WO2022114634A1 - 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치 및 방법 - Google Patents
신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치 및 방법 Download PDFInfo
- Publication number
- WO2022114634A1 WO2022114634A1 PCT/KR2021/016627 KR2021016627W WO2022114634A1 WO 2022114634 A1 WO2022114634 A1 WO 2022114634A1 KR 2021016627 W KR2021016627 W KR 2021016627W WO 2022114634 A1 WO2022114634 A1 WO 2022114634A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- abnormal state
- abnormal
- state determination
- neural network
- network model
- Prior art date
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 188
- 238000003062 neural network model Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title claims description 31
- 230000005856 abnormality Effects 0.000 claims abstract description 8
- 238000012800 visualization Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 239000002826 coolant Substances 0.000 claims description 7
- 238000009795 derivation Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 208000024891 symptom Diseases 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 12
- 238000012544 monitoring process Methods 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- ZZUFCTLCJUWOSV-UHFFFAOYSA-N furosemide Chemical compound C1=C(Cl)C(S(=O)(=O)N)=CC(C(O)=O)=C1NCC1=CC=CO1 ZZUFCTLCJUWOSV-UHFFFAOYSA-N 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21D—NUCLEAR POWER PLANT
- G21D3/00—Control of nuclear power plant
- G21D3/04—Safety arrangements
- G21D3/06—Safety arrangements responsive to faults within the plant
-
- 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
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21C—NUCLEAR REACTORS
- G21C17/00—Monitoring; Testing ; Maintaining
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21D—NUCLEAR POWER PLANT
- G21D3/00—Control of nuclear power plant
- G21D3/001—Computer implemented control
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/10—Plc systems
- G05B2219/16—Plc to applications
- G05B2219/161—Nuclear plant
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24063—Select signals as function of priority, importance for diagnostic
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E30/00—Energy generation of nuclear origin
- Y02E30/30—Nuclear fission reactors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Definitions
- the present invention relates to an apparatus and method for tracking an abnormal state determination basis, and more particularly, to an apparatus and method for determining an abnormal state using a neural network model and tracking the basis for the determination.
- the operator determines what kind of abnormal operation condition has occurred based on the alarm of the power plant, and takes appropriate action according to the procedure for abnormal operation condition.
- the method of determining and providing the abnormal state of the nuclear power plant by learning the operation data of the abnormal state using the neural network model so that the operator can quickly determine the abnormal state due to the failure of the equipment or the abnormality of the equipment and take appropriate measures. is being studied
- the neural network model it is difficult to track which power plant operation data changes are the basis for judging an abnormal state.
- Patent Document 1 Korean Patent Registration 2095653 (Apparatus and Method for Determining Abnormal Driving State Using Neural Network Model, Korea Hydro & Nuclear Power Co., Ltd.)
- Patent Document 2 US Registered Patent 10452845 (Generic framework to detect cyber threats in electric power grid, GENERAL ELECTRIC COMPANY)
- Patent Document 2 US Patent Publication 20190164057 (Mapping and quantification of influence of neural network features for explainable artificial intelligence, INTEL CORP)
- the present invention is to solve the problems of the prior art as described above, and when determining the abnormal state of a nuclear power plant to which the neural network model is applied, a method for estimating the operating variable that is the basis of the judgment and the abnormal state determination using the neural network model
- An object of the present invention is to provide an evidence tracking device.
- an apparatus for tracking the basis for determining an abnormal state using a neural network model includes an abnormality type classification unit that classifies the abnormal state into a plurality of failures in an abnormal driving scenario in which a plurality of scenarios related to the abnormal state are stored, and the An operation variable derivation unit for deriving an operating variable that affects the abnormal state determination result for each of the plurality of classified failures, and a weighting unit for each power plant operation variable that assigns weights to variables related to the abnormal state among the operation variables and an abnormal state determination basis generating unit that tracks an abnormal state determination basis from the abnormal state determination result generated through the weighted power plant operation variable.
