Deep learning algorithms provide plausible benefits for efficient prediction and analysis of nuclear reactor safety phenomena. However, research works that discuss the critical challenges with deep learning models from the reactor safety perspective are limited. This article presents the state-of-the-art in deep learning application in nuclear reactor safety analysis, and the inherent limitations in deep learning models. In addition, critical issues such as deep learning model explainability, sensitivity and uncertainty constraints, model reliability, and trustworthiness are discussed from the nuclear safety perspective, and robust solutions to the identified issues are also presented. As a major contribution, a deep feedforward neural network is developed as a surrogate model to predict turbulent eddy viscosity in Reynolds-averaged Navier–Stokes (RANS) simulation. Further, the deep feedforward neural network performance is compared with the conventional Spalart Allmaras closure model in the RANS turbulence closure simulation. In addition, the Shapely Additive Explanation (SHAP) and the local interpretable model-agnostic explanations (LIME) APIs are introduced to explain the deep feedforward neural network predictions. Finally, exciting research opportunities to optimize deep learning-based reactor safety analysis are presented. 

Deep learning for safety assessment of nuclear power reactors: Reliability, explainability, and research opportunities

Advances in nuclear power system design and fault-based condition monitoring towards safety of nuclear-powered ships

Accurate estimation and prediction of the State-of-Health (SOH) and Remaining Useful Life (RUL) are fundamental to optimal maintenance strategies formulation for Prognostics and Health Management (PHM) of degraded equipment. However, the performance assessment of health state prognostics and RUL prediction is strongly dependent on the errors and uncertainties in physical measurements, and heterogeneous degradation of equipment in time-varying operating conditions. The objective of the paper is to provide a hybrid prognostic framework that integrates a two-phases clustering scheme and a PF-LSTM learning algorithm based on Particle Filter (PF) and Long Short-Term Memory (LSTM) networks for dynamic classification of SOH and long-term RUL prediction in the absence of future observations. The proposed generic hybrid PF-LSTM prognostic approach is demonstrated and compared with other adaptive learning and machine learning methods such as Unscented Particle Filter (UPF) and Radial Basis Function Network (RBFN) on the degradation modeling and RUL prediction for Lithium-ion batteries. The comparison results show that robust prediction performance can be obtained by the hybrid PF-LSTM prognostic approach with accurate characterization of equipment degradation states based on the integrated subtractive-fuzzy clustering analysis. The more accuracy on prognostic estimations in Probability Density Function (PDF) of prior and posterior distributions of battery capacity and RUL that are achieved by particle filtering can gain extensive insights to predictive maintenance action guide. 

An improved generic hybrid prognostic method for RUL prediction based on PF-LSTM learning

Industry 4.0 is strongly intertwined with big data streaming flows from intelligent sensors and machinery installed in industrial facilities. Failures can disrupt production and lead the supply chain ecosystem to malfunction. Maintenance strategies are necessary to safeguard the continuous operation of production lines, minimize supply chain disruptions, and improve sustainability indicators. Within the context of smart manufacturing, predictive maintenance (PdM) approaches could decrease downtimes, reduce operational costs, and increase productivity, improving system performance and decision-making. The overarching aim of this research is to systematically review state-of-the-art predictive maintenance applications across diverse manufacturing sectors to provide customized insights from academic and operational perspectives, summarized into a comparative decision support map. The study classifies predictive maintenance solutions based on prevailing methodologies, input features, predicted variables, applied algorithms, evaluation metrics, and state-of-the-art software tools per industry sector. The outcomes highlight that data-driven predictive maintenance constitutes a cutting-edge solution with a growing interest in modern manufacturing. Moreover, this research provides insights into the technology readiness of each industrial sector, covering modern areas for PdM implementation, while raising the extant challenges. The proposed multi-sector framework is expected to act as a guiding light for researchers and practitioners towards the development of PdM driven applications in data driven industries. 

