摘要
介绍了采用长短期记忆网络方法(LSTM)对核电机组在蒸汽发生器传热管破裂事故(SGTR)下状态参数的趋势预测,为应急人员的事故管理提供技术支持和决策参考。以M310机组为研究对象,并利用通用热工分析程序RELAP5构建仿真计算模型,构建不同SGTR类型事故序列,生成大量事故样本,然后利用样本数据训练得到基于LSTM神经网络的预测模型。分析结果表明模型可以提供对运行重要参数的准确预测,应急人员提前了解事故走向,进而提前干预,保障核电机组的安全、稳定运行。
This paper introduces the use of long short-term memory network(LSTM)to predict the trend of state parameters of Steam Generator Tube Rupture(SGTR)accidents of nuclear power units,providing technical support and decision-making reference for emergency personnel in accident management.Taking the M310 unit as the research object,a simulation calculation model was constructed using RELAP5 to calculate and generate a large number of SGTR accident sample data.Then,a prediction model based on LSTM neural network was trained using the sample data.The analysis results show that the model can provide accurate predictions of important operating parameters,allowing emergency personnel to understand the direction of the accident in advance,and then intervene in advance to ensure the safe and stable operation of nuclear power units.
作者
吕联鑫
马国扬
黄雄
魏巍
谢政权
冉晓隆
Lv Lianxin;Ma Guoyang;Huang Xiong;Wei Wei;Xie Zhengquan;Ran Xiaolong(China Nuclear Power Operation Technology Corporation,LTD,Wuhan 430223,China;CNNC Key Laboratory on Nuclear Industry Simulation,Wuhan 430074,China)
出处
《电子技术应用》
2024年第S01期109-115,共7页
Application of Electronic Technique
作者简介
吕联鑫(1994-),男,硕士研究生,主要研究方向:核安全仿真。
