S-Space College of Engineering/Engineering Practice School (공과대학/대학원) Dept. of Civil & Environmental Engineering (건설환경공학부) ICASP13
Pathways for Uncertainty Quantification through Stochastic Damage Constitutive Models of Concrete
- Wan, Zhiqiang; Chen, Jianbing; Beer, Michael
- Issue Date
- 13th International Conference on Applications of Statistics and Probability in Civil Engineering(ICASP13), Seoul, South Korea, May 26-30, 2019
- The constitutive model of concrete is of paramount significance for the design of concrete structures and the corresponding reliability assessment. In the present paper, the uniaxial damage model of concrete based on Chinese design code is introduced. It is noticed that there are seven crucial parameters in this model, while five of them are of physical significance and generally should be regarded as random variables. Therefore, the major task of the present paper is to study the effects, variations and randomness of these five parameters. Starting with the fuzzy analysis method (FAM), a brief uncertainty quantification scheme is described. This method is straightforward and easy to implement. Nevertheless, the prior knowledge (i.e., the engineering experience of designers or published literature) is required in FAM. Alternatively, the probability density evolution method (PDEM) is utilized with less needs of prior knowledge, while the type of marginal distribution of parameters is still required or assumed. Thus the epistemic uncertainty may be, more or less, brought in when applying these two methods. To improve this situation, i.e., to reduce the involvement of prior knowledge, a probabilistic learning method (PLM) is applied, in which the prior knowledge is reduced as it is nearly of data-driven background. The research results indicate that these three different methods of uncertainty quantification provide some basic and common conclusions, showing that all of them can capture the main characters of the experimental data. In addition, they individually offer various aspects of information due to different perspectives of these three methods. Therefore, these three methods might derive a series of powerful tools for uncertainty quantification in structural engineering, and be of future interest for opening new perspectives.