摘要
本文成功地训练了一个用于预测流动不稳定性条件下CHF的人工神经网络,利用所训练的人工神经网络,分析了流动不稳定性对无量纲因子F的影响规律。分析结果表明,系统压力对F的影响是非单值的;平均质量流速增大,总体上F也增大。还分析了流动不稳定性条件下系统主要参数对临界热流密度的影响。结果表明:当质量流速的振幅与平均质量流速的比增加时,CHF减小;质量流速的振幅与平均质量流速的比不变时,随着周期增大,CHF减小。
An Artificial Neural Network, denoted by ANNosci, has been trained successfully to predict the critical heat flux （CHF） under flow oscillations in this paper. The effects of flow oscillations on nondimensionalized factor F is analyzed by using the ANNosci. The analyzed results show that the influence of the system pressure on the nondimensionalized factor F is multivalued due to the effects of other parameters. The nondimensionalized factor F will increase with an increase of mean mass flow rate, in general. The influences of flow oscillation on the critical heat flux are also analyzed. The results show that these effects are very complex. Shortly, CHF decreases with increasing the ratio of oscillation amplitude of mass flow rate to mean mass flow rate. When the ratio of oscillation amplitude of mass flow rate to mean mass flow rate is constant, CHF decreases with increasing the oscillation oeriod.
出处
《核动力工程》
EI
CAS
CSCD
北大核心
2007年第3期19-21,60,共4页
Nuclear Power Engineering
基金
陕西省自然科学基金(2003E217)
教育部留学归国人员基金(03回国基金05)资助
