Detail publikace

Pilot comparison of safety formats for reliability assessment of RC structures

Originální název

Pilot comparison of safety formats for reliability assessment of RC structures

Anglický název

Pilot comparison of safety formats for reliability assessment of RC structures

Jazyk

en

Originální abstrakt

ib Model Code 2010 and its ongoing revision consider a comprehensive system for treating reliability and uncertainty in reliability assessment and design of reinforced concrete (RC) structures using a nonlinear analysis. This study analyses a simply supported RC beam for which shear resistance has been experimentally verified, and steel yielding or concrete failure can be observed by varying shear reinforcement ratio. The effect of different failure modes is investigated by comparing design resistance obtained by selected safety formats: probabilistic assessment using LHS, estimate of coefficient of variation ECoV, global resistance format, and the partial factor method. The MC2010 Level of Approximation 3 (LoA3) is applied to represent models that combine the reinforcement and concrete contributions. A reference solution is provided by the probabilistic approach that inherently takes into account the correlation between the failure modes. It appears that the partial factors for materials, commonly adopted in design, are low and fail to adequately reflect the increased model uncertainty related to LoA3. The ECoV estimates are very close to the reference design values and in this case the flexibility of the method is demonstrated. The care needs to be taken when interpreting design values based on a few simulations – the effect of statistical uncertainty may be substantial.

Anglický abstrakt

ib Model Code 2010 and its ongoing revision consider a comprehensive system for treating reliability and uncertainty in reliability assessment and design of reinforced concrete (RC) structures using a nonlinear analysis. This study analyses a simply supported RC beam for which shear resistance has been experimentally verified, and steel yielding or concrete failure can be observed by varying shear reinforcement ratio. The effect of different failure modes is investigated by comparing design resistance obtained by selected safety formats: probabilistic assessment using LHS, estimate of coefficient of variation ECoV, global resistance format, and the partial factor method. The MC2010 Level of Approximation 3 (LoA3) is applied to represent models that combine the reinforcement and concrete contributions. A reference solution is provided by the probabilistic approach that inherently takes into account the correlation between the failure modes. It appears that the partial factors for materials, commonly adopted in design, are low and fail to adequately reflect the increased model uncertainty related to LoA3. The ECoV estimates are very close to the reference design values and in this case the flexibility of the method is demonstrated. The care needs to be taken when interpreting design values based on a few simulations – the effect of statistical uncertainty may be substantial.

BibTex


@inproceedings{BUT157215,
  author="Miroslav {Sýkora} and Jan {Červenka} and Vladimír {Červenka} and Jan {Mlčoch} and Drahomír {Novák} and Lukáš {Novák}",
  title="Pilot comparison of safety formats for reliability assessment of RC structures",
  annote="ib Model Code 2010 and its ongoing revision consider a comprehensive system for treating reliability and uncertainty in reliability assessment and design of reinforced concrete (RC) structures using a nonlinear analysis. This study analyses a simply supported RC beam for which shear resistance has been experimentally verified, and steel yielding or concrete failure can be observed by varying shear reinforcement ratio. The effect of different failure modes is investigated by comparing design resistance obtained by selected safety formats: probabilistic assessment using LHS, estimate of coefficient of variation ECoV, global resistance format, and the partial factor method. The MC2010 Level of Approximation 3 (LoA3) is applied to represent models that combine the reinforcement and concrete contributions. A reference solution is provided by the probabilistic approach that inherently takes into account the correlation between the failure modes. It appears that the partial factors for materials, commonly adopted in design, are low and fail to adequately reflect the increased model uncertainty related to LoA3. The ECoV estimates are very close to the reference design values and in this case the flexibility of the method is demonstrated. The care needs to be taken when interpreting design values based on a few simulations – the effect of statistical uncertainty may be substantial.",
  address="International Federation for Structural Concrete",
  booktitle="Proceedings of the fib Symposium 2019: Concrete - Innovations in Materials, Design and Structures",
  chapter="157215",
  howpublished="print",
  institution="International Federation for Structural Concrete",
  year="2019",
  month="may",
  pages="2076--2083",
  publisher="International Federation for Structural Concrete",
  type="conference paper"
}