MODELLING THE DYNAMICS OF THE LEAD BISMUTH EUTECTIC EXPERIMENTAL ACCELERATOR DRIVEN SYSTEM BY AN INFINITE IMPULSE RESPONSE LOCALLY RECURRENT NEURAL NETWORK

ENRICO ZIO,NICOLA PEDRONI,MATTEO BROGGI,LUCIA ROXANA GOLEA 한국원자력학회 2009 Nuclear Engineering and Technology Vol.41 No.10

In this paper, an infinite impulse response locally recurrent neural network (IIR-LRNN) is employed for modelling the dynamics of the Lead Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS). The network is trained by recursive back-propagation (RBP) and its ability in estimating transients is tested under various conditions. The results demonstrate the robustness of the locally recurrent scheme in the reconstruction of complex nonlinear dynamic relationships.

RISK-INFORMED REGULATION: HANDLING UNCERTAINTY FOR A RATIONAL MANAGEMENT OF SAFETY

ENRICO ZIO 한국원자력학회 2008 Nuclear Engineering and Technology Vol.40 No.5

A risk-informed regulatory approach implies that risk insights be used as supplement of deterministic information for safety decision-making purposes. In this view, the use of risk assessment techniques is expected to lead to improved safety and a more rational allocation of the limited resources available. On the other hand, it is recognized that uncertainties affect both the deterministic safety analyses and the risk assessments. In order for the risk-informed decision making process to be effective, the adequate representation and treatment of such uncertainties is mandatory. In this paper, the risk-informed regulatory framework is considered under the focus of the uncertainty issue. Traditionally, probability theory has provided the language and mathematics for the representation and treatment of uncertainty. More recently, other mathematical structures have been introduced. In particular, the Dempster-Shafer theory of evidence is here illustrated as a generalized framework encompassing probability theory and possibility theory. The special case of probability theory is only addressed as term of comparison, given that it is a well known subject. On the other hand, the special case of possibility theory is amply illustrated. An example of the combination of probability and possibility for treating the uncertainty in the parameters of an event tree is illustrated.

MODELLING THE DYNAMICS OF THE LEAD BISMUTH EUTECTIC EXPERIMENTAL ACCELERATOR DRIVEN SYSTEM BY AN INFINITE IMPULSE RESPONSE LOCALLY RECURRENT NEURAL NETWORK

Zio, Enrico,Pedroni, Nicola,Broggi, Matteo,Golea, Lucia Roxana Korean Nuclear Society 2009 Nuclear Engineering and Technology Vol.41 No.10

In this paper, an infinite impulse response locally recurrent neural network (IIR-LRNN) is employed for modelling the dynamics of the Lead Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS). The network is trained by recursive back-propagation (RBP) and its ability in estimating transients is tested under various conditions. The results demonstrate the robustness of the locally recurrent scheme in the reconstruction of complex nonlinear dynamic relationships.

RISK-INFORMED REGULATION: HANDLING UNCERTAINTY FOR A RATIONAL MANAGEMENT OF SAFETY

Zio, Enrico Korean Nuclear Society 2008 Nuclear Engineering and Technology Vol.40 No.5

A risk-informed regulatory approach implies that risk insights be used as supplement of deterministic information for safety decision-making purposes. In this view, the use of risk assessment techniques is expected to lead to improved safety and a more rational allocation of the limited resources available. On the other hand, it is recognized that uncertainties affect both the deterministic safety analyses and the risk assessments. In order for the risk-informed decision making process to be effective, the adequate representation and treatment of such uncertainties is mandatory. In this paper, the risk-informed regulatory framework is considered under the focus of the uncertainty issue. Traditionally, probability theory has provided the language and mathematics for the representation and treatment of uncertainty. More recently, other mathematical structures have been introduced. In particular, the Dempster-Shafer theory of evidence is here illustrated as a generalized framework encompassing probability theory and possibility theory. The special case of probability theory is only addressed as term of comparison, given that it is a well known subject. On the other hand, the special case of possibility theory is amply illustrated. An example of the combination of probability and possibility for treating the uncertainty in the parameters of an event tree is illustrated.

