and accident history, and time information, and to select an optimal backup generator based on the learning model for an input accident-fault generator based on the constructed learning model.
[0011] The purpose of the present invention is not limited to the aforementioned purposes, and other unmentioned purposes and advantages of the present invention can be understood through the following description and will be more clearly understood through the embodiments of the present invention. Furthermore, it will be readily apparent that the purposes and advantages of the present invention can be realized by the means and combinations thereof set forth in the claims.
[0012] According to one embodiment of the present invention, the operating system of a generator according to one embodiment is:
[0013] A memory that stores computer-readable instructions; and
[0014] comprising at least one processor implemented to execute instructions stored in the memory;
[0015] At least one processor,
[0016] A data collection unit that collects real-time information for each of multiple generators interconnected through a mutually distributed communication network; and
[0017] It is characterized by including a learning model construction unit that constructs a learning model for selecting an optimal power backup generator matching each generator based on the collected real-time information.
[0018] Preferably, the above real-time information
[0019] For multiple generators, at least one of regional information including generator distance and transmission network conditions, operational information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history may be provided.
[0020] Preferably, said at least one processor,
[0021] The system may further include a learning unit that learns about the input of a failed generator based on a learning model to select an optimal power backup generator and connects a network between the selected power backup generator and the failed generator.
[0022] A method of operating a generator according to another embodiment of the present invention is
[0023] At least one processor comprising the operating system of the generator,
[0024] A data collection step for collecting real-time information on each of multiple generators interconnected through a mutually distributed network; and
[0025] It is characterized by including a learning model construction step for constructing a learning model to select the optimal power backup generator based on real-time information collected from each generator.
[0026] Preferably, the above real-time information
[0027] For multiple generators, at least one of regional information including generator distance and transmission network conditions, operational information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history may be provided.
[0028] Preferably, said at least one processor,
[0029] The method may further include a learning step of selecting an optimal power backup generator based on a learning model for input of a generator that has experienced an accidental failure, and connecting a network between the selected power backup generator and the generator that has experienced an accident.
[0030] According to these characteristics, for each of multiple generators, the backup generator is optimized by reflecting regional information including the distance between adjacent generators, transmission network conditions, generator capacity, operation information including the characteristics of energy sources, maintenance period, and maintenance and accident history, and a learning model is constructed, and based on the constructed learning model, the optimal backup generator is selected based on the learning model for the input accident-faulted generator, thereby fundamentally improving the efficiency of generator operation.
[0031] This can minimize the loss and backup charging time for generators that have malfunctioned due to the selection of the optimal power backup generator.
[0032] In addition, according to the present invention, the versatility of the generator's operating system can be improved by selecting an optimal power backup generator based on a lightweight device.
[0033] The following drawings attached to this specification illustrate preferred embodiments of the present invention, and together with the detailed description of the invention described below, serve to further understand the technical idea of the present invention, and therefore, the present invention should not be interpreted as being limited to matters described in such drawings.
[0034] Figure 1 is a diagram showing a generator using a distributed communication network to which one embodiment is applied.
[0035] Figure 2 is a configuration diagram of the operating system of the generator of Figure 1.
[0036] Figure 3 is a detailed configuration diagram of the processor of Figure 2.
[0037] Figure 4 is a diagram illustrating a concept of selecting an optimal backup generator according to one embodiment.
[0038] Figure 5 is a flowchart showing the overall operation process of a generator of another embodiment.
[0039] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement the invention. However, the present invention can be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts irrelevant to the description are omitted for clarity of description, and similar parts are designated with similar reference numerals throughout the specification.
[0040] The following embodiment specifically describes a configuration for constructing a learning model by optimizing backup generators by reflecting regional information including the distance between adjacent generators, transmission network conditions, and operation information including the characteristics of the generator capacity and energy source, as well as timing information such as maintenance periods and maintenance and accident history, for each of a plurality of generators, and selecting an optimal backup generator based on the learning model for an input accident-fault generator based on the constructed learning model.
[0041] FIG. 1 is a diagram showing the structure of a distributed communication network to which one embodiment is applied, FIG. 2 is a configuration diagram of an operating system of a generator of FIG. 1, FIG. 3 is a detailed configuration diagram of at least one processor of FIG. 2, and FIG. 4 is an exemplary diagram explaining the concept of selecting an optimal backup generator by performing the process based on the processor of FIG. 2. Referring to FIG. 1, the operating system (S) of the generator of one embodiment learns real-time information on generators A to N included in a plurality of power plants 1 connected to a distributed communication network as input and selects an optimal backup generator for a corresponding generator Error that has failed.