- the weighting unit for each power plant operation variable that gives weight to the variable related to the abnormal state among the operating variables is classified in consideration of the physical correlation of the system related to the abnormal state, and weights the physical variable related to the abnormal state It is characterized in that it is given.
- the abnormal type classification unit includes at least one of a valve leakage, a pump failure, a heat exchanger failure, and a coolant leakage in the abnormal operation scenario.
- the operation variable derivation unit for deriving an operation variable affecting the abnormal state determination result for each of the plurality of classified failures, when classified as the valve leakage the flow rate of the system related to the valve is converted to the pump failure.
- classification it includes deriving the flow rate and pressure of the system related to the valve, the temperature of the system related to the heat exchanger when classified as a failure of the heat exchanger, and deriving the radiation level of the leakage area when classified as the coolant leakage do.
- the physical variables related to the abnormal state are prepared with reference to the abnormal procedure or actual power plant operation history.
- the reason for determining the abnormal state is the driving variable that can be distinguished from other abnormal states, and is used to verify the abnormal state determination logic used in the abnormal state determination system.
- the reason for determining the abnormal state is used to verify the symptoms described in the abnormal procedure, and the abnormal procedure describes the operating variable that is changed when the abnormal state occurs.
- the method includes the steps of: generating an abnormal state determination result at the last end of the neural network model by performing learning using power plant operation data and a neural network model; and performing influence analysis on the abnormal state determination result in a fully connected layer before generating the state determination result, and extracting variable values affecting the abnormal state determination result.
- the step of extracting the variable values affecting the abnormal state determination result is to apply a visualization algorithm to virtually generate visualized input change data, and to use the virtual input change data as an input through calculation of the neural network model. and analyzing the influence of the input change data on the change of the abnormal state determination result, and extracting the input change data that most contributes to deriving the change of the abnormal state determination result.
- the apparatus and method for tracking abnormality determination grounds using a neural network model can determine the type of the abnormal state within a short time and provide it to the operator when various abnormal states occur, and accordingly, the nuclear power plant It is possible to quickly and accurately respond to abnormal conditions in nuclear power plants, thereby improving the safety of nuclear power plants.
- FIG. 1 is a diagram schematically illustrating an apparatus for determining an abnormal state to which a conventional neural network model is applied.
- FIG. 2 is a diagram schematically illustrating an operation process of an abnormal state determination basis tracking apparatus to which a neural network model is applied according to an embodiment of the present invention.
- FIG. 3 is a diagram schematically illustrating an operation process of extracting a driving variable affecting a change in an abnormal state determination result to which a neural network model is applied according to an embodiment of the present invention.
- FIG. 4 is a diagram illustrating an operation process of generating an abnormal state determination basis according to an embodiment of the present invention.
- FIG. 5 is a schematic block diagram of an apparatus for extracting an abnormal state determination basis according to an embodiment of the present invention.
- FIG. 6 is a diagram illustrating the use of a basis for determining an abnormal state derived by applying a neural network model according to an embodiment of the present invention.
- FIG. 1 is a diagram schematically illustrating an apparatus for determining an abnormal state to which a conventional neural network model is applied.
- the abnormal driving state determination apparatus 100 to which the neural network model is applied includes an abnormal driving state data generating unit 110 , an abnormal driving state data learning unit 130 , an abnormal driving state determining unit 150 , It may include an abnormal driving state monitoring unit 170 .
- the abnormal driving state data generator 110 is configured to virtually generate abnormal driving state data based on information about the abnormal driving state, and may include a scenario database 111 and a simulator 112 .
- the scenario database 111 is a configuration in which a plurality of scenarios related to an abnormal driving state are provided. These scenarios include a scenario according to an operating variable related to a temperature change of nuclear power plant devices, a scenario according to an operating variable related to turbine bearing vibration, and the like, and scenarios related to various abnormal operating conditions are stored in the scenario database 111 .
