Uncertainty quantification of core characteristics parameter based on covariance data of evaluated nuclear data library

In the present paper, uncertainty quantification (UQ) due to covariance data of evaluated nuclear data library will be presented, especially for the light water reactor (LWR) core analysis. In order to perform the practical UQ for the LWR, the random sampling (RS) method is applied instead of the "Sandwich formula" using sensitivity coefficients. As an example of the RS-based UQ, we will discuss uncertainties of assembly power distributions of BWR, compared with that of PWR. Through this comparison, it is confirmed that the uncertainties of assembly power distributions of BWR are significantly smaller than that of PWR under operating condition, because of the strong void feedback effect in the BWR core. (author)