Predictive maintenance in Industry 4.0: A systematic multi-sector mapping

Deep learning–based nuclear intelligent fault detection and diagnosis (FDD) methods have been widely developed and have achieved very competitive results with the progress of artificial intelligence technology. However, the pretrained model for diagnosis tasks is hard in achieving good performance when the reactor operation conditions are updated. On the other hand, retraining the model for a new data set will waste computing resources. This article proposes an FDD method for cross-condition and cross-facility tasks based on the optimized transferable convolutional neural network (CNN) model. First, by using the pretrained model's prior knowledge, the model's diagnosis performance to be transferred for source domain data sets is improved. Second, a model-based transfer learning strategy is adopted to freeze the feature extraction layer in a part of the training model. Third, the training data in target domain data sets are used to optimize the model layer by layer to find the optimization model with the transferred layer. Finally, the proposed comprehensive simulation platform provides source and target cross-condition and cross-facility data sets to support case studies. The designed model utilizes the strong nonlinear feature extraction performance of a deep network and applies the prior knowledge of pretrained models to improve the accuracy and timeliness of training. The results show that the proposed method is superior to achieving good generalization performance at less training epoch than the retraining benchmark deep CNN model.

https://www.frontiersin.org/articles/10.3389/fenrg.2022.823395/pdf

Model-Based Deep Transfer Learning Method to Fault Detection and Diagnosis in Nuclear Power Plants

In a carbon-constrained world, future uses of nuclear power technologies can contribute to climate change mitigation as the installed electricity generating capacity and range of applications could be much greater and more diverse than with the current plants. To preserve the nuclear industry competitiveness in the global energy market, prognostics and health management (PHM) of plant assets is expected to be important for supporting and sustaining improvements in the economics associated with operating nuclear power plants (NPPs) while maintaining their high availability. Of interest are long-term operation of the legacy fleet to 80 years through subsequent license renewals and economic operation of new builds of either light water reactors or advanced reactor designs. Recent advances in data-driven analysis methods—largely represented by those in artificial intelligence and machine learning—have enhanced applications ranging from robust anomaly detection to automated control and autonomous operation of complex systems. The NPP equipment PHM is one area where the application of these algorithmic advances can significantly improve the ability to perform asset management. This paper provides an updated method-centric review of the full PHM suite in NPPs focusing on data-driven methods and advances since the last major survey article was published in 2015. The main approaches and the state of practice are described, including those for the tasks of data acquisition, condition monitoring, diagnostics, prognostics, and planning and decision-making. Research advances in non-nuclear power applications are also included to assess findings that may be applicable to the nuclear industry, along with the opportunities and challenges when adapting these developments to NPPs. Finally, this paper identifies key research needs in regard to data availability and quality, verification and validation, and uncertainty quantification.

/pdf/prognostics-and-health-management-in-nuclear-power-plants-an-4jkdy10w3k.pdf

Prognostics and Health Management in Nuclear Power Plants: An Updated Method-Centric Review With Special Focus on Data-Driven Methods

Decarbonization efforts in the United States has resulted in the increasing interest in flexible operation of nuclear power plants or thermal storage-based hybrid energy systems integrated with the power grid. Consequently, the need for rapid and effective modeling and simulation of such integrated systems is becoming critical. This requires modeling a approach that offers capabilities for multi-domain simulation that allows to model coupled phenomena accurately. This work focused on developing a proof of concept of such multi-domain models by designing a nominal model of a nuclear power plant balance of plant (BOP) system using the Modelica libraries. To model the thermofluidic domain, three libraries are evaluated, namely, ORNL’s TRANSFORM, Modelon’s ThermalPower, and Casella’s ThermoPower. Next, the thermofluidic model is coupled with an electrical grid model built using the OpenIPSL. To understand the tradeoffs of each library, the response of each different model to power transients in the nuclear power plant was analyzed. Simulation results and modeling methods were compared. The BOP model in TRANSFORM was found to be stiff and may require more detailed component models, e.g. condenser and feed water heater, to model the BOP more realistically and rapidly.