- the operating variables are operating factors for the operating conditions of the devices of the nuclear power plant, and may be included in about 1,000 to 2,000 for each device. These operating variables may include pressure, temperature, flow rate, and the like.
- the simulator 112 is configured to simulate an abnormal driving state with respect to a scenario selected from among the scenarios related to the abnormal driving state stored in the scenario database 111 . Accordingly, data on the abnormal driving state may be virtually generated.
- the abnormal driving state data learning unit 130 is configured to visualize and learn the abnormal driving state by applying a visualization algorithm based on the abnormal driving state data generated by the abnormal driving state data generating unit 110 .
- the abnormal driving state data learning unit 130 may include a first visualization arrangement unit 131 and a second visualization arrangement unit 132 .
- the first visualization arrangement unit 131 is configured to arrange based on the physical locations of operating variables of devices provided in the nuclear power plant. That is, the operating variables may be arranged in the same arrangement as the structure of an actual nuclear power plant.
- the second visualization arrangement unit 132 is configured to preferentially arrange the physically identical driving variables. For example, by arranging temperature-related operating variables in the same zone, event-specific characteristics appear when temperature changes occur.
- the abnormal operation state determination unit 150 learns the abnormal operation state based on the operation variables indicated by the abnormal operation state data learning unit 130 applying a visualization algorithm, and the device acquired from the process monitoring and alarm system of the nuclear power plant It is a configuration that determines whether an abnormal driving state has occurred based on the driving variables.
- the abnormal driving state determining unit 150 may include a neural network model 151 and a signal matching unit 152 .
- the neural network model 151 is configured to learn the abnormal driving state data visualized by the first visualization arrangement unit 131 and the second visualization arrangement unit 132 based on the visualization algorithm.
- the signal matching unit 152 is configured to transmit information on a monitoring signal including information on an abnormal driving state to the devices.
- the abnormal operation state monitoring unit 170 is configured to monitor whether the operation state of each device provided in the nuclear power plant is within a normal range.
- the abnormal driving state monitoring unit 170 may periodically acquire a monitoring signal including information on driving variables of each device and transmit it to the abnormal driving state determining unit 150 .
- FIG. 2 is a diagram schematically illustrating an operation process of an abnormal state determination basis tracking apparatus to which a neural network model is applied according to an embodiment of the present invention.
- the abnormal state determination basis tracking apparatus 200 performs learning using the power plant operation data 210 and the neural network model 230 to generate an abnormal state determination result 250 .
- the neural network model 230 is calculated through multiple layers of neural networks (Deep Learning) in order to effectively learn each abnormal state.
- An abnormal state determination result 250 is generated at the last end of the neural network model 230.
- an abnormal state determination result 250 is performed in the Fly Connected Layer before the abnormal state determination result 250 is generated.
- By performing the influence analysis 270 on variable values affecting the abnormal state determination result 250 are extracted.
- FIG. 3 is a diagram schematically illustrating an operation process of extracting a driving variable affecting a change in an abnormal state determination result to which a neural network model is applied according to an embodiment of the present invention.
- the change data 310 of the visualized input is virtually generated by first applying a visualization algorithm. Using this virtual input change data 310 as an input, the effect of the input change data 310 of each item on the result change 350 is analyzed through the calculation of the neural network model 330, and the result change ( 350), the change data 310 of the input that contributes the most to deriving is extracted.
- FIG. 4 is a diagram illustrating an operation process of generating an abnormal state determination basis according to an embodiment of the present invention.
- FIG. 5 is a schematic block diagram of an apparatus for extracting an abnormal state determination basis according to an embodiment of the present invention.
- the abnormal state determination basis tracking device 500 includes an abnormal operation scenario 510 , an abnormal type classification unit 520 , a driving variable derivation unit 530 , and a weighting unit for each operating variable of the power plant. It may include whether or not 540 and an abnormal state determination basis generating unit 550 .
