用于核反应堆安全壳关键参数快速预测系统及模型的构建方法,用于非能动安全壳冷却系统的事故预测,所述方法包括以下步骤:初始化安全壳参数,构建数据集;构建LSTM快速预测模型,LSTM快速预测模型包括单参量模型和多参量协同模型,利用单参量模型和多参量协同模型进行预测安全参数瞬态;利用数据集对模型进行训练得到最优LSTM快速预测模型;输入安全壳参数到最优LSTM快速预测模型对安全壳安全性能进行预测。所述系统中包括输入参数获取模块、数据处理模块、预测模块、预测结果检验模块。本发明成果可为核反应堆安全壳关键安全参数的敏感性、不确定性分析、系统设计优化提供快速预测智能分析模型。
A method of constructing a fast prediction system and a model for critical parameters of a nuclear reactor containment for accident prediction of a non-energetic containment cooling system, said method comprising the following steps: initialising containment parameters, constructing a dataset; constructing an LSTM fast prediction model, the LSTM fast prediction model comprising a single parameter model and a multi-parameter synergistic model, and predicting the safety parameter transients; using the data set to train the model to obtain the optimal LSTM fast prediction model; inputting the safety shell parameters to the optimal LSTM fast prediction model to predict the safety performance of the safety shell. Said system includes an input parameter acquisition module, a data processing module, a prediction module, and a prediction result checking module. The results of the present invention can provide a fast prediction intelligent analysis model for the sensitivity and uncertainty analysis of the key safety parameters of the nuclear reactor containment and system design optimisation.