基于ν-支持向量机的事故工况下反应堆功率预测  

Prediction of Reactor Power under Accident Conditions of Nuclear Power Plant Using ν-Support Vector Machine

作　　者：蒋波涛 黄新波[1] Hines JWesley 赵福宇[3] Jiang Botao;Huang Xinbo;Hines JWesley;Zhao Fuyu(School of Electronics and Information,Xi’an Polytechnic University,Xi’an,710048,China;Department of Nuclear Engineering,University of Tennessee,Knoxville,37996,U.S.A.;School of Nuclear Science and Technology,Xi’an Jiaotong University,Xi’an,710049,China)

机构地区：[1]西安工程大学电子信息学院,西安710048 [2]田纳西大学核工系,田纳西诺克斯维尔37996 [3]西安交通大学核科学与技术学院,西安710049

出　　处：《核动力工程》2019年第6期105-108,共4页Nuclear Power Engineering

基　　金：国家自然科学基金青年项目(11705135);国家留学基金资助项目（201508610045）;陕西省教育厅专项科研计划项目（15JK1297）;西安工程大学博士科研启动基金项目(BS1339)

摘　　要：针对事故工况下堆芯功率变化的特点和神经网络(ANNs)易陷极小值、收敛速度慢等问题,提出一种基于ν-支持向量回归机(ν-SVR)的事故工况下堆芯功率预测方法。该方法运用k重交叉验证(k-CV)完成对ν-SVR参数的优化,然后在此基础上设计2种不同的ν-SVR预测器并将其用于弹棒事故(REA)和落棒事故(RDA)工况下的堆芯功率预测。研究表明,与ANNs相比,该方法具有更高的预测精度和更短的响应时间。Aiming at the characteristics of core power change under accident conditions and the problems of artificial neural network(ANNs) such as easy to trap minimum and slow convergence speed, a prediction method of core power under accident conditions based on ν-support vector regression(ν-SVR) was proposed. This method used a k-CV to optimize the parameters of ν-SVR, and then two different ν-SVR predictors were designed. These two predictors were applied to the prediction of core power of rod ejection accident(REA) and rod drop accident(RDA). The results have shown that this method has higher prediction accuracy and shorter response time than the ANNs.

关 键 词：支持向量机 弹棒事故 落棒事故 

分 类 号：TL48[核科学技术—核技术及应用]