福岛核事故典型气载放射性核素的弥散及辐射剂量研究
发布时间：2019-01-07 16:38
【摘要】：2011年3月11日，日本本州岛仙台港以东约130km处海底发生了9.0级大地震，该地震引发的巨大海啸，导致福岛第一核电站机组严重受损以及大规模放射性物质的释放。事故的发生再次引起了民众和科研人员对于事故应急和辐射安全的普遍关注。 本文应用MVK模型验证工具对ARTM、AERMOD和CALPUFF大气扩散模式进行了统计验证，结果表明：三种模型都基本能够达到推荐模型标准，ARTM模型和CALPUFF模型在0～3km距离段内有低估趋势，AERMOD模型有高估趋势，CALPUFF模型在模拟区域大于50km的模拟预测方面具有一定的优越性。 本文应用CALPUFF大气扩散模式模拟了以福岛第一核电站为中心160km160km的区域，事故后30天~(131)I的浓度分布和沉积量分布情况并使用等效衰减法对计算结果进行了放射性衰变修正。事故期间空气浓度峰值出现在事故发生后的第5天第13时(79,81)网格，空气浓度峰值为6.7010~6Bq/m~3，地面沉积量为1.8110~5Bq/m~2，释放源西北方向为污染最严重的陆域方向。 通过系统分析法对研究区域的生态系统进行简化，用一系列微分动力学方程描述核素的生态转移机制，编写程序实现核素的转移过程，确定了受照的关键人群、关键途径和关键器官，，研究区域内的四个居民点中位于源西北方位的葛尾村公众的甲状腺当量剂量最大。 鉴于福岛核事故早期全厂断电，堆芯受损情况不明，不能通过堆芯的运行参数和损坏程度有效的估计源项，本文通过将混合遗传算法与烟团扩散模式的耦合，对采用环境监测数据反演核事故源项研究进行了初步探索，实现了对源项释放率、位置和高度的快速估计。遗传算法-单纯形法与遗传算法-模式搜索法、遗传算法和单纯形法等三种算法的比较结果表明：遗传算法-单纯形法结合了遗传算法和单纯形法两种算法的优势，也弥补了各自的缺陷，其反算值可以与期望值较精确符合、通用性强、计算花费较少，能够适用于核事故源项的快速估计。
[Abstract]:On March 11, 2011, a magnitude 9.0 earthquake struck the sea floor about 130km east of Sendai Port, Honshu Island, Japan. The earthquake triggered a huge tsunami that caused serious damage to the units of the Fukushima Daiichi Nuclear Power Station and the release of large-scale radioactive material. The accident has once again aroused widespread concern for emergency and radiation safety among people and researchers. In this paper, MVK model verification tool is used to verify the atmospheric diffusion models of ARTM,AERMOD and CALPUFF. The results show that the three models can basically reach the recommended model standard, and the ARTM model and the CALPUFF model tend to be underestimated in the range of 0~3km. The AERMOD model tends to be overestimated, and the CALPUFF model has some advantages in the simulation prediction of the simulation region larger than that of the 50km model. In this paper, the CALPUFF atmospheric diffusion model is used to simulate the distribution of ~ (131) I concentration and deposition in the area with Fukushima Daiichi nuclear power station as the center. The radioactive decay correction is carried out by using the equivalent attenuation method. During the accident, the peak of air concentration appeared at 13:00 on the fifth day after the accident, the peak of air concentration was 6.7010 ~ 6Bq / m ~ (-1), the amount of surface deposition was 1.81 ~ 10 ~ 5Bq / m ~ (2), and the northwestern direction of the release source was the most polluted land area. The ecological transfer mechanism of nuclides is described by a series of differential dynamics equations, and the process of radionuclide transfer is realized by a program. Among the four residential areas in the study area, the thyroid equivalent dose of the public in Gewei Village, located in the northwest of the source, was the largest. In view of the power failure of the whole plant in the early stage of the Fukushima nuclear accident, the damage of the reactor core is unknown, and the effective source term of the operation parameters and damage degree of the reactor core can not be estimated, the hybrid genetic algorithm is coupled with the smoke diffusion model in this paper. A preliminary study on the inversion of source term of nuclear accident using environmental monitoring data is carried out, and the fast estimation of release rate, location and height of source term is realized. The comparison of genetic algorithm-simplex method and genetic algorithm-pattern search method, genetic algorithm and simplex method shows that genetic algorithm-simplex method combines the advantages of genetic algorithm and simplex algorithm. It can be used to estimate the source term of a nuclear accident because it is more accurate than the expected value and has the advantages of high generality and less calculation cost.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TL73
