Details

Autor(en) / Beteiligte

• Dundulis, Gintautas
• Kulak, Ronald F.
• Alzbutas, Robertas
• Uspuras, Eugenijus

Titel

Integrated probabilistic analysis of nuclear power plant building damage due to an aircraft crash

Ist Teil von

• International journal of crashworthiness, 2011-02, Vol.16 (1), p.49-62

Ort / Verlag

Cambridge: Taylor & Francis Group

Erscheinungsjahr

2011

Quelle

Taylor & Francis Journals Auto-Holdings Collection

Beschreibungen/Notizen

• In order to ensure that nuclear power plant buildings are reliable and safe in case of external loading, it is very important to evaluate uncertainties associated with loads, material properties, geometrical parameters, boundaries and other parameters. Therefore, a probability-based analysis was developed as the integration of deterministic and probabilistic methods using existing state-of-the-art software. The subject of this paper is the integrated analysis of building failure due to impact by a commercial aircraft. The Monte Carlo Simulation, First-Order Reliability and the combined Monte Carlo Simulation and Response Surface methods were used for the probabilistic analyses. During an aircraft crash, the dynamic impact loading is uncertain. Therefore a relation expressed by the probability of failure of impacted wall and loading function was determined. With failure defined as concrete cracking and rebar rupture, the failure probabilities of the impacted wall were calculated as a function of the peak impact load. The integrated deterministic and probabilistic analysis approach was applied to the Ignalina Nuclear Power Plant in Lithuania. The conclusions from this analysis was that a through-the-wall crack in the concrete element of a plant wall may occur with a probability of 0.0266, but the failure probability of the reinforcement bars is very small, that is, near zero. Thus, no perforation of the impacted wall by structures of the aircraft should occur. The importance of performing a probabilistic analysis of crash events is shown by comparing results to those obtained by a mean value deterministic approach.