[0042] Referring to FIGS. 2 and 3, the operating system (S) of the generator of one embodiment is equipped with a memory (1) and at least one processor (2), and at least one processor (2) may include a data collection unit (21), a learning model construction unit (22), a learning unit (23), and a backup unit (24).
[0043] The data collection unit (21) is configured to collect real-time information for each of a plurality of generators interconnected through a mutually distributed communication network. Here, the real-time information may be at least one of regional information including generator distance and transmission network conditions for a plurality of generators, operational information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history.
[0044] That is, the data collection unit (21) collects, for each of a plurality of generators connected through a mutually distributed communication network, the distance between generators and transmission network conditions, the capacity and energy source characteristics of each generator, maintenance period and accident history information, and then matches them with the identifier of each generator and stores them.
[0045] The learning model construction unit (22) constructs a learning model that selects the optimal power backup generator for a randomly set generator by inputting real-time information between each collected generator.
[0046] The learning model construction unit (22) derives a probability value for each collected real-time information through AI (artificial intelligence) based on a deep learning neural network such as convolution, and creates a weight for each real-time information using the probability value for each derived real-time information.
[0047] Accordingly, the learning model construction unit (22) selects a power backup generator matching the generator based on each real-time information and weight, and constructs a learning model in the form of a pair of labels for the generator that has failed and the power backup generator matching it.
[0048] The learning unit (23) learns the input of the generator in which an accident has occurred based on a learning model, selects the optimal power backup generator matching the generator in which an accident has occurred, and outputs the selected power backup generator.
[0049] For example, referring to (a) of Fig. 4, each power plant 1 can have multiple generators A to D that can be operated independently, and power plant 2 can have generators AB and generator CD that can be operated interdependently.
[0050] For example, referring to (b) of FIG. 4, if it is assumed that power plant 1, in which multiple generators A to C are each operated at 100% of the power capacity and generator D is operated at 20%, and the corresponding generator E has failed, the learning unit (23) selects the generator D, which is operating at 100% of the power capacity, as the optimal backup generator based on the input real-time information of power plant 1, and provides 80% of the power capacity of the selected optimal backup generator D to the corresponding generator E, which has failed. Accordingly, the optimal backup generator D, which has received the power supply instruction of the learning unit (23), additionally generates 80% of the power capacity, and 80% of the additionally generated power capacity is transferred to the failed generator E to perform the backup function.
[0051] As another example, referring to (c) of FIG. 4, assuming that generators A and B of power plant 2, which operate interdependently, generate 100% of the power, and generators C and D generate 50% of the power, and that generator E fails, the learning unit (23) selects generators C and D, which are operating at 100%, as optimal backup generators based on the input real-time information of power plant 2, and additionally generates 80% and 70% of the power of the selected optimal backup generators C and D, and provides the additionally generated power to the corresponding generator E, which failed. Accordingly, the optimal backup generator CD, which received the power dispatch instruction from the learning unit (23), additionally generates 80% and 70% of the power, and the additionally generated power is provided to the corresponding generator E, which failed.
[0052] Meanwhile, the learning unit (23) can evaluate the performance of the learning model based on the learning results and train the learning model by reflecting the evaluation results. At this time, the evaluation index of the learning model includes precision, recall, and accuracy.
[0053] Here, precision is the ratio of the optimal power backup generator selected by the learning model to the actual power backup generator, recall is the ratio of the actual power backup generator to the optimal power backup generator in the learning model, and accuracy is the ratio of the optimal power backup generator to the total predicted power backup generators.
[0054] Meanwhile, as described above, the backup unit (24) receives power from the selected optimal power backup generator and charges the generator in which an accident has occurred.
[0055] In this embodiment, for each of a plurality of generators, a learning model is constructed by optimizing a backup generator by reflecting regional information including the distance between adjacent generators, transmission network conditions, operation information including generator capacity and energy source characteristics, maintenance period, and maintenance and accident history, and selecting an optimal backup generator based on the learning model for an input accident-fault generator based on the constructed learning model.
[0056] FIG. 5 is a flowchart showing the operation process of the generator of FIG. 2, and an operation method of a generator according to another embodiment of the present invention is described with reference to FIG. 5.
[0057] In the data collection step (S1), at least one processor (2) of the generator operating system according to one embodiment collects real-time information among regional information including generator distance and transmission network conditions, operation information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history for each of a plurality of generators interconnected through a mutually distributed network.
[0058] And the collected real-time information is provided to the learning model construction stage (S3).
[0059] The learning model construction stage (S3) builds a learning model that selects the optimal power backup generator based on real-time information collected from each generator.
[0060] In other words, the learning model selects the optimal power backup generator matching each generator based on real-time information, and constructs a pair of labels for the generators matching the selected optimal power backup generator.