Risk-informed approach to the safety improvement of the reactor protection system of the AGN-201K research reactor

Ahmed, Ibrahim,Zio, Enrico,Heo, Gyunyoung Korean Nuclear Society 2020 Nuclear Engineering and Technology Vol.52 No.4

Periodic safety reviews (PSRs) are conducted on operating nuclear power plants (NPPs) and have been mandated also for research reactors in Korea, in response to the Fukushima accident. One safety review tool, the probabilistic safety assessment (PSA), aims to identify weaknesses in the design and operation of the research reactor, and to evaluate and compare possible safety improvements. However, the PSA for research reactors is difficult due to scarce data availability. An important element in the analysis of research reactors is the reactor protection system (RPS), with its functionality and importance. In this view, we consider that of the AGN-201K, a zero-power reactor without forced decay heat removal systems, to demonstrate a risk-informed safety improvement study. By incorporating risk- and safety-significance importance measures, and sensitivity and uncertainty analyses, the proposed method identifies critical components in the RPS reliability model, systematically proposes potential safety improvements and ranks them to assist in the decision-making process.

Application of particle fi ltering for prognostics with measurement uncertainty in nuclear power plants

김기범,김현민,Enrico Zio,허균영 한국원자력학회 2018 Nuclear Engineering and Technology Vol.50 No.8

For nuclear power plants (NPPs) to have long lifetimes, ageing is a major issue. Currently, ageing managementfor NPP systems is based on correlations built from generic experimental data. However, eachsystem has its own characteristics, operational history, and environment. To account for this, it is possibleto resort to prognostics that predicts the future state and time to failure (TTF) of the target system byupdating the generic correlation with specific information of the target system. In this paper, we presentan application of particle filtering for the prediction of degradation in steam generator tubes. With a casestudy, we also show how the prediction results vary depending on the uncertainty of the measurementdata.

On-line process monitoring during transient operations using weighted distance Auto Associative Bilateral Kernel Regression

Ahmed, Ibrahim,Heo, Gyunyoung,Zio, Enrico Elsevier 2019 ISA transactions Vol.92 No.-

<P><B>Abstract</B></P> <P>In this paper, a new data-driven auto associative bilateral kernel regression (AABKR) method based on weighted distance is proposed for the on-line monitoring of transient process operations. A bilateral approach to the kernel regression formulates a representative model that uses both the spatial and temporal information in the data, and a new weighted-distance algorithm captures temporal information. Moreover, an adaptive approach is proposed to dynamically compensate for faulty process inputs in the bilateral kernel evaluations, providing a robust model with little spillover. The proposed weighted-distance AABKR is first implemented using numerical process examples and then applied to the transient start-up operation of a nuclear power plant. Monte Carlo simulation results are provided by randomly assigning fault sensors and fault magnitudes. The results demonstrate the feasibility and efficiency of the proposed method.</P> <P><B>Highlights</B></P> <P> <UL> <LI> A method for on-line monitoring in process transient operations is proposed. </LI> <LI> The method is a weighted-distance Auto Associative Bilateral Kernel Regression. </LI> <LI> The proposed method captured both the spatial and temporal information in the data. </LI> <LI> Applications results demonstrate the efficiency of the proposed method. </LI> </UL> </P>

Physics informed neural networks for surrogate modeling of accidental scenarios in nuclear power plants

Antonello Federico,Buongiorno Jacopo,Zio Enrico 한국원자력학회 2023 Nuclear Engineering and Technology Vol.55 No.9

