A Method to Minimize the Sampling Sample Size for Uncertainty Analysis of Nuclear Reactor Physics

Abstract

A method to minimize the sampling sample size of nuclear reactor physical uncertainty analysis, firstly determine the multi-group cross-sectional population covariance matrix of the nuclear reactor inner element to be analyzed, and then use the Latin hypercube sampling to obtain a group from the same dimension standard normal distribution population The sample, so that the rank of its covariance matrix is ​​not less than the rank of the overall covariance matrix of the multi-group section, linearly transforms the sample, and solves a transformation matrix for linear transformation to ensure that the transformed sample mean and covariance are respectively equal to The mean and covariance of the multi-group cross-section population, and then obtain the calculation samples of the multi-group cross-section input parameters. The invention reconstructs the uncertainty of nuclear data with the minimum sample size, ensures the convergence of uncertainty analysis results, and solves the problems of loss of nuclear data uncertainty information and huge sample size required by traditional sampling methods; the inventive method is easy to implement and easy to use. The calculation efficiency is significantly improved, and the convergence result is accurate and reliable, which is of great significance to the analysis of nuclear reactor physical uncertainty.

Description

technical field [0001] The invention relates to the field of nuclear reactor physical uncertainty analysis, and relates to a method for minimizing the sampling sample size of nuclear reactor physical uncertainty analysis. Background technique [0002] The results of nuclear reactor physical uncertainty analysis are of great significance to ensure the safety and economy of the reactor. Statistical sampling method is an effective and widely used method for uncertainty analysis. This method is simple and easy to implement, and has strong applicability to various types of input-output systems. Nuclear data uncertainty is a major source of nuclear reactor physics uncertainty. In the process of analyzing the uncertainty of nuclear data based on the sampling method, the uncertainty information of multi-group cross-section input parameters is carried by the sample, which is transferred from the nuclear data to the key response of the reactor physics calculation. Therefore, the deg...

AI Technical Summary

Problems solved by technology

[0004] In addition, due to the large number of input parameters of multi-group cross-sections for nuclear reactor physical calculations, traditional sampling methods face the problem of "curse of dimensionality"

If the nuclear data uncertainty information is described as completely as possible, the sample size required by the traditional sampling method is very large, while the reactor physical calculation model is complex, the program scale is large, and the calculation takes a long time. During the uncertainty analysis process, each sample needs a calculation program. Executed repeatedly, the calculation cost of a large number of samples is very huge, even unaffordable

Images