- the abnormal driving scenario 500 is a configuration in which a plurality of scenarios related to an abnormal driving state are provided. These scenarios include a scenario according to an operating variable related to temperature change of nuclear power plant devices and a scenario according to an operating variable related to turbine bearing vibration, and the abnormal type classification unit 520 divides the abnormal operation scenario 500 into valve leakage, A pump failure, a heat exchanger failure, a coolant leak, etc. are classified, and the operation variable derivation unit 530 for the classified failure derives an operation variable that affects the abnormal determination result.
- the flow rate of the system related to the valve when classified as a pump failure, the flow rate and pressure of the system related to the valve, and when classified as a heat exchanger failure, related to the corresponding heat exchanger If the system temperature is classified as coolant leakage, the radiation level at the leakage site is derived.
- the input range that affects the abnormal state result includes a number of uncertainties, and in order to extract the basis for the abnormal state judgment result, it is important to change the input physically related to the abnormal state.
- the extracted driving variables as the basis for judging the abnormal state are used as the basis for judging the neural network in consideration of the physical correlation of related systems.
- An abnormal state is physically classified based on the information on the abnormal state, and a weight is given to a physical variable related to each corresponding abnormal state.
- the abnormal state is classified as valve leakage, and a high weight is given to the flow rate of the system that is thermally hydraulically related to the system, and is used to explain the basis for determining the abnormal state.
- the selection of operating variables physically related to the abnormal state should be prepared by referring to the abnormal procedure or actual power plant operation history.
- the weighting unit 540 for each operating variable of the power plant gives weight to the physical variable related to the abnormal state among the operating variables for each failure, and finally, the abnormal state determination basis generating unit 550 provides a more accurate basis through the weighted result. tracking becomes possible.
- FIG. 6 is a diagram illustrating the use of a basis for determining an abnormal state derived by applying a neural network model according to an embodiment of the present invention.
- the abnormality determination basis 620 derived from the simulated abnormal driving simulation data 610 for a scenario selected from among the abnormal driving scenarios 600 having a plurality of scenarios related to the abnormal state is It is a power plant operation variable that can be distinguished from other abnormal conditions in the case of an abnormality. This can be utilized for verification 630 of the abnormal state determination logic being used in the abnormal state determination system. That is, it can be checked whether or not the abnormal state determination logic has been developed by using the operation variable that is changed due to the abnormal state.
- the abnormal procedure describes the operating variables that change when the abnormal condition occurs. It can be used to verify and revise the procedure so that the operator can more effectively determine the abnormal condition by verifying the symptoms regarding the abnormal procedure through the abnormal procedure-related symptom verification 640 using the abnormal condition determination basis 620. have.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Plasma & Fusion (AREA)
- High Energy & Nuclear Physics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Emergency Management (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
Description
Claims (9)
- 비정상 상태에 관한 복수의 시나리오가 저장된 비정상 운전 시나리오에서 상기 비정상 상태를 복수의 고장으로 분류하는 비정상 종류 분류부;상기 분류된 복수의 고장의 각각에 대해 비정상 상태 판단 결과에 영향을 주는 운전 변수를 도출하는 운전 변수 도출부;상기 운전 변수 중 상기 비정상 상태와 관련 있는 변수에 가중치를 부여하는 발전소 운전 변수별 가중치 부여부; 및상기 가중치가 부여된 발전소 운전 변수를 통하여 생성된 상기 비정상 상태 판단 결과에서 비정상 상태 판단 근거를 추적하는 비정상 상태 판단 근거 생성부를 포함하는 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치.
- 제1항에 있어서,상기 운전 변수 중 상기 비정상 상태와 관련 있는 변수에 가중치를 부여하는 발전소 운전 변수별 가중치 부여부는상기 비정상 상태와 관련된 계통의 물리적 상관관계를 고려하여 분류되고 상기 비정상 상태와 관련 있는 물리적 변수에 상기 가중치를 부여하는 것을 특징으로 하는신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치.
