• 한국산업정보학회논문지
• /
• 제25권6호
• /
• pp.1-13
• /
• 2020

본 논문에서는 원전 비상 상황 발생 시 다수의 신호 오류가 발생했을 때 어떤 신호에 오류가 발생했는지를 추정하는 신호 오류 식별 (Fault identification) 방법론을 개발하였다. 변분 오토인 코더 (Variational autoencoder; VAE) 기반 모델은 기존의 이상 탐지 방법론과 같이 정상 신호 데이터만을 이용하여 훈련이 진행되며, 이후 각 신호에 대한 복원 오차 (Reconstruction error)와 복원 오차를 입력의 특정 부분으로 미분한 값을 이용하여 어떤 부분에 오류가 포함되어 있는지를 예측한다. 데이터 취득을 위하여 시뮬레이션을 수행하였으며, 일련의 실험으로부터 제시한 신호 오류 식별 방법이 적절한 오차 범위 내에서 오류가 발생한 신호를 특정할 수 있음을 확인하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 B
• /
• pp.868-870
• /
• 1999

In this paper we obtain a discrete mathmatical model of a Boiler control system from expermental data, we find appropriate input signal and parameter estimation algorithm for identification of the Boiler control system in power plant. Under these conditions experimental data are collected from real system and parameters are estimated by the Recursive Least Square algorithm. The computer simulation results show the parameter estimation algorithm for identification and the effectiveness of controller design of the Boiler control system.

• 한국환경과학회지
• /
• 제26권12호
• /
• pp.1333-1339
• /
• 2017

Coal-fired power plants emit various Particulate Matter(PM) at coal storage pile and ash landfill as well as the stack, and affect the surrounding environment. Field Emission Scanning Electron Microscopy and Energy Dispersive X-ray analyzer(FE-SEM/EDX) were used to develop identification factor and the physico-chemical analysis of PM emitted from a power plant. In this study, three samples of pulverized coal, bottom ash, and fly ash were analyzed. The pulverized coal was spherical particles in shape and the chemical composition of C-O-Si-Al and C/Si and C/Al ratios were 200~300 on average. The bottom ash was spherical or non-spherical particles in shape, chemical composition was O-C-Si-Al-Fe-Ca and C/Si and C/Al ratios were $4.3{\pm}4.6$ and $8.8{\pm}10.0$. The fly ash was spherical particles in shape, chemical composition was O-Si-Ai-C-Fe-Ca and C/Si and C/Al ratios were $0.5{\pm}0.2$ and $0.8{\pm}0.5$.

• Nuclear Engineering and Technology
• /
• 제53권9호
• /
• pp.2888-2898
• /
• 2021

Vital area identification (VAI) is an essential procedure for the design of physical protection systems (PPSs) for nuclear power plants (NPPs). The purpose of PPS design is to protect vital areas. VAI has been improved continuously to overcome the shortcomings of previous VAI generations. In first-generation VAI, a sabotage fault tree was developed directly without reusing probabilistic safety assessment (PSA) results or information. In second-generation VAI, VAI model was constructed from all PSA event trees and fault trees. While in third-generation VAI, it was developed from the simplified PSA event trees and fault trees. While VAIs have been performed for NPPs in full-power operations, VAI for NPPs in low-power and shutdown (LPSD) operations has not been studied and performed, even though NPPs in LPSD operations are very vulnerable to sabotage due to the very crowded nature of NPP maintenance. This study is the first to research and apply VAI to LPSD operation of NPP. Here, the third-generation VAI method for full-power operation of NPP was adapted to the VAI of LPSD operation. In this study, LPSD VAI for a few plant operational states (POSs) was performed. Furthermore, the operation strategy of vital areas for both full-power and LPSD operations was discussed. The LPSD VAI method discussed in this paper can be easily applied to all POSs. The method and insights in this study can be important for future LPSD VAI that reflects various LPSD operational states. Regulatory bodies and electric utilities can take advantage of this LPSD VAI method.

• Nuclear Engineering and Technology
• /
• 제53권6호
• /
• pp.2038-2045
• /
• 2021

The main objective of this work is to propose a reliable routine standard operation procedures (SOP) for structural health monitoring and diagnosis of nuclear power plants (NPPs). At present, NPPs have monitoring systems that can be used to obtain the quantitative health record of containment (CTMT) buildings through system identification technology. However, because the measurement signals are often interfered with by noise, the identification results may introduce erroneous conclusions if the measured data is directly adopted. Therefore, this paper recommends the SOP for signal screening and the required identification procedures to identify the dynamic characteristics of the CTMT of NPPs. In the SOP, three recommend methods are proposed including the Recursive Least Squares (RLS), the Observer Kalman Filter Identification/Eigensystem Realization Algorithm (OKID/ERA), and the Frequency Response Function (FRF). The identification results can be verified by comparing the results of different methods. Finally, a preliminary CTMT healthy record can be established based on the limited number of earthquake records. It can be served as the quantitative reference to expedite the restart procedure. If the fundamental frequency of the CTMT drops significantly after the Operating Basis Earthquake and Safe Shutdown Earthquake (OBE/SSE), it means that the restart actions suggested by the regulatory guide should be taken in place immediately.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1008-1011
• /
• 1996

