Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

 
Session Overview
Session
6.06-1: PRA - I
Time:
Tuesday, 17/Mar/2020:
1:45pm - 3:15pm

Session Chair: Didier De Bruyn, SCK•CEN, Belgium
Session Chair: Hyun Gook Kang, RPI, United States of America
Location: B-1049

Show help for 'Increase or decrease the abstract text size'
Presentations

Electrical Cross-tie Option for Extended Station Blackout in Multi-Unit NPP Site

Jamila Khamis Alsuwaidi1,2, Ho Joon Yoon1, Hyun Gook Kang3

1Khalifa University of Science and Technology; 2Federal Authority for Nuclear Regulation (FANR); 3Rensselaer Polytechnic Institute

From the probabilistic evaluation of Nuclear Power Plant (NPP) risk, it is found that the initiating event of Station Blackout (SBO) is one of the main contributors to the core damage risk. Fukushima Daiichi accident has led to the situation where all AC power sources are lost for an extended period and consequently loss of core cooling and loss of heat removal.

Both Loss of Offsite Power and loss of the unit’s Emergency Diesel Generators will lead to the SBO, and if the shared Diesel Generator (AAC) is lost, the SBO event is extended. The electrical cross-tie is a sharing of other unit’s dedicated (EDG) to the unit under extended SBO of multi-unit NPP site. The study aims to re-evaluate the extended SBO risk with Electrical crosstie option in Single Unit Probabilistic Risk Assessment (PRA) with the impact of Multi-Unit Features (2 Units/Site, 3 and 4 Units/Site). To examine the extended SBO and get a more realistic estimation of Core Damage Frequency (CDF) from the PRA model, the (KU Pilot Model) is developed. The study has examined the factor of redundancy of the on-site AC sources on the CDF risk by the addition of Third and Fourth EDGs to the KU Pilot base Model. The research has evaluated the CDF risk with EDG Crosstie option from other units by developing a methodology and modelling of the crosstie taking into consideration the following aspects:

• Common Cause Failure between AC sources (DGs) in single and multi-unit site

• Study of Human Reliability Analyses (HRAs) for SBO and Extended SBO

• DC Battery load shedding to extend the battery life

• Modelling of Occupancy Factor of shared AAC and crosstie EDG

• Modelling of Multi-unit LOOP and Multi-unit SBO in Multi-unit Site



Dynamic Risk Assessment Using Bayesian Network with Unsupervised Machine Learning

Junyung Kim, Hyun Gook Kang

RPI

Dynamic probabilistic risk assessment (DPRA) inherently possess a fundamental challenge in data analytics: One should consider a large number of scenarios which is initiated from single transient. Furthermore, there is another challenge in DPRA to quantify the risk and its associated uncertainties and to draw meaningful risk insight from such a large amount of information. In this paper, we present a probabilistic mapping technique with a discretization of system space in time, which enables both risk quantification of a target system in a probabilistic manner as well as fault propagation tracking. Dynamic Bayesian network with a clustering algorithm for a dynamic system under uncertainties is proposed to develop risk insights that contribute to system safety improvement. The multilevel flow modeling and mean-shift clustering methods are used to identify principal system parameters which can represent system dynamics and to cluster similar scenario dataset so that the proposed method can be applied in practice. The risk effect quantification of variations in control units’ configuration would lead to the verification and improvement of dynamic system safety without the burden of additional thousands of simulations iteratively for a large variety of plant operational conditions.



Lessons Learned from Recent Seismic PRAs in USA

Ram Srinivasan1, Gabriel Toro3, Robert Sewell2

1Consultant; 2RTSA Consulting; 3Lettis Consultants International, Inc.

In response to the USNRC Request for Information following the Fukushima Incident, seventeen operating nuclear sites in the US are expected to complete a Seismic Probabilistic Risk Assessment (SPRA) by the end of 2019. The primary author of this proposed paper has been involved in ten of those SPRAs, either as a consultant to the utility or as an independent peer reviewer. Based on this experience and the information submitted to the USNRC, the authors have prepared a number of lessons learned to be shared with the international nuclear power plant community.

The seventeen operating sites cover a variety of BWRs and PWRs, designed and built in the 1960’s through 1980’s. They cover a wide range of site conditions (soil and rock) and different geographic locations of the US (East, Southeast, Midwest and West). The SPRAs were performed following the requirements specified in the ASME/ANS RA-Sb-2013 Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plants (Part 5) or the Code Case 1 to the above standard. Guidelines provided in the EPRI document (1025287 endorsed by the USNRC), Screening, Prioritization and Implementation Details (SPID) for the Resolution of Fukushima Near Term Task Force Recommendation 2.1; Seismic. The plants in the central and eastern parts of the US follow the seismic hazard methodology laid out in the USNRC (NUREG-2116)/USDOE (DOE/NE-0140)/EPRI (1021097) Central and Eastern United States Seismic Source Characterization for Nuclear Facilities Report [2013]. The two plants in the Western US performed a site-specific seismic hazard analysis generally following the SPID document.

The lessons learned would cover the breadth of the SPRA technical elements: Seismic Hazard Analysis, Seismic Structural Response and Fragility Analysis and the Plant Response Analysis.



 
Contact and Legal Notice · Contact Address:
Conference: ICAPP 2020
Conference Software - ConfTool Pro 2.6.134+TC
© 2001 - 2020 by Dr. H. Weinreich, Hamburg, Germany