- 제1항에 있어서,상기 비정상 종류 분류부는 상기 비정상 운전 시나리오를 밸브 누설, 펌프 고장, 열교환기 고장, 냉각재 누설 중 적어도 하나를 포함하도록 분류하는 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치.
- 제1항에 있어서,상기 분류된 복수의 고장의 각각에 대해 비정상 상태 판단 결과에 영향을 주는 운전 변수를 도출하는 운전 변수 도출부에서는 상기 밸브 누설로 분류된 경우에는 해당 밸브과 관련된 계통의 유량을, 상기 펌프 고장으로 분류된 경우에는 해당 밸브와 관련된 계통의 유량 및 압력을, 상기 열교환기 고장으로 분류된 경우에는 해당 열교환기와 관련된 계통의 온도를, 상기 냉각제 누설로 분류된 경우에는 누설부위 방사선 준위를 도출하는 것을 포함하는 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치.
- 제1항에 있어서,상기 비정상 상태와 관련 있는 물리적 변수는 비정상 절차서 또는 실제 발전소 운전이력을 참조하여 작성되는 신경망 모델을 이용한 비정상 상태 판단 근거 추적장치.
- 제1항에 있어서,상기 비정상 상태 판단 근거는 상이한 비정상 상태와 구분할 수 있는 상기 운전 변수로서, 비정상 상태 판단 시스템에서 사용하는 비정상 상태 판단 논리의 검증에 사용되는 신경망 모델을 이용한 비정상 상태 판단 근거 추적장치.
- 제1항에 있어서,상기 비정상 상태 판단 근거는 비정상 절차서에서 기술하고 있는 증상을 검증하는데 사용되며, 상기 비정상 절차서는 상기 비정상 상태 발생 시 변화하는 상기 운전 변수를 기술하고 있는 비정상 상태 판단 근거 추적장치.
- 신경망 모델을 이용한 비정상 상태를 판단하는 근거를 생성하는 방법에 있어서,발전소 운전데이터와 신경망 모델을 활용하여 학습을 진행하여 상기 신경망 모델의 최후단에서 비정상 상태 판단 결과를 생성하는 단계; 및상기 비정상 상태 판단 결과를 생성하기 전의 완전결합층(Fully Connected Layer)에서 상기 비정상 상태 판단 결과에 대해 영향분석을 수행하여 상기 비정상 상태 판단 결과에 영향을 주는 변수 값들을 추출하는 단계를 포함하는신경망 모델을 이용한 비정상 운전 상태 판단 근거를 생성하는 방법.
- 제8항에 있어서,상기 비정상 상태 판단 결과에 영향을 주는 변수 값들을 추출하는 단계는 시각화 알고리즘을 적용하여 시각화된 입력 변화 데이터를 가상으로 생성하며, 상기 가상의 입력 변화 데이터를 입력으로 상기 신경망 모델의 계산을 통하여 상기 입력 변화 데이터가 상기 비정상 상태 판단 결과의 변화에 미치는 영향을 분석하여, 상기 비정상 상태 판단 결과의 변화를 도출하는데 가장 크게 기여하는 상기 입력 변화 데이터를 추출하는 단계를 포함하는 신경망 모델을 이용한 비정상 운전 상태 판단 근거를 생성하는 방법.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2023527774A JP2023548414A (ja) | 2020-11-25 | 2021-11-15 | ニューラルネットワークモデルを利用した非正常状態判断根拠追跡装置及び方法 |
US18/033,477 US20230394301A1 (en) | 2020-11-25 | 2021-11-15 | Device and method for tracking basis of abnormal state determination by using neural network model |
CN202180079289.9A CN116490933A (zh) | 2020-11-25 | 2021-11-15 | 使用神经网络模型跟踪异常状态判断依据的装置和方法 |
EP21898460.7A EP4254430A1 (en) | 2020-11-25 | 2021-11-15 | Device and method for tracking basis of abnormal state determination by using neural network model |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020200160251A KR102537723B1 (ko) | 2020-11-25 | 2020-11-25 | 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치 및 방법 |
KR10-2020-0160251 | 2020-11-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022114634A1 true WO2022114634A1 (ko) | 2022-06-02 |
Family
ID=81756116
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2021/016627 WO2022114634A1 (ko) | 2020-11-25 | 2021-11-15 | 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치 및 방법 |