Licensing the next-generation of nuclear reactor designs requires extensive use of Modeling and Simulation (M&S) to investigate system response to many operational conditions, identify possible accidental scenarios and predict their evolution to undesirable consequences that are to be prevented or mitigated via the deployment of adequate safety barriers. Deep Learning (DL) and Artificial Intelligence (AI) can support M&S computationally by providing surrogates of the complex multi-physics high-fidelity models used for design. However, DL and AI are, generally, low-fidelity ‘black-box’ models that do not assure any structure based on physical laws and constraints, and may, thus, lack interpretability and accuracy of the results. This poses limitations on their credibility and doubts about their adoption for the safety assessment and licensing of novel reactor designs. In this regard, Physics Informed Neural Networks (PINNs) are receiving growing attention for their ability to integrate fundamental physics laws and domain knowledge in the neural networks, thus assuring credible generalization capabilities and credible predictions. This paper presents the use of PINNs as surrogate models for accidental scenarios simulation in Nuclear Power Plants (NPPs). A case study of a Loss of Heat Sink (LOHS) accidental scenario in a Nuclear Battery (NB), a unique class of transportable, plug-and-play microreactors, is considered. A PINN is developed and compared with a Deep Neural Network (DNN). The results show the advantages of PINNs in providing accurate solutions, avoiding overfitting, underfitting and intrinsically ensuring physics-consistent results

Cyber attack taxonomy for digital environment in nuclear power plants

Kim, Seungmin,Heo, Gyunyoung,Zio, Enrico,Shin, Jinsoo,Song, Jae-gu Korean Nuclear Society 2020 Nuclear Engineering and Technology Vol.52 No.5

With the development of digital instrumentation and control (I&C) devices, cyber security at nuclear power plants (NPPs) has become a hot issue. The Stuxnet, which destroyed Iran's uranium enrichment facility in 2010, suggests that NPPs could even lead to an accident involving the release of radioactive materials cyber-attacks. However, cyber security research on industrial control systems (ICSs) and supervisory control and data acquisition (SCADA) systems is relatively inadequate compared to information technology (IT) and further it is difficult to study cyber-attack taxonomy for NPPs considering the characteristics of ICSs. The advanced research of cyber-attack taxonomy does not reflect the architectural and inherent characteristics of NPPs and lacks a systematic countermeasure strategy. Therefore, it is necessary to more systematically check the consistency of operators and regulators related to cyber security, as in regulatory guide 5.71 (RG.5.71) and regulatory standard 015 (RS.015). For this reason, this paper attempts to suggest a template for cyber-attack taxonomy based on the characteristics of NPPs and exemplifies a specific cyber-attack case in the template. In addition, this paper proposes a systematic countermeasure strategy by matching the countermeasure with critical digital assets (CDAs). The cyber-attack cases investigated using the proposed cyber-attack taxonomy can be used as data for evaluation and validation of cyber security conformance for digital devices to be applied, and as effective prevention and mitigation for cyber-attacks of NPPs.

A Siamese hybrid neural network framework for few-shot fault diagnosis of fixed-wing unmanned aerial vehicles

Li Shaobo,Li Chuanjiang,Zhang Ansi,Yang Lei,Zio Enrico,Pecht Michael,Gryllias Konstantinos 한국CDE학회 2022 Journal of computational design and engineering Vol.9 No.4

As fixed-wing unmanned aerial vehicles (FW-UAVs) are used for diverse civil and scientific missions, failure incidents are on the rise. Recent rapid developments in deep learning (DL) techniques offer advanced solutions for fault diagnosis of unmanned aerial vehicles. However, most existing DL-based diagnostic models only perform well when trained on massive amounts of labeled data, which are challenging to collect due to the complexity of the FW-UAVs systems and service environments. To address these issues, this paper presents a novel framework, Siamese hybrid neural network (SHNN), to achieve few-shot fault diagnosis of FW-UAVs in an intelligent manner. “State map” strategy is firstly proposed to transform raw flight data into similar and dissimilar sample pairs as input. The proposed SHNN framework consists of two identical networks that share weights with each other, and each subnetwork is designed with a hybrid one-dimensional conventional neural network and long short-term memory model as feature encoder, whose generated feature embedding is used to measure the similarity of input pairs via a distance function in the metric space. In comprehensive experiments on a real flight dataset of an FW-UAV, the SHNN framework achieves competitive results compared to other models, demonstrating its effectiveness in both binary and multi-class few-shot fault diagnosis.