• Journal of the Korean Society of Safety
• /
• v.29 no.5
• /
• pp.160-165
• /
• 2014

The realistic cases causing severe core damage should be analyzed and arranged systematically for preparing an accident management of the specific nuclear power plant. The objective of this paper is to establish basic technical information for reactor safety and reactor building integrity management strategies in CANDU reactor severe accident. For the development of severe accident management strategies, plant specific features and behaviors must be studied by detailed analysis works. This analysis scope will serve to cover overall methods and analyzing results to understand the reactor building integrity status in the most likely severe accident sequences that could occur at CANDU reactor. Also analysis results could help prevent or mitigate severe accidents for the identification of any plant specific vulnerabilities to severe accidents using the probabilistic safety assessment (PSA) quantified results.

• Nuclear Engineering and Technology
• /
• v.41 no.7
• /
• pp.921-928
• /
• 2009

As part of an evaluation for an in-vessel severe accident management strategy, a coolant injection into the reactor vessel under depressurization of the reactor coolant system (RCS) has been evaluated in detail using the SCDAP/RELAP5 computer code. A high-pressure sequence of a small break loss of coolant accident (SBLOCA) has been analyzed in the Optimized Power Reactor (OPR) 1000. The SCDAP/RELAP5 results have shown that safety injection timing and capacity with RCS depressurization timing and capacity are very effective on the reactor vessel failure during a severe accident. Only one train operation of the high pressure safety injection (HPSI) for 30,000 seconds with RCS depressurization prevents failure of the reactor vessel. In this case, the operation of only the low pressure safety injection (LPSI) without a HPSI does not prevent failure of the reactor vessel.

• Nuclear Engineering and Technology
• /
• v.38 no.8
• /
• pp.697-732
• /
• 2006

In this paper, a historical review of the developments in the safety of LWR power plants is presented. The paper reviews the developments prior to the TMI-2 accident, i.e. the concept of the defense in depth, the design basis, the large LOCA technical controversies and the LWR safety research programs. The TMI-2 accident, which became a turning point in the history of the development of nuclear power is described briefly. The Chernobyl accident, which terrified the world and almost completely curtailed the development of nuclear power is also described briefly. The great international effort of research in the LWR design-base and severe accidents, which was, respectively, conducted prior to and following the TMI-2 and Chernobyl accidents is described next. We conclude that with the knowledge gained and the improvements in plant organisation/management and in the training of the staff at the presently-installed nuclear power stations, the LWR plants have achieved very high standards of safety and performance. The Generation 3+LWR power plants, next to be installed, may claim to have reached the goal of assuring the safety of the public to a very large extent. This review is based on the historical developments in LWR safety that occurred primarily in USA, however, they are valid for the rest of the Western World. This review can not do justice to the many fine contributions that have been made over the last fifty years to the cause of LWR safety. We apologize if we have not mentioned them. We also apologize for not providing references to many of the fine investigations, which have contributed towards LWR safety earning the conclusions that we describe just above.

• Journal of Power System Engineering
• /
• v.22 no.6
• /
• pp.5-10
• /
• 2018

Nuclear energy issues such as safety and social acceptance can not only influence the production costs of generating nuclear power, but also the external costs that are not reflected in market prices. Consequently, the social issues affiliated with nuclear power, beyond a severe accident, require some form of financial expense. The external social issues considered here are accident risk and realization, regulatory costs, and nuclear energy policy costs. Through several calculations and analyses of these external costs for nuclear power generation, it is concluded that these costs range from 7 to 27 \/kWh. Considering external costs are required for making energy plans, it could have an influence on generation costs.

• Proceedings of the Korean Nuclear Society Conference
• /
• 2002.05a
• /
• pp.337-337
• /
• 2002

• Nuclear Engineering and Technology
• /
• v.28 no.5
• /
• pp.467-481
• /
• 1996

An integrated framework of modeling the human operator cognitive behavior during nuclear power plant accident scenarios is presented. It incorporates both plant and operator models. The basic structure of the operator model is similar to that of existing cognitive models, however, this model differs from those existing ones largely in too aspects. First, using frame and membership function, the pattern matching behavior, which is identified as the dominant cognitive process of operators responding to an accident sequence, is explicitly implemented in this model. Second, the non-task-related human cognitive activities like effect of stress and cognitive biases such as confirmation bias and availability bias, are also considered. A computer code, OPEC is assembled to simulate this framework and is actually applied to an SGTR sequence, and the resultant simulated behaviors of operator are obtained.