Country Status (6)
Country | Link |
---|---|
US (1) | US20230394301A1 (ko) |
EP (1) | EP4254430A1 (ko) |
JP (1) | JP2023548414A (ko) |
KR (1) | KR102537723B1 (ko) |
CN (1) | CN116490933A (ko) |
WO (1) | WO2022114634A1 (ko) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102685434B1 (ko) * | 2024-03-04 | 2024-07-16 | 성균관대학교산학협력단 | 인공신경망 기반의 고장 분류 모델을 이용하여 하이브리드 전력망의 고장을 식별하는 방법 및 장치 |
WO2024209450A1 (en) * | 2024-04-24 | 2024-10-10 | Ahmadkhan Kamelia | Automated fault detection algorithm reporting and resolving issues in any type of industrial production lines based on artificial intelligence |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3087974B2 (ja) * | 1991-08-28 | 2000-09-18 | 株式会社日立製作所 | プラントの異常時運転支援方法及びその装置 |
US20070073518A1 (en) * | 2005-09-26 | 2007-03-29 | Lockheed Martin Corporation | Method and system of monitoring and prognostics |
KR101829137B1 (ko) * | 2016-08-29 | 2018-02-13 | 한국수력원자력 주식회사 | 기기 중요도와 경보 유효성 판단을 포함한 기기 이상징후 조기경보 방법 및 시스템 |
US20190164057A1 (en) | 2019-01-30 | 2019-05-30 | Intel Corporation | Mapping and quantification of influence of neural network features for explainable artificial intelligence |
US10452845B2 (en) | 2017-03-08 | 2019-10-22 | General Electric Company | Generic framework to detect cyber threats in electric power grid |
KR102062992B1 (ko) * | 2018-09-13 | 2020-01-06 | 한국수력원자력 주식회사 | 발전소 경보나 증상 및 주요 운전변수의 변화에 기반한 비정상 상태 판단 시스템 및 방법 |
KR102095653B1 (ko) | 2018-10-12 | 2020-03-31 | 한국수력원자력 주식회사 | 신경망 모델을 이용한 비정상 운전 상태 판단 장치 및 방법 |
-
2020
- 2020-11-25 KR KR1020200160251A patent/KR102537723B1/ko active IP Right Grant
-
2021
- 2021-11-15 EP EP21898460.7A patent/EP4254430A1/en active Pending
- 2021-11-15 US US18/033,477 patent/US20230394301A1/en active Pending
- 2021-11-15 CN CN202180079289.9A patent/CN116490933A/zh active Pending
- 2021-11-15 JP JP2023527774A patent/JP2023548414A/ja active Pending
- 2021-11-15 WO PCT/KR2021/016627 patent/WO2022114634A1/ko active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3087974B2 (ja) * | 1991-08-28 | 2000-09-18 | 株式会社日立製作所 | プラントの異常時運転支援方法及びその装置 |
US20070073518A1 (en) * | 2005-09-26 | 2007-03-29 | Lockheed Martin Corporation | Method and system of monitoring and prognostics |
KR101829137B1 (ko) * | 2016-08-29 | 2018-02-13 | 한국수력원자력 주식회사 | 기기 중요도와 경보 유효성 판단을 포함한 기기 이상징후 조기경보 방법 및 시스템 |
US10452845B2 (en) | 2017-03-08 | 2019-10-22 | General Electric Company | Generic framework to detect cyber threats in electric power grid |
