Variance decomposition is an effective sensitivity analysis method to screen the key parameters influencing passive safety system operation based on the uncertainties of thermal-hydraulic (T-H) model inputs. However, such method needs a large number of samples gained from T-H model running with the inputs sampled randomly from their probabilistic distributions, and the T-H model always takes quite long time to run once, then it will be a heavy calculation burden to do the analysis. In this paper, we propose a method to improve the analyzing efficient: based on the system T-H characteristics, the system behavior in a short time after an accident happening can represent the system T-H performance and be used to do the sensitivity analysis. Passive Containment Cooling System (PCCS) in AP1000 is used as a case study in our analysis, by which the heat produced in the containment can be transferred to the atmosphere through natural circulations. After steam line break (SLB) accident, the peak value of pressure in the containment appears within 1000s, we do the sensitivity analysis to screen key parameters in two ways: firstly, we screen the key inputs with variance decomposition method directly, 100 samples are gained from T-H model simulating 1000s, the inputs are sampled based on their probabilistic distributions, and the results show that air pressure is the most important parameter and the others don't have enough differentiation degrees. Then we get more 100 samples under the condition that air pressure is supposed as 0.1 MPa, here air temperature and steam mass flow are important ones besides air pressure. In another way, we analyze the correlation between pressure in the containment in a short time after SLB and the peak value according to the system T-H behavior, and get 600 samples from T-H model simulating 50s, the results are in accordance with that from T-H model simulating 1000s, air pressure, air temperature and steam mass flow are important parameters, and it just needs 1.55 h to calculate the important factor for one input.

An efficient method of key parameter screening for PCCS under SLB accident in AP1000

Di Maio F.;Zio E.
2020-01-01

Abstract

Variance decomposition is an effective sensitivity analysis method to screen the key parameters influencing passive safety system operation based on the uncertainties of thermal-hydraulic (T-H) model inputs. However, such method needs a large number of samples gained from T-H model running with the inputs sampled randomly from their probabilistic distributions, and the T-H model always takes quite long time to run once, then it will be a heavy calculation burden to do the analysis. In this paper, we propose a method to improve the analyzing efficient: based on the system T-H characteristics, the system behavior in a short time after an accident happening can represent the system T-H performance and be used to do the sensitivity analysis. Passive Containment Cooling System (PCCS) in AP1000 is used as a case study in our analysis, by which the heat produced in the containment can be transferred to the atmosphere through natural circulations. After steam line break (SLB) accident, the peak value of pressure in the containment appears within 1000s, we do the sensitivity analysis to screen key parameters in two ways: firstly, we screen the key inputs with variance decomposition method directly, 100 samples are gained from T-H model simulating 1000s, the inputs are sampled based on their probabilistic distributions, and the results show that air pressure is the most important parameter and the others don't have enough differentiation degrees. Then we get more 100 samples under the condition that air pressure is supposed as 0.1 MPa, here air temperature and steam mass flow are important ones besides air pressure. In another way, we analyze the correlation between pressure in the containment in a short time after SLB and the peak value according to the system T-H behavior, and get 600 samples from T-H model simulating 50s, the results are in accordance with that from T-H model simulating 1000s, air pressure, air temperature and steam mass flow are important parameters, and it just needs 1.55 h to calculate the important factor for one input.
2020
Key parameters screening; Passive safety system; Thermal-hydraulic characteristics; Variance decomposition
File in questo prodotto:
File Dimensione Formato  
An efficient sensitivity analysis method for PCCS in AP1000-revise4.pdf

accesso aperto

: Pre-Print (o Pre-Refereeing)
Dimensione 344 kB
Formato Adobe PDF
344 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1134281
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 2
social impact