Comparing Thermal Library Modeling Suites for Integrated Modeling of Nuclear Power Plant and Power Grids

This paper presents a prognostic model for sodium-cooled fast reactor (SFR) steam generators (SGs). The purpose of the model is to estimate the remaining useful life of SFR SGs and thus to support the decision-making of autonomous control. SFR SGs are of great interest for plant integrity due to their harsh operation environment. They operate at higher temperatures and higher coolant-to-steam pressure differences than those of current light water reactors (LWR). The severity of the SFR SG failure consequences, which include water-sodium contact, is another reason. Understanding its failure mechanisms is important for the development of its prognostic model. Based on our literature review and physics-based analysis, we concluded that creep would be a dominant degradation mode of SFR SGs due to SFR's high-temperature and high-pressure environment. Thus, creep is the focus of the prognostic model. Various other failure modes were also investigated in this study. The mechanical fatigue due to flow-induced vibration is usually observed in early developmental phases and would not be an important issue during normal operation. The thermal fatigue due to thermal stripping is occasionally observed in other components in SFRs but does not affect SG integrity. Pure water stress corrosive crack, fretting, and so on are commonly observed in LWR SGs but are significantly less important in SFR SGs because of the high temperatures, high pressure differences, and chemical properties of liquid sodium. Based on these investigations, a prognostic model focusing on creep failures was developed. It estimates the failure probability profile by sampling the Larson-Miller parameter (LMP) and temperature and associated uncertainties through Monte Carlo methods. Two case studies were presented. The first one demonstrated the model's capability to calculate the failure probability for a new specimen within a given time. If the tube is working under temperatures of 500C ± 3 and pressures of 15.2 MPa, which leads to LMP of 20,100 ± 50, the failure probabilities within 50, 70, and 90 years are approximately 0.1 %, 3.2 %, and 17.5 %. The second demonstrated how adjusting the workload can help to protect the integrity of the component. For an old tube reaching 99 % of its lifespan, continuing to run at temperatures of 500C ± 3 and LMP of 20,100 ± 50 leads to a failure probability of about 29.0 % within a year. If the temperature is reduced by 5C, the failure probability can be reduced to 2.5 %, and, if the pressure is also reduced such that the LMP is increased by 100, the failure probability can further be reduced to 0.12 %. 

Prognostic model and failure mechanisms of steam generators in Sodium-Cooled fast reactors

This work documents the development and testing of a multi-domain Rankine cycle-based balance of plant and power grid model developed using the Modelica language. Steady-state and transient behavior of the coupled systems is analyzed. Results indicate that simplified models of either steam turbines or generators used in traditional single-domain simulations do not sufficiently capture all dominant system dynamics. Initiation of power transients in the power plant clearly impacted grid operating current, and faults in the power grid were shown to cause variations in torque experienced by the shaft of the steam turbine. Therefore, multi-domain models such as this are necessary for analysis in the power generation industries in order to capture the interconnected behavior of complex energy systems.

Multi-domain Modeling of a Steam Power Plant with Power Grid

To address the challenge of interpreting the decisions made by machine learning-based control algorithms, the Integrated Artificial Reasoning Framework (IARF) introduced in Part I of this work was developed. Further implementing this framework within a Markov Decision Process (MDP) yielded a decision-making system which produces optimal decisions whose logic can be easily traced and understood. The IARF with MDP formulated in Part I of this study was applied to two case studies of relevance to the nuclear power community: 1) operational decision-making while experiencing the degradation of key components, and 2) optimal control of a nuclear power and hydrogen co-generation facility. This work details the steps of implementing and analyzing these two studies, beginning with the creation of simulation models using the Modelica programming language, covering the formulation of the MDP using the IARF, and concluding with the execution and analysis of the scenarios. Technical challenges related to implementation, such as maintaining physically understandable traceability and efficiently producing large amount of data for calculating state transition probabilities, are addressed and their resolutions are suggested. Emphasis is placed on the benefits of relying on a replaceable modeling framework for simulating system behavior. Results indicate that optimal operational trajectories were successfully found by the MDP, and through utilization of the IARF, these results are easily interpretable and understandable to any user.

Decision-making based on Markov decision process in integrated artificial reasoning framework—Part 2: Applications

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Kyle Warns

Author Tools

Chat about Author