[0061] And, based on the learning step (S5), at least one processor (2) learns about the generator in which the input fault occurred based on the learning model and selects the optimal power backup generator. And, based on the backup step (S7), at least one processor (2) charges the generator in which the fault occurred through the selected optimal power backup generator.
[0062] Accordingly, one embodiment constructs a learning model by optimizing backup generators by reflecting regional information including the distance between adjacent generators, transmission network conditions, and operation information including the characteristics of the generator capacity and energy source, maintenance period, and maintenance and accident history, and selecting an optimal backup generator based on the constructed learning model for an input accident-fault generator, thereby selecting an optimal power backup generator.
[0063] For ease of understanding, the processor is sometimes described as being used alone. However, those skilled in the art will recognize that a processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processor may include multiple processors or a single processor and a single controller. Additionally, other processing configurations, such as parallel processors, are also possible.
[0064] Here, the software may include a computer program, code, instructions or a combination of one or more of these, which may configure a processing device to perform a desired operation or may independently or collectively command a processing device.
[0065] Software and / or information, signals and data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage media or device, or transmitted signal wave for interpretation by a control unit or for providing instructions or data to a processing unit.
[0066] The software may be distributed across networked computer systems, stored and executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.
[0067] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., either singly or in combination. The program commands recorded on the medium may be those specifically designed and configured for the embodiment or may be known and usable by those skilled in the art of computer software.
[0068] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specifically configured to store and execute program instructions such as ROMs, RAMs, and flash memory.
[0069] Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.
[0070] The above hardware device may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.
[0071] Although the embodiments described above have been described by way of limited examples and drawings, those skilled in the art will appreciate that various modifications and variations can be made based on the above teachings. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.
[0072] Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the claims described below but also by equivalents of the claims.
[0073] [Explanation of symbols]
[0074] 1: Memory
[0075] 2: At least one processor
[0076] 21: Data Collection Department
[0077] 22: Learning Model Construction Department
[0078] 23: Learning Department
[0079] 24: Backup Department
Claims
1. Memory that stores computer-readable instructions; and At least one processor configured to execute instructions stored in the memory, At least one processor, A data collection unit that collects real-time information for each of multiple generators interconnected through a mutually distributed communication network; and An operating system for a generator, including a learning model construction unit that constructs a learning model for selecting an optimal power backup generator matching each generator based on the collected real-time information.
2. In paragraph 1, the real-time information is An operating system for a generator, which includes at least one of regional information including generator distance and transmission network conditions for multiple generators, operational information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history.
3. In the first paragraph, the at least one processor, An operating system for a generator further comprising a learning unit that learns about the input of a failed generator based on a learning model to select an optimal power backup generator and connects a network between the selected power backup generator and the failed generator.
4. In the method of operating a generator performed in the generator operating system of paragraph 1, At least one processor comprising the operating system of the generator, A data collection step for collecting real-time information on each of multiple generators interconnected through a mutually distributed network; and A method for operating a generator, including a learning model construction step for constructing a learning model that selects the optimal power backup generator based on real-time information collected from each generator.
5. In paragraph 4, the real-time information is A method of operating a generator, comprising at least one of regional information including generator distance and transmission network conditions for multiple generators, operational information including generator capacity and energy source characteristics, and timing information such as maintenance period and maintenance and accident history.
6. In the fourth paragraph, the at least one processor, A method for operating a generator further comprising a learning step of selecting an optimal power backup generator based on a learning model for input of a generator in which an accident has occurred and connecting a network between the selected power backup generator and the generator in which an accident has occurred.
7. A computer-readable recording medium having recorded thereon a program for executing the operating method of the generator of any one of the preceding clauses 4 to 6 on a computer.
8. In a computer program stored in a computer-readable recording medium to cause the computer to execute a method of operating a generator by being coupled with a computer, The method of operating the above generator is as follows: A data collection step for collecting real-time information on each of multiple generators interconnected through a mutually distributed network; and A learning model construction step for building a learning model that selects the optimal power backup generator based on real-time information collected from each generator; and An operation program for a generator that includes a learning step for selecting an optimal power backup generator by learning based on a learning model for the input of a failed generator, and for connecting a network between the selected power backup generator and the failed generator, thereby deriving the selected optimal power backup generator.
Citation Information
Patent Citations
A data acquisition and supervisory control system
KR100964296B1
Stand-alone microgrid control system and control method thereof
KR1020140060401A
Demand Response Market participation system through a emergency generator of customer site.
KR1020140128722A
Operation system and method for virtual power plant using risk analysis
KR102105194B1
KR20210143045A