KR102062992B1 (ko) * | 2018-09-13 | 2020-01-06 | 한국수력원자력 주식회사 | 발전소 경보나 증상 및 주요 운전변수의 변화에 기반한 비정상 상태 판단 시스템 및 방법 |
KR102095653B1 (ko) | 2018-10-12 | 2020-03-31 | 한국수력원자력 주식회사 | 신경망 모델을 이용한 비정상 운전 상태 판단 장치 및 방법 |
US20190164057A1 (en) | 2019-01-30 | 2019-05-30 | Intel Corporation | Mapping and quantification of influence of neural network features for explainable artificial intelligence |
Also Published As
Publication number | Publication date |
---|---|
KR102537723B1 (ko) | 2023-05-26 |
EP4254430A1 (en) | 2023-10-04 |
CN116490933A (zh) | 2023-07-25 |
JP2023548414A (ja) | 2023-11-16 |
US20230394301A1 (en) | 2023-12-07 |
KR20220072533A (ko) | 2022-06-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020076081A1 (ko) | 신경망 모델을 이용한 비정상 운전 상태 판단 장치 및 방법 | |
WO2022114634A1 (ko) | 신경망 모델을 이용한 비정상 상태 판단 근거 추적 장치 및 방법 | |
CN1991652B (zh) | 可编程逻辑控制器和中央单元之间通信的通信方法及自动远程监控和诊断系统 | |
WO2017191872A1 (ko) | 플랜트 이상 감지 방법 및 시스템 | |
CN103149475B (zh) | 电力设备故障诊断方法及系统 | |
CN103178615A (zh) | 电力设备故障监控方法及其系统 | |
CN108508852A (zh) | 隔离管理系统与隔离管理方法 | |
WO2019083126A1 (ko) | 원자력 발전소 계측제어시스템의 검증방법과 이를 위한 검증장치 | |
KR101865666B1 (ko) | 원자로노심보호계통 소프트웨어 검증용 시뮬레이션 장치 및 시스템 | |
JP6812312B2 (ja) | プラント支援評価システム及びプラント支援評価方法 | |
US10580541B2 (en) | Apparatus and system for simulating maintenance of reactor core protection system | |
WO2021172723A1 (ko) | 원전의 지능형 상태감시 방법 및 시스템 | |
CN117783769B (zh) | 基于可视平台的配电网络故障定位方法、系统、设备及存储介质 | |
Kościelny et al. | Towards a unified approach to detection of faults and cyber-attacks in industrial installations | |
KR102136956B1 (ko) | 입출력 핫스와퍼블 배선 인터페이스 제공 방법 및 그를 위한 장치 및 시스템 | |
KR20230102431A (ko) | 인공지능 기반의 오일가스 플랜트 설비 고장 예측 및 진단시스템 | |
KR20220084709A (ko) | 가상화 기술을 적용한 발전소의 모의 시스템 | |
WO2023128320A1 (ko) | 인공지능 검증 시스템 및 방법 | |
CN104460337B (zh) | 一种基于修正β因子的控制系统共因失效分析方法 | |
WO2022146020A1 (ko) | 회전기계의 결함을 종합적으로 진단하는 방법 및 시스템 | |
Bushby et al. | The virtual cybernetic building testbed–a building emulator | |
WO2020004998A1 (ko) | 발전소 고장 예측 및 진단시스템의 화면표시방법 | |
WO2024205009A1 (ko) | 원자력 발전소의 사고해석 자동화 시스템 및 자동화 방법 | |
WO2024053913A1 (ko) | 발전소 필드값 기반의 발전소 상태를 실시간 추적하는 시뮬레이션 시스템 및 방법 | |
CN116931546A (zh) | 一种基于虚实结合技术的dcs可靠性验证方法及系统 |
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: 21898460 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023527774 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180079289.9 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2021898460 Country of ref document: EP Effective date: 20230626 |