基于对抗神经网络模型的中国铀资源可持续发展安全评价

Safety Evaluation of Sustainable Uranium Development in China Combined with an Analytical GAN Framework

随着我国经济的快速发展,以化石能源为主的能源结构对我国环境压力与日俱增.相比之下,核能具有安全、高效、清洁、可持续性等诸多优点.铀作为核能的基础物质,目前已经受到国际各界的高度重视.如何保证铀的安全供应也已成为核可持续发展的重要保障.本文的工作是在动态系统建模的基础上,结合生成式对抗网络(GAN)模型,对我国铀资源可持续发展进行综合评价.采用8个基本指标和42个子指标进一步详细量化描述,本文通过制定一个框架,根据5种燃料循环过渡方案的量化绩效,在大量文献阅读和广泛征询专家意见的基础上,采用传统的问卷调查方法,对中国铀资源特有的可持续性进行评估和排名.首先利用KMO样本测量和Bartlett球度检验来确定因子分析的适用性和修正模型图与观测数据的适用性.然后从权重的角度分析了不同利益相关者代表所扮演的角色及其对综合排名的影响.结果表明,情景1对铀资源供应要求高,经济性最差,各部门之间的对抗性最大,对于协调和可持续发展会有更大的障碍,是最不可持续和高度对抗的情景.相反,情景5对技术的要求更高,各部门之间的对抗性更小,基本上满足了当前铀可持续性和低对抗性定义的基本要求.本文采用GAN框架计算铀供应安全评价指标体系,为铀的可持续供应和发展提供了灵活性,以应对技术发展的演变和内在不确定性.因此在进行铀资源可持续发展安全评价时需要进行博弈分析,找出影响铀矿供应安全的主导因素,协调各个部门及公众的利益关系,进而提出针对性对策,保障我国铀矿可持续安全和经济持续健康发展.

Uranium is the basic raw material for nuclear energy and is quite highly regarded.Developing a safe supply of uranium is important for safeguarding sustainable nuclear development.The purpose of this study is to evaluate the sustainability of uranium development in China based on dynamic system modeling combined with GAN(Generative Adversarial Network)analysis.We considered eight essential indicators and 42 sub-indicators as part of a detailed quantitative description,and then developed a framework to evaluate and rank China-specific sustainability in light of the quantitative performance of five options for fuel cycle transition scenarios.We began by using KMO sample measurements and the Bartlett Test of Sphericity to determine the suitability of factor analysis and the fitness of the corrected model map and observation data.We then analyzed the roles of different representatives of the decision makers and their impacts on the overall ranking by applying GAN methods from a weighted perspective.Five transition scenarios identified are 1)Pressurized Heavy Water Reactors,2)Mixed Light Water Reactor+Fast Reactor,3)Mixed LWR+FR fuel cycle scheme with heterogeneous irradiation,4)Mixed Pressurized Water Reactor+FR fuel cycle scheme with plutonium recycled directly and repeatedly,and 5)Sodium-cooled fast breeder reactor power plant.The results showed that scenario 1 is the most unsustainable and highly confrontational scenario with a high demand for uranium resources,the lowest sustainability and a high level of antagonism among departments.On the other hand,Scenario 5 requires more advanced technology but exhibits less antagonism among the departments,and thus it largely satisfies the basic requirements for uranium sustainability and low levels of antagonism.In this paper,a safety assessment index system for the uranium supply is computed using a GAN framework.This system plays a crucial role in the sustainable supply and development of uranium,and provides flexibility for coping with the evolution and inherent uncertainties of the necessary technological developments.