交互式之核燃料束及核燃料套管材料的等效热性质之应用程序之建立

The Establishment of the Interactive Program on the Analysis of the Equivalent Thermal Properties of the Nuclear Fuel Bundle and Nuclear Fuel Casing Materials

核能是对人类来说是相当重要的能源,但使用核能发电后会有一些放射性产物,如用过核燃料,这些用过核燃料因为会持续发出衰变热及放射线,必须小心贮存,若电厂没有足够的空间贮存这些乏燃料,电厂将强制停止运转,不安全的贮存则可能造成危险甚至是燃料毁损。很多研究显示,干式贮存系统能有效的解决用过燃料的屯贮问题。建构适合台湾使用的系统或是系统正式启用前的审核工作,都必须进行详细的热流模拟分析,对于这项需求,核能研究所引进商用CFD软件Fluent来辅助分析工作的实行。目前的分析流程已趋于完善,并且在此套流程下所进行的分析工作在精确度上确实达到建构系统所需要的水平,但部份的流程却为分析人员所垢病;例如过多重复性且无必要的计算工作,以及Fluent数据库更新上的不便。本研究利用Visual Basic软件撰写自定义软件ADSS,从以上问题点切入,将干贮系统热流分析中常需简化之燃料组件与燃料提篮之均质化模式以及功率分布预先模块化,并与Fluent数据库作联结,配合窗口化的操作接口,使热流仿真分析工作能更有效率。本研究结果显示以模式建立计算后之Keff值代入FLUENT进行运算,其燃料束温度分布透过新方法得出温度略低于传统之旧有之方法,其误差不超过10%。

Nuclear is now an important energy source for people, however, when the operation of the nuclear power plants, spent fuel is definitely produced as the waste material. Unfortunately, these spent fuel not only undisposed but also a big problem for the storage. Insufficient of storage space can let power plants stop operating and unsafe storage can cause the nuclear disaster. A well-developed spent fuel dry storage system is one of the methods to solve the storage problem. For last decade, Taiwan Taipower Company introduces a system module from USA, but it turns out not a suitable program for spent fuel dry storage system in Taiwan. To design and construct a suitable system for Taiwan, we need to do CFD simulations. FLUENT is a commercial CFD program with strong ability. With the help of FLUENT, INER get the accurate result of simulations. However, some of the simulation steps were inconvenient, such as the parameters were not friendly and hard on database renewing for FLUENT. With Visual Basic, we produce a new pro- gram, namely ADSS, to solve above problems. This program contains modularize calculation models for fuel assembly, homogenization model for fuel bundles and power distribution. With the help of the windows operation interface, the program connects with FLUENT very easy and makes the simulation processes be more effective as well. The simu- lation results shown the error were within 10% for the temperature distribution compare with the temperature use by the traditional method.