• Nuclear Engineering and Technology
• /
• v.53 no.8
• /
• pp.2652-2660
• /
• 2021

The gamma-ray exposure doses in decontamination operation after a nuclear accident were evaluated with a consideration of various geometrical conditions and specific gamma-ray energies. The calculation domain is organized with three residence types and each form is divided into two kinds of geometrical arrangements. The position-wise air KERMA values were calculated with an assumption of evenly distributed gamma-ray source based on Monte Carlo radiation transport analysis using the MCNP code. The radioactivity is initially set to be unity to be multiplied by the deposition value measured in the actual accident condition. The workforce data set depending on the target object was determined by modifying the Fukushima report. The external exposure doses for decontamination workers were derived from the calculated KERMA values and the workforce analysis. These results can be used to efficiently determine the workforce required by the characteristics of the area and the structure to be decontaminated within the dose limits.

• Nuclear Engineering and Technology
• /
• v.49 no.2
• /
• pp.426-433
• /
• 2017

This paper describes the development process, the innovative techniques used and insights gained from the latest integrated, full scope, multistate Level 2 PSA analysis conducted at the Leibstadt Nuclear Power Plant (KKL), Switzerland. KKL is a modern single-unit General Electric Boiling Water Reactor (BWR/6) with Mark III Containment, and a power output of $3600MW_{th}/1200MW_e$, the highest among the five operating reactors in Switzerland. A Level 2 Probabilistic Safety Assessment (PSA) analyses accident phenomena in nuclear power plants, identifies ways in which radioactive releases from plants can occur and estimates release pathways, magnitude and frequency. This paper attempts to give an overview of the advanced modeling techniques that have been developed and implemented for the recent KKL Level 2 PSA update, with the aim of systematizing the analysis and modeling processes, as well as complying with the relatively prescriptive Swiss requirements for PSA. The analysis provides significant insights into the absolute and relative importances of risk contributors and accident prevention and mitigation measures. Thanks to several newly developed techniques and an integrated approach, the KKL Level 2 PSA report exhibits a high degree of reviewability and maintainability, and transparently highlights the most important risk contributors to Large Early Release Frequency (LERF) with respect to initiating events, components, operator actions or seismic component failure probabilities (fragilities).

• Nuclear Engineering and Technology
• /
• v.49 no.2
• /
• pp.434-441
• /
• 2017

Containment venting is one of several essential measures to protect the integrity of the final barrier of a nuclear reactor during severe accidents, by which the uncontrollable release of fission products can be avoided. The authors seek to develop an optimization approach to venting operations, from a simulation-based perspective, using an integrated severe accident code, THALES2/KICHE. The effectiveness of the containment-venting strategies needs to be verified via numerical simulations based on various settings of the venting conditions. The number of iterations, however, needs to be controlled to avoid cumbersome computational burden of integrated codes. Bayesian optimization is an efficient global optimization approach. By using a Gaussian process regression, a surrogate model of the "black-box" code is constructed. It can be updated simultaneously whenever new simulation results are acquired. With predictions via the surrogate model, upcoming locations of the most probable optimum can be revealed. The sampling procedure is adaptive. Compared with the case of pure random searches, the number of code queries is largely reduced for the optimum finding. One typical severe accident scenario of a boiling water reactor is chosen as an example. The research demonstrates the applicability of the Bayesian optimization approach to the design and establishment of containment-venting strategies during severe accidents.

• Nuclear Engineering and Technology
• /
• v.54 no.8
• /
• pp.2859-2870
• /
• 2022

Because nuclear power plants (NPPs) are safety-critical infrastructure, it is essential to increase their safety and minimize risk. To reduce human error and support decision-making by operators, several artificial-intelligence-based diagnosis methods have been proposed. However, because of the nature of data-driven methods, conventional artificial intelligence requires large amount of measurement values to train and achieve enough diagnosis resolution. We propose a graph neural network (GNN) based accident diagnosis algorithm to achieve high diagnosis resolution with limited measurements. The proposed algorithm is trained with both the knowledge about physical correlation between components and measurement values. To validate the proposed methodology has a sufficiently high diagnostic resolution with limited measurement values, the diagnosis of multiple accidents was performed with limited measurement values and also, the performance was compared with convolution neural network (CNN). In case of the experiment that requires low diagnostic resolution, both CNN and GNN showed good results. However, for the tests that requires high diagnostic resolution, GNN greatly outperformed the CNN.