Most control systems of power plants are using classical PID controllers for their process control. In order to get the desired control performances, the correct tuning of PID controllers is very important. Sometimes, it is necessary to retune PID controllers after the change of system operating condition and system design change, etc. Commercial auto-tuning controllers such as relay feedback controller can be used for this purpose. However, using these controllers to the safety-critical systems of nuclear power plants may be cause of unsafe operation, because they are using test signals for tuning. A new system identification auto-tuning method without using test signal has been developed in this paper. This method uses process input/output signals for system identification of unknown control process. From the model information of control process which was obtained from system identification approach, the optimal PID parameters can be calculated. The method can be used in the safety-critical systems because it is not using test signals during system modeling process.

• 동력기계공학회지
• /
• 제3권2호
• /
• pp.34-41
• /
• 1999

This paper identifies a linear combustion model of Refuse Incineration Plant(RIP) which characterizes its combustion dynamics, where the proposed model has thirteen-inputs and one-output. The structure of the RIP model is given as an ARX model which obtained from the theoretical analysis. And then, some unknown model parameters are decided from experimental input-output data sets, using system identification algorithm based on Instrumental Variables(IV) method. In result, it is shown that the proposed model well approximates the input-output combustion characteristics riven by experimental data sets.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2001년도 추계학술대회논문집 II
• /
• pp.897-902
• /
• 2001

This paper describes the evaluation of noise influence for residental and boundary areas in 75MW thermal power plant. It includes the measurements of noise level around the boundary area of the plant, identification of noise propagating path, and discussion on the measures. Noise assessments are carried out based on the ISO 3744, ISO 9613-1 and ISO 9613-2 to predict the noise distribution of specific locations from the noise sources such as power transformers, flash vent-pipe, I.D.fan, and stack. It is identified the vent-pipe of flash tank in thermal power plant as the root cause of the environmental noise during the plant commissioning process.

• 한국건설관리학회논문집
• /
• 제18권1호
• /
• pp.48-57
• /
• 2017

1971년 우리나라 최초의 원자력발전소인 고리 1호기가 착공되어 2015년 현재 24기가 가동 중에 있으며 4기가 건설 중으로 약 40여 년간의 원자력발전소 건설 역사 동안 원전 건설의 성공적인 시공을 위한 CSFs 도출에 관한 연구는 아주 미흡한 실정이다. 본 연구에서는 시공단계에서 원전 건설의 성공적인 시공을 위한 CSFs를 도출하기 위하여 기존 연구에서 제안된 원전 및 일반 건설의 성공요인과 고찰한 원전 건설의 시공요인을 바탕으로 원전 전문가와 포커스 그룹 인터뷰를 실시하여 원전 건설의 성공적인 시공을 위한 CSFs를 도출하였다. 포커스 그룹 인터뷰 결과의 검증을 위하여 현재 원전 건설 프로젝트에 참여하고 있는6 14명의 실무 전문가들의 설문조사를 통해 도출한 CSFs의 유용성을 검증하였다. 설문 결과 분석을 통해 합리적인 공사비산정기준 수립, 원전 건설경험 인력 관리시스템 구축 및 활용, 경쟁력 있는 우수 협력업체 발굴 및 육성 방안 마련 및 검사 절차 개선 등으로 작업 대기시간의 최소화 방안 등에 대해 우선 검토 및 개선이 이루어져야 함을 도출 하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1449-1454
• /
• 2003

For many industrial processes there are good static models used for process design and steady state operation. By using system identification techniques, it is possible to obtain black-box models with reasonable complexity that describe the system well in specific operating conditions [1]. But black-box models using inductive modeling(IM) is not suitable for model based control because they are only valid for specific operating conditions. Thus we need to use deductive modeling(DM) for a wide operating range. Furthermore, deductive modeling is several merits: First, the model is possible to be modularized. Second, we can increase and decrease the model complexity. Finally, we are able to use model for plant design. Power plant must be able to operate well at dramatic load change and consider safety and efficiency. This paper proposes a simplified nonlinear model of an industrial boiler, one of component parts of a power plant, by DM method and applies optimal control to the model.