本文编号：2403885
[Abstract]:On March 11, 2011, a magnitude 9.0 earthquake struck the sea floor about 130km east of Sendai Port, Honshu Island, Japan. The earthquake triggered a huge tsunami that caused serious damage to the units of the Fukushima Daiichi Nuclear Power Station and the release of large-scale radioactive material. The accident has once again aroused widespread concern for emergency and radiation safety among people and researchers. In this paper, MVK model verification tool is used to verify the atmospheric diffusion models of ARTM,AERMOD and CALPUFF. The results show that the three models can basically reach the recommended model standard, and the ARTM model and the CALPUFF model tend to be underestimated in the range of 0~3km. The AERMOD model tends to be overestimated, and the CALPUFF model has some advantages in the simulation prediction of the simulation region larger than that of the 50km model. In this paper, the CALPUFF atmospheric diffusion model is used to simulate the distribution of ~ (131) I concentration and deposition in the area with Fukushima Daiichi nuclear power station as the center. The radioactive decay correction is carried out by using the equivalent attenuation method. During the accident, the peak of air concentration appeared at 13:00 on the fifth day after the accident, the peak of air concentration was 6.7010 ~ 6Bq / m ~ (-1), the amount of surface deposition was 1.81 ~ 10 ~ 5Bq / m ~ (2), and the northwestern direction of the release source was the most polluted land area. The ecological transfer mechanism of nuclides is described by a series of differential dynamics equations, and the process of radionuclide transfer is realized by a program. Among the four residential areas in the study area, the thyroid equivalent dose of the public in Gewei Village, located in the northwest of the source, was the largest. In view of the power failure of the whole plant in the early stage of the Fukushima nuclear accident, the damage of the reactor core is unknown, and the effective source term of the operation parameters and damage degree of the reactor core can not be estimated, the hybrid genetic algorithm is coupled with the smoke diffusion model in this paper. A preliminary study on the inversion of source term of nuclear accident using environmental monitoring data is carried out, and the fast estimation of release rate, location and height of source term is realized. The comparison of genetic algorithm-simplex method and genetic algorithm-pattern search method, genetic algorithm and simplex method shows that genetic algorithm-simplex method combines the advantages of genetic algorithm and simplex algorithm. It can be used to estimate the source term of a nuclear accident because it is more accurate than the expected value and has the advantages of high generality and less calculation cost.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TL73
【参考文献】
相关期刊论文 前10条
1 刘立平,牛熠;遗传算法综述[J];东莞理工学院学报;2005年03期
2 邢可霞;郭怀成;;环境模型不确定性分析方法综述[J];环境科学与技术;2006年05期
3 蔡旭晖,陈家宜,康凌;核事故条件下的大气扩散模式及应用[J];辐射防护;2003年05期
4 姚仁太,徐向军,郝宏伟,姚青山;远东地区核风险厂址释放的气载放射性污染物长距离迁移的模拟[J];辐射防护;2005年04期
5 邹敬,曲静原,曹建主;RODOS系统中RIMPUFF模型的验证与比对[J];核动力工程;2005年05期
6 伯鑫;丁峰;徐鹤;李时蓓;;大气扩散CALPUFF模型技术综述[J];环境监测管理与技术;2009年03期
7 王淑兰,孟志鹏,丁信伟;适用于固定工业源泄放的扩散模型-AERMOD[J];化学工业与工程;2003年04期
8 徐志新;奚树人;曲静原;;核事故源项反演技术及其研究现状[J];科技导报;2007年05期
9 邹旭东;杨洪斌;刘玉彻;;CALPUFF在沈阳地区大气污染模拟研究中的应用[J];气象与环境学报;2008年06期
10 刘爱华;蒯琳萍;;放射性核素大气弥散模式研究综述[J];气象与环境学报;2011年04期
本文编号：2403885
本文链接：https://www.wllwen.com/kejilunwen/anquangongcheng/2403885.html
