Back to home
Using bootstrap method to establish surrogate safety goals for a society with operating NPPs   WANG Yan1, ZHANG Zhijian2, ZHANG Huazhi3, and XU Anqi4   1. Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, 150001, Harbin, China (wangyan622@hotmail.com) 2. Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, 150001, Harbin, China (zhangzhijian@hrbeu.edu.cn) 3. Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, 150001, Harbin, China (zhanghz@163.com) 4. Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, 150001, Harbin, China (anqixu520@foxmail.com)   Abstract: Quantitative risk values for nuclear power plants can be obtained by conducting Probabilistic Safety Assessment (PSA). However, people cannot judge the risk level without comparing the risk values from PSA with the standards of acceptable risk in society. Acceptable risk standards are affected by many factors, and those factors are preferentially considered in specified applications. In the U.S.A., both qualitative safety goals and quantitative health objectives (QHOs) for the current generation of light water reactors have been established by the comparative methods which are described in the Safety Goals Policy Statement Published by the U.S. Nuclear Regulatory Commission. In U.S.A. most of the existing nuclear power plants (NPPs) have conducted Level 1 PSA or simplified Level 2 PSA models, wherein surrogate safety goals are needed to evaluate whether or not the risks of those NPPs satisfy the QHOs. The existing surrogate safety goals had been established based on the QHOs. However, there are some problems in the derivation process, for example, unreasonable assumptions and logic. Apart from the U.S.A., most other countries with operating NPPs simply adopt the surrogate safety goals recommended by International Atomic Energy Agency without considering the society-specific risk and the development status of nuclear power. In this paper, the process of how to derive QHOs from qualitative safety goals and a model of quantitative health risk are first introduced, and then the models using Core Damage Frequency and Large Early Release Frequency based on QHOs are introduced. Generally speaking, there are multiple types of NPPs utilized in an individual society, and the environmental conditions of the plants are different from each other. So the specified society has the CDF and LERF samples satisfying the QHOs. The Bootstrap method is a resampling method and is suitable to the problems of small sample size. The upper limits of the confidence interval of CDF and LERF samples are obtained as surrogate safety goals by Bootstrap method. Lastly the QHOs of Chinese NPPs regarding individual early fatality and individual latent cancer fatality are estimated from the statistical data published by the government, and the CDF sample collected from the reference is applied to determine the surrogate safety goal CDF from the Bootstrap method. Keywords: safety goals; surrogate safety goals; bootstrap method
	Click here for viewing full text.